CHAPTER 12: SURFACE ELECTROCARDIOGRAPHY

The surface electrocardiogram (ECG), introduced more than 100 years ago by Willem Einthoven, is the most common technique for the study of heart diseases. It is very useful for the evaluation of acute chest pain, palpitations, syncope, and acute dyspnea, and it is the gold standard for the diagnosis of cardiac arrhythmias, conduction disturbances, preexcitation syndromes, channelopathies, and some aspects of acute ischemic heart disease (IHD). Additionally, it is a fundamental tool to assess the evolution of heart diseases, in particular ischemic diseases, as well as in situations such as electrolyte disorders and drug therapy. It is also useful for epidemiologic studies and for screening and evaluation of athletes.

Despite its invaluable usefulness if used correctly, the interpretation of an ECG recording of normal appearance must be performed with caution. We should bear in mind that a relatively high percentage of patients with coronary heart disease, in the absence of chest pain, show a normal ECG recording. In approximately 5% to 10% of acute coronary syndromes (ACSs), the ECG is normal or borderline, especially in the early phases. Furthermore, the ECG may appear normal after a myocardial infarction (MI). Therefore, a normal ECG is not completely reassuring, because a patient may die from cardiac causes even on the same day a normal recording was taken. However, in the absence of any clinical symptoms or signs or family history of sudden death, the possibility of this occurring is very remote.

On the other hand, some subtle ECG abnormalities without evidence of heart disease may be observed occasionally. Notwithstanding, one must be cautious in such circumstances, and IHD, channelopathies, and preexcitation syndromes should be ruled out before relegating minor changes to nonspecific changes that are not meaningful. Therefore, it is necessary to read the ECG recordings while keeping in mind the clinical setting (eg, family history, chest pain, syncope) and, if necessary, taking sequential recordings.

In addition, normal variants may be observed in the ECG recording that are related to constitutional habits, chest malformations, age, or other factors. Even transient abnormalities may be detected due to a number of causes (eg, hyperventilation, hypothermia, glucose or alcohol intake, ionic abnormalities, or effect of certain drugs).

Today, the importance of the ECG goes beyond its established role of diagnosing abnormal patterns. It has become a tool to determine prognosis and perform risk stratification in many clinical situations, because it can provide insight into basic electrophysiology by ascertaining abnormalities at the molecular level, such as in channelopathies. Furthermore, it is sometimes the key technique to inform the need for such invasive devices as permanent pacemaker or cardiac resynchronization therapy. Therefore, as has been recently stated,¹ it could be said that we face a new renaissance of ECG.

These facts should be borne in mind before starting to learn a technique such as ECG, because the ECG assessment has to be done within the context of a given clinical condition. For example, the occurrence of minor ST-segment depression, if in conjunction with precordial pain, is more likely to be secondary to ischemia (ACS) than it is in the asymptomatic patient, when it is more likely explained by such noncardiac causes as ingestion of drugs or alcohol or hyperventilation.

In this chapter, the basis of the normal ECG and of the abnormal morphologic changes that may occur as a consequence of the enlargement of cavities, atrial and ventricular conduction disorders, preexcitation, and ischemia will be discussed. After that, the most relevant ECG abnormalities found in ischemic and other heart diseases, as well as in some special situations, such as ionic imbalances, hypothermia, and athletes, will be summarized. ECG of arrhythmias and the other noninvasive (exercise ECG, Holter technology, body mapping) and invasive electrical technologies used for study of heart diseases are discussed elsewhere.

Different Methods for Recording the Electrical Activity of the Heart

The ECG is a linear recording of the electrical activity of the heart that repeats over time, taken from outside the body by surface electrodes. For each cardiac cycle, different successive morphologies are recorded: atrial depolarization (P wave); ventricular depolarization, called the QRS complex; and ventricular repolarization (T wave). Occasionally, other waves, such as U waves, may be recorded. The smooth wave of atrial repolarization usually remains hidden in the QRS complex (Fig. 12–1A).

Between these waves, there are different intervals. Figure 12–1A shows these different waves and intervals that represent the successive activation (depolarization and repolarization) of atria and ventricles, which repeat one cycle after another. In this figure, the ECG morphology corresponds to the electrical activity measured from the surface of the left ventricle (Fig. 12–1B). The P, QRS, and T waves are positive because the electrode faces the head of the main depolarization vector (→) of the atria and left ventricle (see Fig. 12–1B) that moves toward the electrode and the head of the repolarization vector of the ventricles; yet during repolarization, the sense of the phenomenon (see Figs. 12–1B and 12–5) moves away from the electrode (from epicardium to endocardium). The vector of depolarization and repolarization of the ventricles is the vectorial expression on the exterior of the cell of a dipole caused by the propagation of depolarization and repolarization fronts (couple of electrical changes, – +, that initiate the process of depolarization and repolarization; see Fig. 12–1B). The small initial and terminal morphologies of QRS are negative because the initial and terminal forces of ventricular depolarization move away from the left ventricle electrode (see Fig. 12–1B).

Other forms of recording the electrical activity of the heart may be used, such as vectorcardiography (VCG), body mapping, and intracavitary recordings. VCG records the path that follows the electrical activity of the heart through closed loops that represent atrial depolarization (P loop), ventricular depolarization (QRS loop), and ventricular repolarization (T loop) (see Fig. 12–12). Each loop is formed by the sum of its multiple vectors and represents the pathway followed by the successive dipoles of depolarization of atria and ventricles and of repolarization of the ventricles. The head of these vectors is always located toward the positive part of the depolarization or repolarization dipoles (see Fig. 12–12).

A close correlation exists between VCG loops and the ECG curves. The ECG morphology may be deducted from the morphology of the VCG loops and vice versa. This reflects the loop-hemifield correlation theory (see The Dipole-Vector-Loop-Hemifield Correlation). According to this correlation (see Figs. 12–10 and 12–12), the morphology of different waves (P, QRS, and T) recorded from different sides (leads) varies. Because the heart is a three-dimensional organ, the projection of the loop’s maximum vectors in the frontal and horizontal planes, in the positive and the negative hemifield of each lead (see Leads and Hemifields: Electrocardiography-Vectorcardiography Correlation]) is required to ascertain exactly the loop’s location and deduct ECG morphology (Fig. 12–2; see also Fig. 12–10). However, the morphology of ECG not only depends on the maximum vector of a given loop but also on its rotation (see Figs. 12–2 and 12–10). This underscores the importance of considering the loop in its entirety, and not only its maximum vector, as most of the available texts do, to explain the ECG morphology.

Presently, the recording of the VCG has little practical application,^2,3,4 although it may be of interest for research purposes, and in the assessment of select conduction blocks and/or characterization of tissue necrosis. In addition, the correlation of the ECG with the VCG loop is extremely useful for educational purposes.

To summarize:

1. To have a clear understanding of ECG curves, it is essential to perform a deductive analysis of the electrical activation of the heart under normal and pathologic conditions with the use of VCG loop projection, knowing its clockwise or counterclockwise rotation, on the positive and negative hemifields of different leads (see The Dipole-Vector-Loop-Hemifield Correlation).

2. The ECG-VCG correlation is a useful method to understand the ECG curves and is the method that we use for teaching purposes (see Figs. 12–2 and 12–10 and The Dipole-Vector-Loop-Hemifield Correlation).

The Origin of Electrocardiography Morphology: From Cellular Electrogram to Human Electrocardiogram

When a microelectrode is placed outside and another inside a myocardial cell, a transmembrane diastolic potential, is recorded, which is stable in contractile cells (Fig. 12–3) and ascendant in pacemaker cells (see Fig. 12–11). Then, if is stimulated properly the contractile cell, the monophasic curve called the transmembrane action potential (TAP) is recorded, which corresponds to the depolarization and repolarization of the cell (Fig. 12–4).

The correlation between the TAP of the contractile cells and the human ECG is as follows (see Fig. 12–4): phase 0 corresponds with the QRS complex, phase 1 and the initial part of phase 2 correspond with the J point and start of the ST segment, the middle part of phase 2 corresponds with the isoelectric ST line, and the end of phase 2 and phase 3 correspond with the T wave.

The cellular electrogram and the human ECG represent the sum of the cellular or ventricular depolarization plus repolarization. Both cell and ventricle recordings are characterized by having a similar depolarization (QRS complex), but different repolarization (T wave). This is negative in the cellular electrogram and positive in the human ECG. In both cases, the T wave follows a period (ST segment) that in normal conditions is isoelectric.

Because of the work by Wilson et al,⁵ it is accepted that this difference between the cellular electrogram and human ECG may be explained by the opposite direction of the human heart repolarization wave (T wave) with respect to the depolarization (QRS complex) (Fig. 12–5), whereas in the cellular electrogram, the repolarization wave follows the same direction.

It is well known⁶ that the depolarization starts in the endocardium, specifically in three areas of the left ventricle (see Fig. 12–13), and that the epicardium depolarizes after the endocardium. Many experimental works performed in animals^7,8 and in human beings, through the separate recording of endocardium and epicardium TAPs during catheterization and cardiac surgery,⁹ have shown that the areas with the shortest TAP, and thus the earliest repolarization, correspond to the epicardium. Therefore, it seems evident that a transmural repolarization gradient exists; although for some authors,¹⁰ the middle section of the ventricular wall (M-cell zone) takes even longer to repolarize than the endocardium, whereas for other authors,^11,12 the repolarization gradient, more than transmural, is transregional (eg, between the apex and the base).

We explain the origin of the human ECG based on the fact that the epicardium repolarizes before the endocardium; this is because repolarization always starts in areas with more abundant perfusion, and physiologically (in normal conditions), the subendocardium is relatively underperfused (see Fig. 12–5). At a cellular level, the opposite happens because the zone distal to the electrode (equivalent to the endocardium) does not have this physiologic hypoperfusion, and therefore, the first zone to be depolarized (the zone distal to the electrode) is also the first to be repolarized.

The approach to explain the human ECG arises from two different theories:

1. The first theory considers, in light of what has been discussed, that the curves of the human ECG are the sum of the successive dipoles (– +) of depolarization and repolarization recorded by an electrode located on the opposite side (epicardium) relative to the place (endocardium) where the stimulation initiates (depolarization) (see Fig. 12–5).

2. The second theory considers that the human ECG is the sum of the TAPs of the subendocardium, the farthest part to the recording electrode (which TAP starts before and ends later), and the subepicardium, the closest part to the electrode (which TAP starts later and ends before).

This is a simplification of how the ECG is produced because we have considered the left ventricle alone, and thus its depolarization can be represented by a single vector. Actually, this is not the case because there are more vectors in the human ECG. At least one vector is required, as the sum of all atrial vectors, to explain the pathway that is followed by the stimulus to depolarize the atrium. This is known as the atrial depolarization loop (P wave) (see Figs. 12–1A and 12–12A). Three vectors explain the pathway of the ventricular depolarization loop (QRS complex) (see Fig. 12–1B). The QRS complex is followed by the ventricular repolarization loop (T wave) (see Fig. 12–1C). Three vectors are required to explain the morphology of the ventricular depolarization loop (qRs or rSr′) and depend on the placement of the recording lead, although they actually represent many more vectors (see Figs. 12–1 and 12–11).

To summarize:

1. There is considerable evidence that, in the human heart, the epicardium (the zone nearest to the exploring electrode) is the first part to repolarize and, therefore, the human ECG records a positive T wave (see Fig. 12–5), the opposite of what the cellular electrogram shows. The depolarization and repolarization dipole is expressed as a vector. The sum of the successive depolarization and repolarization dipoles and the vectors that express them, from an electrode located on the opposite side relative to the site were the stimulus originates, can explain the human ECG curve (see Fig. 12–5).

2. The human ECG can also be considered as the result of summation of the TAPs of the farthest (the subendocardium) and the nearest (the subepicardium) zones from the exploring electrode (Fig. 12–6).

Leads and Hemifields: Electrocardiography-Vectorcardiography Correlation

Leads and Hemifields

The ECG presents different morphologies as it is recorded from different sites, named leads. We currently use six frontal (I, II, III, VR, VL, and VF) and six horizontal (V₁ to V₆) leads (Figs. 12–7A, 12–7B, and 12–8C).

There are three bipolar leads (I, II, and III) in the frontal plane, which, according to the Einthoven law, should accomplish the equation II = I + III. These three leads form what we call the Einthoven triangle (see Fig. 12–7A). Figure 12–7C depicts how the projection of different vectors (or loops) explains the different morphologies found in leads I, II, and III. Bailey, by shifting the three leads toward the center, obtained a reference figure (Bailey triaxial system) (Fig. 12–8A). In the frontal plane, there are also three monopolar leads (VR, VL, and VF) (see Fig. 12–7B). By adding these three leads to the Bailey triaxial system, the Bailey hexaxial system is obtained (Fig. 12–8B).

All of the 12 leads located around the circumference in the frontal and horizontal planes have a positive line (Fig. 12–8D; see Fig. 12–8B) that goes from the place where the electrode is located up to the center of the heart. If lines perpendicular to the frontal and horizontal leads are drawn passing through the center of the heart, the positive and negative hemifields of these leads may be obtained (Fig. 12–9). The positive hemifield of lead I extends from +90° to –90° passing through 0°; that of lead II extends from –30° to +150° passing through +60°; that of lead III extends from +30° to –150° passing through +120°; that of VR extends from +120° to –60° passing through –150°; that of VL extends from –120° to +60° passing through –30°; that of VF extends from 0° to ±180° passing through +90°; that of V₂ extends from 0° to 180° passing through +90°; and that of V₆ extends from –90° to +90° passing through 0°. The rest of the hemifields corresponding to the horizontal plane leads can be obtained in the same manner, drawing lines that are perpendicular to the corresponding lead, passing through the center of the heart (see Fig. 12–9; Fig. 12–10). In all cases, the negative hemifields are opposed to the positive ones.

A loop of P, QRS, or T, or their maximum vectors, located in the positive or the negative hemifield or on the interface between both hemifields in any of the 12 leads, gives rise to, respectively, positive, negative, or isodiphasic deflections of P, QRS, or T waves in that given lead. An isodiphasic deflection has a maximum vector but may have a different morphology; it can be positive/negative or negative/positive, according to the direction of the loop rotation representing the pathway followed by the stimulus (see Fig. 12–2). The degree of positivity or negativity in each morphology depends on two factors: the magnitude and the direction of the loops or vectors. For a given magnitude, the vectorial result will have a greater positivity or negativity according the direction of the maximal vector of the loop; and for a given direction, the vector of the loop with a greater magnitude will cause a greater positivity or negativity (see Fig. 12–9).

The Dipole-Vector-Loop-Hemifield Correlation

As previously mentioned, a pair of electric charges called a dipole (– +) is formed in both depolarization and repolarization processes of activation of the heart. The existence of these dipoles results from ionic changes and explains the formation of TAP (see Fig. 12–4 and legend). These dipoles have a vectorial expression, with the head of the vector located in the positive part of a dipole. An electrode that faces the head of the vector records positive deflection regardless of whether this dipole approaches the electrode or moves away. It has already been explained how the cellular and ventricular electrograms are formed and that, in a human ECG, the repolarization wave (T wave) is positive because physiologically there is less perfusion in the subendocardial zone and the process of repolarization always starts in the more perfused zone. Therefore, in the human ECG, this process begins in the subepicardium, the opposite of what occurs at the cellular level (see Fig. 12–5).

It is relevant to highlight that the P, QRS, and T loops are formed from the vectorial sum of depolarization and repolarization dipoles that express the pathway followed by electric stimulus during these processes (Fig. 12–11). As stated, only the simultaneous projection on two planes, frontal and horizontal, may provide exact information as to the direction of respective electric forces (frontal plane: upward-downward and rightward-leftward; horizontal plane: rightward-leftward and frontward-backward) (see Fig. 12–11). Each of these loops has a maximum vector, which is considered to be the result of all instantaneous vectors (see Fig. 12–11) and expresses the magnitude and the overall direction of a loop. Nevertheless, the morphology of a loop, especially the initial and terminal parts, and the loop rotation (clockwise or counterclockwise) have relevant added value (see Figs. 12–10 and 12–11; see also Fig. 12–2).

The tracing of atrial (P) and ventricular (QRS) wave activation explained by the VCG curve may be translated into ECG curves in different leads according to the VCG-ECG correlation. Figure 12–10 shows this correlation with VF and V₂ leads.

To summarize:

1. The following chain of events explains the ECG morphology: dipole → vector → loop → hemifield → ECG.

2. The P, QRS, and T loops overall have an orientation that may be expressed by a maximum vector. Although these vectors provide important information on ECG morphology in different leads, only the global contour of the loop, its sense of rotation, and the loop-hemifield correlation may explain the total ECG morphology (see Figs. 12–2, 12–10, and 12–11).

3. The rotation of the loop, clockwise or counterclockwise, explains why, for a given maximal vector, the morphology of the QRS may be either or (see Figs. 12–2 and 12–10).

Activation of the Heart

The ECG tracing reflects the activation sequence (depolarization and repolarization) of the heart, starting with the stimulus arising in the sinus node (this is the structure with greater automaticity), up to the ventricular Purkinje network through the specific conduction system. Figures 12–1B and 12–11 show a summary of atrial depolarization and ventricular depolarization and repolarization that explains the ECG according to the dipole → vector → loop → hemifield theory, and Fig. 12–12 shows how the ECG may be also considered as the sum of different TAPs originating along the heart activation sequence from the sinus node to the ventricular muscle. The conduction speeds of different parts of the specialized conduction system are shown at the right in Fig. 12–12.

The union of the heads of all atrial depolarization vectors represents the P loop, which is recorded on the ECG as the initial deflection, the P wave (see Fig. 12–11A). The loop-hemifield correlation explains the morphology of the P wave in the different leads. Generally, atrial repolarization (Ta wave) is seldom seen, being masked by the significant forces generated by ventricular depolarization that give rise to the QRS complex (see Fig. 12–1A).

Once atrial depolarization ends and prior to starting ventricular depolarization (PR segment in ECG), the electric stimulus depolarizes small structures, and therefore, no waves are recorded on the surface ECG (see Fig. 12–1). Nonetheless, depolarization of the bundle of His and its branches can be recorded with intracavitary recording techniques (His recording) (see Fig. 12–1C).

Ventricular depolarization occurs in three successive steps that give rise to the generation of three vectors (the expression of three dipoles) (see Fig. 12–11B and C). Each of the three vectors explains a deflection of the QRS.¹³ Ventricular depolarization starts at three different sites in the left ventricle⁶: the areas of the anterior and posterior papillary muscles and a midseptal area (Fig. 12–13A and C). At almost the same time, the right ventricle begins its depolarization. These three initial depolarization sites in the left ventricle (black asterisks in Fig. 12–13A) dominate the small initial forces of the right ventricle and originate a joint depolarization dipole (vector), which is called the first vector (1) (Fig. 12–13B). This first vector is directed front-rightward and, generally, upward (see Fig. 12–11B), although in some patients, especially obese individuals, it may be also directed downward. Once this initial area in the left ventricle is depolarized, most of the right and left ventricular mass is depolarized at the same time, giving rise to a right depolarization vector (2r) and a left depolarization vector (2i). The sum of these vectors is directed toward the left, somewhat backward, and, generally, downward (see Fig. 12–11B) and is known as the second vector. In obese individuals, it is usually located around 0°. Finally, the last areas to depolarize in both ventricles (the areas with fewer Purkinje fibers—that is, the basal septal areas) originate a third vector (3), which is directed upward, somewhat to the right, and backward (see Fig. 12–11B). As stated, the union of the heads of these three vectors, which is merely a simplification of the union of the heads of all the instantaneous vectors originated during ventricular depolarization, represents the pathway that the electrical stimulus follows when it depolarizes the ventricles and is called the QRS loop, which originates the QRS complex in the ECG (see Figs. 12–1B and 12–11). The loop-hemifield correlation explains the morphology of QRS in different leads (see Figs. 12–2 and 12–10).

Finally, ventricular repolarization takes place, and is mainly represented by the repolarization of the left ventricular free wall. From a physiologic viewpoint, the subendocardial area is relatively underperfused (physiologic hypoperfusion). As previously stated (see Fig. 12–1B), this explains the positivity in the last part of repolarization in the leads facing the left ventricle and the negativity in the opposite leads (VR). The initial pathway followed by repolarization does not produce any expression in the ECG and is recorded as an isoelectric ST segment. Later, when a repolarization dipole is formed, the union of the heads of all instantaneous vectors originates the T loop, which is recorded as a T wave in the ECG (see Fig. 12–12C).

After the T wave, which represents the end of ventricular systole, and prior to onset of the next atrial systole, an isoelectric line, corresponding to the resting phase of all cardiac cells, is recorded (see Fig. 12–12C). Sometimes, a small wave, called a U wave, which forms part of the repolarization process, is recorded after the T wave (see Fig. 12–1A).

Electrocardiographic Devices and Recording Techniques

The equipment for performing an ECG is a conventional device that produces a paper recording of tracings. However, we are now immersed in the digital era, and in the field of ECG, this has given rise to the need for functional programs and devices that are portable, versatile, and interactive.

This means working in an integrated fashion with management and quality assurance in the different health institutional settings, independently of the types of technologies used. Currently, these systems allow us to work online and advance rapidly in the implementation of telemedical services, including diagnosis performed by an expert in real time through the Internet, which can be applied to multiple scenarios that otherwise would not have the same quality of assistance because of the geographic location or the associated economic costs.

The recording technique includes the following: (1) connection of the device and preparation of the patient, (2) placement of electrodes at their correct locations, (3) adjustment of baseline tracing, (4) checking of calibration, and (5) ensuring that the tracing is correctly obtained according to the Einthoven law (II = I + III). The misplacement of the electrodes and other clinical mistakes, that will be discussed later, may lead to several errors.

Systematic Interpretation of the Electrocardiogram

The names given to different waves and intervals are shown in Fig. 12–1, and the different morphologies of P, QRS, and T waves are shown in Fig. 12–14. In some pathologic situations, other waves may be recorded—a delta wave in Wolff-Parkinson-White pattern, an epsilon wave in arrhythmogenic right ventricular cardiomyopathy, and a J wave in hypothermia (Osborne wave) or other situations such as early repolarization.

Different items should be routinely assessed when reading an ECG. The following sections will briefly discuss the most important aspects of the normal limits of these ECG parameters, as well as some relevant aspects of the abnormal values. For more detailed information, consult Bayés de Luna² and McFarlane et al.¹⁴

Lastly, to perform accurate measurements of different waves or intervals, it is sometimes necessary to use magnifying glasses and/or calipers.

Heart Rate

Sinus rhythm at rest normally ranges from 60 to 90 beats/min. Considering that the recording paper is divided into 5 mm rectangles (when the paper runs at 25 mm/s, it is equivalent to 0.20 second) each containing five smaller (1 mm) rectangles, the following methods may be used to measure the heart rate: (1) Observe the number of R-R intervals occurring every 6 seconds (every five 5 mm rectangles represent 1 second) and multiply this number by 10 (this is the best method when arrhythmia is present; or (2) use a proper ruler (Fig. 12–15).

Rhythm

Rhythm can be normal sinus rhythm or ectopic rhythm. Sinus rhythm produces positive P waves in leads I, II, aVF, and V₂ to V₆; positive or positive/negative waves in III and V₁; positive or negative/positive waves in aVL; and negative waves in aVR. Figure 12–16 explains, according to rotation of the loop (counterclockwise in sinus rhythm or clockwise in ectopic rhythm), the different bimodal morphologies in the P wave in both cases. For example, when the axis of the P loop is located around +30°, the morphology of a sinus P wave in III will be positive/negative (counterclockwise rotation of P wave; see Fig. 12–16) and negative/positive in case of ectopic rhythm (clockwise rotation). The same correlation is useful to explain the morphologies of P, QRS, or T waves seen in other leads.

The morphology of the P wave in reentrant atrioventricular (AV) junctional arrhythmias and the morphology of atrial activation waves in other supraventricular arrhythmias will be discussed in Chap. 84.

PR Interval and Segment

PR interval is the distance from the beginning of the P wave to the beginning of the QRS complex (see Fig. 12–1A). The procedure for this measurement is shown in Fig. 12–15. Normal PR interval values in adult individuals range from 0.12 to 0.20 second (up to 0.22 second in the elderly and even < 0.12 second in the newborn). Longer PR intervals are seen in cases of AV block, and shorter PR intervals are seen in preexcitation syndromes.

The PR segment is the distance from the end of the P wave to the QRS onset and is usually isoelectric. However, with intracardiac recordings, the depolarization of the bundle of His may be seen. Figure 12–1C shows the different spaces of PR interval taken with this technique. Sympathetic overdrive may cause a descent in PR segment that forms part of an arch of circumference together with an ascendant ST segment (Fig. 12–17C). In pericarditis and other diseases affecting the atrial myocardium, as in atrial infarction, PR segment in lead II is depressed or, more frequently, an elevated PR segment in lead VR may be seen (see Fig. 12–25).

QT Interval

The QT interval represents the sum of depolarization (QRS complex) and repolarization (ST segment and T wave). Often, particularly in the presence of a flat T wave or a U wave, it is difficult to measure the QT interval properly. It is commonly accepted that this measurement should be performed using a consistent method to ensure that equivalent measurements are taken when the QT interval is studied sequentially.¹⁵ The most suitable method entails considering the end of repolarization in a point where the tangent line following the descendent slope of the T wave crosses the isoelectric line (see Fig. 12–15A and B). It is necessary to correct the QT interval by heart rate (QTc). In clinical practice, QTc may be measured with a ruler (see Fig. 12–15C), and it is considered that its duration should not exceed approximately 10% of the value corresponding to heart rate.

A long QT interval may be found in congenital long QT syndrome¹⁶ (see Figs. 12–15B and 12–80) and also in diseases such as heart failure and IHD, in some electrolyte imbalances, and after the intake of different drugs. It is considered that a drug should not increase the QTc by more than 30 milliseconds and that a change of 60 milliseconds may result in torsade de pointes (TdP) and sudden cardiac death. Nevertheless, TdP rarely occurs unless the QTc exceeds 500 milliseconds.¹⁷

A short QT interval can be found in cases of early repolarization, digitalis effect, hypermagnesemia, acidosis, and other conditions and, rarely, in some genetic disorders associated with sudden death (short QT syndrome)¹⁸ (see Figs. 12–15B and 12–81A). Usually, in the latter case, the QT is less than 300 milliseconds and the T wave is peaked and tall, especially in V₂ to V₃. In symptomatic short QT syndrome, the distance from the J point to the T peak is shorter (≅100 milliseconds) compared with asymptomatic short QT (see Fig. 12–15B).

P Wave

The P wave is the atrial depolarization wave (see Figs. 12–1 and 12–12A). In general, its height should not exceed 2.5 mm, and its width should be less than 120 milliseconds. It is rounded and positive, except in VR, but may be positive/negative in V₁ and, III and rarely, in II and VF, and negative/positive in VL according to the loop-hemifield correlation (see Figs. 12–1, 12–12A, and 12–16).

The P wave changes seen in atrial abnormalities are discussed later (see Atrial Abnormalities).

QRS Complex

The QRS complex corresponds to ventricular depolarization. Its morphology varies in the different leads according to the loop-hemifield correlation (see Figs. 12–10 and 12–12B). An example of this correlation in a heart without rotations is shown in Fig. 12–12B, and different morphologies of QRS are also presented in this figure. Figure 12–14 shows the different letters used to express these morphologies. The normal values of QRS and T in 12 leads according to age can be found elsewhere.^2,14

The width of the Q wave should be less than 0.10 second, and the R wave height should not exceed 25 mm in leads V₅ and V₆ or 20 mm in leads I and VL, although in VL, heights of more than 15 mm are usually abnormal and suggest left ventricle hypertrophy (see Table 12–3).

Furthermore, the Q wave must be narrow (< 0.04 second) and of quick recording and should not exceed 25% of the following R wave, although some exceptions exist, mainly in leads III, VL, and VF. The presence of abnormal Q waves suggests myocardial necrosis but may be also observed in many other circumstances (see Table 12–14).

ST Segment and T Wave

The T wave, together with the preceding ST segment, is generated during ventricular repolarization (see Figs. 12–1B and 12–12C). According to the loop-hemifield correlation, in the adult, the T wave is positive in all leads, except VR (because the T loop is located in the negative hemifield of that lead) but being the slope up slower that the slope down. The T wave may be negative or flattened or, occasionally, slightly positive in V₁, and sometimes, it may also be flattened or slightly negative in V₂ and even in V₃ in women and black people. In III and VF, the T wave may be flattened or even slightly negative. In children, a negative T wave of characteristic morphology is seen in the right precordial leads (children’s repolarization) (Fig. 12–17E).

Under normal conditions, the ST segment is isoelectric (see Fig. 12–1) or shows only a small downward slope (< 0.5 mm) with ascendant inclination. A small upward slope (1–2 mm) that is convex in relation to the isoelectric line is often seen in V₁ and V₂ as a normal variant (Fig. 12–17B).

Normal ST Variant Patterns

Different examples of normal ST variants are shown in Fig. 12–17, and some of these are discussed here.

The rSr′ in V1 pattern (Fig. 12–17G) can be observed in healthy people, especially in those with pectus excavatum or when the V₁ and V₂ leads are located in a higher position (second intercostal space). In both cases, the P wave in V₁ is negative. This pattern should be differentiated from the type 2 Brugada pattern (see Fig. 12–81C). The best method to differentiate all cases that presents r’ in V₁ is through morphologic features of r’ the Baranchuk algorithm¹⁹ (see Fig. 12–82).

The pattern of early repolarization (see Fig. 12–17D) entailing a 2 to 4 mm ST-segment elevation with downward convexity, often preceded by the J wave, is particularly evident in midprecordial leads. In early repolarization, the ST-segment elevation normalizes with exercise. Recently, it has been described²⁰ that many patients with idiopathic ventricular fibrillation present with this pattern but with the malignant type (rectilinearly or slightly descendent ST in inferior leads especially). However, from an epidemiologic point of view, its presence only represents a slight increase in the risk of sudden death (11 in 100,000 vs 4 in 100,000 in a control group)²¹ (see Fig. 12–81D).

Abnormalities of Repolarization

The abnormalities of repolarization encompass T-wave disturbances and abnormal deviations of the ST segment. We will only comment here that these abnormalities may be primary or secondary. Primary T-wave disturbances are changes in the normal T wave (T wave higher than normal or flat/negative) resulting from delayed TAP in some regions of the heart or some layers of the left ventricle as a consequence of ischemia or other causes (see Changes in T Wave). Primary ST deviations (abnormal elevations and depressions of ST) are caused by shape changes of TAP in some regions of the heart or some layers of the left ventricle as a consequence of ischemia or other causes (see Changes in ST Segment). Finally, secondary abnormalities of repolarization (usually ST depression and asymmetric negative T waves) are caused changes in shape and duration of TAP at the ventricles secondary to conduction delays (bundle branch block) (see Fig. 12–32) and/or ventricular hypertrophy (strain pattern) (see Fig. 12–30). Sometimes a mixed pattern (primary and secondary) may be seen.

All changes in the normal morphology of the T wave and the abnormal deviations of the ST segment secondary to IHD or present in other heart diseases or situations will be discussed later in this chapter.

U Wave

Occasionally, especially in bradycardia, a small wave called the U wave is observed after the T wave, particularly in persons with vagal predominance and in the elderly.² Normally, it is smooth and has the same polarity as the T wave (see Fig. 12–1). If it is not, it is always pathologic. Characteristically, the presence of negative U waves in right precordial leads is suggestive of left anterior descending coronary artery occlusion.

Assessment of the QRS Electrical Axis in a Frontal Plane

When the QRS axis (ÂQRS) is at +60°, the morphology is positive in I, II, and III because the loop and its main vector fall in the positive hemifield of I, II, and III. However, it is more positive in II according to the rule that II = I + III (the same rule may be followed for the assessment of P and T wave axis) (Fig. 12–18A). When the axis shifts to the left from +60° to +30°, up to –120°, the QRS complexes become negative starting from lead III, changing from positive to isodiphasic and then from isodiphasic to negative for each shift of 30° to the left in the electrical axis (Fig. 12–18B). As the ÂQRS shifts to the right from +60° to +90°, up to –120°, the complexes again become negative, but starting from lead I, they change from positive to isodiphasic and then from isodiphasic to negative with every 30° shift to the right in the electrical axis (Fig. 12–18C). All of these changes may be explained by the correlation between loop location and the projection on different hemifields of I, II, and III in the frontal plane (see Fig. 12–18A, right) and the projection of the ÂQRS vector in leads I, II, and III in the Einthoven triangle (see Fig. 12–18A, left). Using this procedure, the ÂQRS may be calculated with a precision of 30°. To locate the ÂQRS more precisely, the morphology in the VR, VL, and VF leads has to be checked. For instance, a positive R wave in I, II, and III means that ÂQRS is approximately +60°. The QRS is exactly at +60° if QRS in VL is isodiphasic (). According to the loop-hemifield correlation, if the complex in VL is more positive than negative (), the major part of the loop is still located in the positive hemifield of VL, and the ÂQRS is between +30° and +60°; if the QRS complex is more negative than positive (), the greatest part of the loop is in the negative hemifield of VL, and the ÂQRS is between +60° and +90°. The same can be done for VR and VF. With such an approach, at a glance, we can calculate the ÂQRS precisely (with an error < 10%).

The normal values of the P, QRS, and T electric axes are as follows. In more than 90% of normal cases, the P axis (ÂP) is located between +30° and +70°. The ÂQRS generally ranges from 0° to +80°, although it can be shifted somewhat more to the left in obese people and more to the right in very lean people. The T axis (ÂT) generally ranges from 0° to +70°. The ÂT is shifted more to the left when the ÂQRS is also shifted to the left. Nevertheless, on certain occasions, when ÂQRS is shifted to the right, the ÂT is between 0° and –30°.

Figure 12–19A shows a normal ECG with an ÂQRS of approximately +30° (the voltage of the QRS complex in lead III is isodiphasic and is positive in I and II).

Electrocardiographic Changes with Age

Infants, Children, and Adolescents

The most important features of the ECG in healthy children (Fig. 12–19B) compared with normal adults can be summarized as follows:

1. There is a faster heart rate and a shorter PR interval.

2. Because of the physiologic right ventricular hypertrophy in infants, the heart is usually vertical, with an ÂQRS shifted to the right and negative or bimodal T waves in V₁ to V₃-V₄ and with a characteristic morphology (children’s repolarization) that may be present until adolescence, particularly in females. The QRS loop goes to the left before going backward, which explains why the V₆ but not the V₁ morphology (there is higher R in V₁ compared with q in V₆) looks like the adult’s morphology. Sometimes an rsr′ pattern is observed in V₁. In infants, especially if they are postterm, even R or qR patterns can be seen at birth with a somewhat positive T wave. The Rs pattern persists for a time, perhaps for years or even until adulthood (see Table 12–2). However, the T wave usually becomes flattened or negative in the days following birth.

3. In some adolescents, an R wave with high voltage in precordial leads (SV₂ + RV₅ > 60 mm) without the existence of left ventricular enlargement may be seen.

4. Sometimes, there is an evident increase in the heart rate with inspiration.

The Elderly

The following phenomena can be considered age-related ECG variants of the elderly (Fig. 12–19C)^2,22:

1. A slower heart rate and longer PR interval (normal up to 0.22 second).

2. Occasionally, an ÂP shifted more to the right is caused by pulmonary emphysema, with the S wave present in lead V₆ and an ÂQRS shifted to the left (from 0° to –30°).

3. A poor r progression from V₁ to V₃, probably resulting from septal fibrosis. This can produce differential diagnosis issues with septal necrosis.

4. Unspecific alterations of repolarization (slightly depressed ST segment and/or flattened T wave). A U wave is frequently present, particularly in the intermediate precordial leads.

5. An increase in the incidence of atrial and ventricular blocks as well as left ventricular hypertrophy.

Rotations of the Heart

In a heart without apparent rotation (intermediate position), the ÂQRS is situated around +30° (between +20° and +40°). The ECG of a heart featuring these characteristics is shown in Fig. 12–19A.

Nevertheless, the heart may present isolated or combined rotations, the most characteristic of which are rotations on the following axes: (1) the anteroposterior axis (vertical or horizontal heart; see VL and VF leads in Fig. 12–20A), and (2) the longitudinal axis (dextro- or levorotation; see precordial leads in Fig. 12–20B).

Certain special types of rotation explain some variants of the normal pattern, such as S_I, S_II, and S_III (differential diagnosis with right ventricular enlargement [RVE])²³ and QR in III (differential diagnosis with previous inferior infarction). In the latter, the Q wave diminishes during inspiration. For more information, consult Bayés de Luna.²

Technical Mistakes during Electrocardiogram Recording

In addition to the knowledge of normal and pathologic patterns, the correct interpretation of ECG recordings requires the use of standard recording procedures.²² The most frequent errors are a consequence of errors in limb and precordial electrode placement, artifacts, and inadequate filter application.

1. Misplacement of electrodes may lead to a diagnosis of ectopic rhythm or dextrocardia (Fig. 12–21A) or to an incorrect diagnosis of Q wave MI.²⁴

2. As a consequence of some artifacts, a false diagnosis of flutter or ventricular tachycardia may be made (Fig. 12–21B).

3. Finally, the inappropriate use of some filters may induce abnormal ST elevation in V₁ to V₂ or may suppress an otherwise visible J wave (Fig. 12–21C).

For more information, consult García-Niebla et al.²⁵

Electrocardiography Criteria of Abnormality

For many conditions, such as blocks and arrhythmias, the ECG is the gold standard for the diagnosis. For other conditions, such as enlargement and some types of ischemia, the ECG diagnostic criteria have to be contrasted against some other gold standard (echocardiography, cardiovascular magnetic resonance [CMR], or coronary angiography). In these cases, it is important to know the diagnostic accuracy of the ECG criteria. For that purpose, we have to study the sensitivity, specificity, negative predictive value, and positive predictive value of each criterion. For further information, consult Bayés de Luna.²

Atrial Abnormalities

The term atrial abnormalities encompasses atrial enlargement, atrial blocks, and abnormalities of atrial repolarization. The following general principles should be emphasized ²:

1. Atria become dilated rather than hypertrophied.

2. P-wave voltage is influenced by extracardiac factors that increase it (eg, hypoxia, increased sympathetic tone) or decrease it (eg, emphysema, atrial fibrosis).

3. The atrial repolarization wave is usually hidden within the QRS complex (see Fig. 12–1).

Right Atrial Enlargement

Right atrial enlargement (RAE) is mainly present in patients with congenital and valvular heart diseases affecting the right side of the heart and in patients with cor pulmonale (see Fig. 12–23B and C) (Figs. 12–22 and 12–23). Typically, the P wave is increased in voltage but not in length.

The diagnostic criteria of RAE are based on P-wave abnormalities (positive voltage of P ≥ 2.5 mm in II and/or positive voltage > 1.5 mm in V₁). These criteria have low sensitivity and somewhat higher specificity.²

Left Atrial Enlargement

Left atrial enlargement (LAE) is seen in patients with mitral and aortic valvular disease, IHD, hypertension and some cardiomyopathies (see Figs. 12–22C and 12–23D). Typically, the P wave is bimodal in some leads and +/– in V₁ with evident final negative deflection.

The diagnostic criteria of LAE are as follows²: (1) P wave with a duration 0.12 second or longer, especially seen in leads I or II, generally bimodal, but with normal height, and (2) diphasic P wave in V₁ with evident final negative deflection of at least 0.04 second of duration because the second part of the loop is directed backward due to LAE (see Fig. 12–23D, horizontal plane). These two criteria have good specificity (close to 90%; few false-positive cases) but discrete sensitivity (< 60%; more false-negative cases). The +/– P-wave morphology in II, III, and VF with a P of at least 0.12 second is a very specific criterion and has a high positive predictive value (100% in valvular heart disease and cardiomyopathies). However, it has a low sensitivity and low negative predictive value for LAE.

Biatrial Enlargement

The most important diagnostic criteria of biatrial enlargement are the following (see Fig. 12–23E)²: (1) P wave in lead II that is taller (≥ 2.5 mm) and wider (≥ 0.12 s) than normal, (2) first part of P wave that is positive and peaked in V₁-V₂ (positive mode > 1.5 mm) with a slow negative deflection (width ≥ 4 mm), (3) signs of LAE with right ÂP; the opposite case is not valid because the ÂP can be on the left side in isolated RAE of patients with congenital heart diseases, and (4) presence of atrial fibrillation (AF) along with QRS changes suggestive of RAE. Frequently, more than one criterion is found (P ≥ 120 ms in FP + P± in V₁ with first part peaked and a slow negative mode).

Interatrial Blocks

The concept of a block means that in a certain part of the heart (sinoatrial junction, atria, AV junction, or ventricles), the electrical stimulus encounters overall significant difficulties for its conduction. A first-degree or partial block (P-IAB) occurs when conduction is slow but the stimulus passes through the area but with slow conduction. A third-degree or advanced block (A-IAB) is when the stimulus does not pass completely through the zone but perhaps could pass at a very slower heart rate; therefore, we call it advanced, not complete. A second-degree block is when the stimulus sometimes passes and sometimes does not².

The conduction delay in interatrial block occurs between the right and left atria (Fig. 12–24). Although usually associated with LAE, it may also exist as an isolated finding in cases of pericarditis, IHD, old age, and other conditions.²²

In P-IAB, the stimulus reaches the left atrium via the normal pathway, but with certain delay. The diagnostic criterion of P-IAB is a P wave with a duration at least 0.12 second in the frontal plane.^26,27,28 The P-wave length and, consequently, the bimodal morphology of the P wave seen in lead II (which is the most typical lead to detect an isolated P-IAB) are similar to the P wave occurring in LAE. In fact, the delay in interatrial conduction, rather than the left atrium dilation, generally explains the morphology observed with an LAE. However, the morphology of the P wave in the horizontal plane, especially in V₁, is usually different. In case of isolated interatrial block (eg, pericarditis), the second part of the loop is not directed so much backward because there is no LAE, and consequently, the P-wave morphology in V₁ is positive or only presents a small negative part.

In A-IAB (see Fig. 12–24),^27,28 the stimulus does not reach the left atrium via the normal path but by retrograde left atrial activation.²⁸ The diagnostic criteria of A-IAB are as follows²⁷: (1) P wave with a duration of at least 0.12 second and +/– in II, III, and VF, and (2) often, P wave +/– in V₁ to V₃ or V₄. The prevalence of A-IAB increases with age, being very low before 50 years of age and approaching 25% in centenarians in whom an increase of dementia and stroke associated with A-IAB has been detected.²² This type of block is frequently accompanied, especially when it is present in patients with advanced heart disease, by supraventricular arrhythmias, particularly AF and/or atrial flutter (Bayes syndrome).^29,30,31

Abnormalities of Atrial Repolarization

The atrial repolarization wave (ST-Ta) usually has a polarity opposed to the P wave and is not visible because it is hidden within the QRS complex. In cases of great sympathetic overdrive in normal individuals, an ST-Ta depression may be seen (see Fig. 12–17C). Moreover, in some cases of remarkable atrial enlargement, pericarditis, or atrial infarction, especially in the presence of a long PR interval, shifts of ST-Ta may be observed (Fig. 12–25).

Ventricular Enlargement

Morphologies of ventricular enlargement are secondary to a hypertrophy rather than to dilatation, unlike what occurs in the atria. Slight or even moderate degrees of enlargement of either of the ventricles, mainly the right, or of both at the same time, may not produce abnormalities in the ECG.

The superiority of echocardiography over the ECG for the diagnosis of ventricular enlargement, mainly of the left ventricle, is evident (the sensitivity is much higher, and the specificity is similar). However, when a ventricular enlargement is diagnosed with an ECG, the accuracy of the ECG is greater than that of the echocardiogram in predicting heart disease evolution and prognosis.

For details on the diagnosis of RVE and/or left ventricular enlargement (LVE) combined with ventricular block (QRS complex duration > 120 milliseconds), the reader is referred to Bayés de Luna² and McFarlane and Lawrie.¹⁴

Right Ventricular Enlargement

RVE is found especially in congenital heart diseases, valvular heart diseases, and cor pulmonale. The changes produced by RVE move the loop rightward and either frontward or backward. This is more a result of a delay of activation of the right ventricle, rather than an increase of the right ventricle mass (which generally never overcomes the mass of the left ventricle). Figure 12–26 shows the changes that RVE may produce in ventricular activation expressed as ventricular loops and how these changes may explain the different ECG patterns. Basically, a large part of the QRS loop is going to the right (presence of S in V₆) but sometimes with an anterior loop (R or RS in V₁) and occasionally with a posterior loop (rS or QS in V₁).

The lower part of Fig. 12–26 shows three cases of RVE of different etiologies (see legend) in which the ECG pattern in V₁ (with more or less R wave) is related more to RVE degree than to RVE etiology.

The ECG criteria (low sensitivity, high specificity) most frequently used for the diagnosis of RVE are shown in Table 12–1.

• Morphology of V₁: (1) Morphologies with dominant or exclusive R wave in V₁ are very specific but not very sensitive (< 10%) for the diagnosis of RVE because the loop that gives rise to them (anterior and to the right; see Fig. 12–26B–D) is observed in a small number of cases with RVE (especially in some congenital heart diseases such as pulmonary stenosis). In these cases, the prominent R in V₁ is also seen in V₂ and V₃. Moreover, the repolarization presents an ST-segment depression and a negative and asymmetric T wave (strain pattern; Fig. 12–26, Fig. 12–27A), except in newborns because they may present with an exclusive R wave with a positive T wave. On the contrary, in cases of pulmonary stenosis from tetralogy of Fallot, the morphology in V₁ is similar to that in isolated pulmonary stenosis, but in V₂, there is an rS morphology. Other causes that may present a dominant R pattern in V₁ must be ruled out (Table 12–2). (2) The presence of rsR′ is especially typical of atrial septal defect and sometimes may also be seen in moderate pulmonary stenosis. (3) The rS or even the QS morphology in V₁, with RS in V₆, may often be observed in the early phases of RVE, or with rS in V₆ in advanced cases of chronic cor pulmonale (see Fig. 12–26B).

• Morphology of V₆. The presence of evident forces directed to the right, which are expressed as an evident S wave in V₅ to V₆, is one of the most important ECG diagnostic criteria (see Figs. 12–26 and 12–27; see Table 12–1).

• S_I, S_II, S_III. This morphology is frequently seen in chronic cor pulmonale with QS pattern in V₁ and RS pattern in V₆. Note that this pattern may be secondary to positional changes or it may simply be a normal variant.² Nonetheless, an abnormal P wave favors the diagnosis of RVE (see Fig. 12–27B).

• Electrical axis: ÂQRS of at least +110°. Inferoposterior hemiblock, vertical heart, and lateral infarction must be ruled out. This criterion is quite specific (> 95%) but relays low sensitivity.

The combination of more than one of these criteria increases the diagnostic possibilities. The differential diagnosis of exclusive or predominant R wave in V₁(R, Rs, or RSR′ pattern) is shown in Table 12–2.

The ECG signs indicative of right ventricle acute overload (decompensation of cor pulmonale or pulmonary embolism) are as follows³²:

• Change in the ÂQRS (> 30° to the right of its usual position)

• Transient negative T waves, sometimes very evident in the right precordial leads

• S_I, Q_III with negative T_III pattern (McGinn-White pattern) in the frontal plane and an RS or rS pattern in V₆ (Fig. 12–28)

• Appearance of a complete right bundle branch block morphology, often with ST-segment elevation

The latter two criteria are highly specific but have low sensitivity for important pulmonary embolism (see Cor Pulmonale). Nevertheless, the clinical setting and comparison with previous ECGs are important to make a differential diagnosis of both processes.

Left Ventricular Enlargement

LVE is found in hypertension, IHD, valvular heart disease, cardiomyopathies, and some congenital heart diseases. In general, in patients with LVE, the maximum QRS vector of the loop increases its voltage and is directed more posteriorly than normal (Fig. 12–29). This explains why the QRS complex negativity predominates in the right precordial leads (see Fig. 12–29A–C). Occasionally, the maximum vector is not directed posteriorly (it is located close to 0°). This implies a tall R wave that is seen even in V₂, especially in cases of apical hypertrophic cardiomyopathy (see Fig. 12–29E).

The ECG pattern changes during disease evolution. The pattern of “strain” appears more in relation with the duration of the disease than with the presence of different types of hemodynamic overload. However, in aortic valve disease, a q wave in V₅ to V₆ is found more frequently in long-standing aortic regurgitation than in aortic stenosis (Fig. 12–30). The disappearance of the q wave in V₆ is probably more related to interstitial septal fibrosis, a substrate of partial left bundle branch block, than to hemodynamic overload³¹ (see Figs. 12–29 and 12–30).

The presence of noticeable signs that are suggestive of LVE (high voltage of the QRS together with inverted ST-T wave—strain pattern) in an asymptomatic patient without heart murmur or hypertension suggests hypertrophic cardiomyopathy. The ECG does not correlate with the gene mutation,³³ but there are two at least ECG changes suggestive of hypertrophic cardiomyopathy: (1) the striking negative T waves (see Fig. 12–29E) and (2) the presence of a deep and narrow q waves (see Fig. 12–29D).

Sometimes the LVE pattern, at least partially, and especially the changes in ST-T may be resolved in a few months with pharmacologic treatment, as occurs in hypertension or after surgery (valvular heart disease).

Various diagnostic criteria exist for LVE,^34,35,36,37 with varying sensitivity and specificity, as shown in Table 12–3. It is usually possible to diagnose an LVE through an ECG in patients with severe hypertension, whereas this diagnosis is difficult in asymptomatic normotensive adults. However, the value of the ECG diagnostic criteria shown in Table 12–3 is lower in hypertensive patients. For these patients, the following criterion is useful: sum of the QRS voltage of 12 ECG leads greater than 120 mm.³⁷

Biventricular Enlargement

The ECG diagnosis of biventricular enlargement is even more difficult than that of the enlargement of just one ventricle. This is true because the increased opposing forces of both ventricles often counterbalance each other or the notable predominance of the enlargement of one ventricle completely masks the enlargement of the other. Therefore, even those criteria that have fair specificity only have a moderate sensitivity.³⁸

The following ECG patterns are used for the diagnosis of biventricular enlargement:

• Tall R wave with s in V₅ and V₆, rSR′ pattern in V₁, and P wave of biatrial enlargement

• Tall R wave in V₅ and V₆, with an ÂQRS shifted to the right (≥ 90°). The presence of an inferoposterior hemiblock associated with LVE and an asthenic body build must be ruled out.

• Small S wave in V₁, deep S wave in V₂, predominant R wave in V₅ and V₆, and an ÂQRS shifted to the right in the frontal plane or an S_I–, S_II–, S_III–type morphology

• The presence, especially in the elderly, of QRS complexes within normal limits but with significant repolarization abnormalities (negative T wave and depression of the ST segment), mainly when the patient presents with AF. This type of ECG can be found in the elderly with advanced heart diseases and biventricular enlargement.

Ventricular Blocks

Blocks at the ventricular level can occur on the right side (Table 12–4) or on the left side (Table 12–5). They can affect the entire ventricle (global block) or only part of it (zonal or divisional block). The blockage of electrical impulses, in the ventricles or in the whole heart (see interatrial blocks) is called a first-degree (partial) block when the impulses pass through the area but with a delay; second-degree block when the stimulus sometimes passes and sometimes does not; and third-degree (advanced) block when the passage of stimulus is blocked.² The term advanced is preferred over complete, because one can not know whether the stimulus would pass at a very lower heart rate. Second-degree block is known as aberrancy of conduction and is explained in Section 6).

In global ventricular block, the conduction delay usually occurs in the proximal part of the right or left branches. For this reason, the global ventricular block is known as bundle branch block.

Advanced or third-degree bundle branch blocks, both right and left, have the following characteristics^2,13:

1. Depolarization of the ventricle corresponding to the blocked branch that occurs transseptally, beginning at the contralateral ventricle. This phenomenon explains the QRS complex widening (≥ 0.12 second) caused by a small number of Purkinje fibers in the septum and the peculiar QRS complex morphology, both in right and left bundle branch blocks, caused by the loop-hemifield correlation (Figs. 12–31, 12–32, and 12–33).

2. Diagnosis is mainly based on data provided by the horizontal plane leads V₁ and V₆ and the frontal plane lead VR.

3. Slurring at the end of the small of the QRS is usually opposed to the T wave.

4. Septal repolarization dominates over that of the left ventricular free wall and is responsible for the ST-T changes.

5. In general, the anatomic changes are more diffuse than the ECG expression.

Partial or first-degree bundle branch block presents a QRS complex with a duration less than 120 milliseconds, which gives rise to morphologies that are sometimes indistinguishable from some patterns seen in homolateral ventricular enlargement (see Fig. 12–32).

Zonal or divisional left blocks (hemiblocks) have been studied more in depth, both from the anatomic and electrophysiologic viewpoints, compared with right zonal blocks.

There are four intraventricular fascicles: right bundle branch, trunk of the left bundle branch, and the superoanterior and inferoposterior divisions of the left bundle. A block of the middle fibers (septal fascicle) of the left bundle can probably also occur (see later). Therefore, in addition to the block of one fascicle, blocks of two fascicles (bifascicular block) or three fascicles (trifascicular) may occur.

Advanced or Third-Degree Right Bundle Branch Block

In third-degree right bundle branch block (RBBB), the depolarization of the right ventricle occurs entirely through the septum from the left side, originating the formation of vectors 3 and 4 and causing the global change in QRS loop. The classic ECG morphologies, which result from the loop-hemifield correlation in the frontal and horizontal planes, are shown in Fig. 12–31. The repolarization is from right to left at septal level in case of proximal RBBB or at peripheral level in case of peripheral block. Consequently, the T wave is negative in V1 and positive in V6.

The blockade location, both in advanced and partial block, is usually proximal. Nevertheless, a block located in the distal part of the branch or in the right ventricle Purkinje network is often seen in some congenital heart diseases (eg, Ebstein anomaly, atrial septal defect, following surgery for tetralogy of Fallot) and in some cardiomyopathies (eg, arrhythmogenic right ventricular cardiomyopathy), giving rise to morphologies similar to those of classic complete or partial bundle branch block but with some specific patterns. See Table 12–2 concerning the differential diagnosis of tall R wave in V₁.

The presence of advanced RBBB does not usually represent a poor prognosis in the absence of overt heart disease. However, it predicts an adverse outcome if it appears after acute MI, concomitantly with acute dyspnea (resulting from probable pulmonary embolism) or during an ACS caused by occlusion of the left anterior descending coronary artery (LAD) proximal to the first septal branch, which is the branch perfusing the right bundle.²

The ECG diagnostic criteria are as follows (see Fig. 12–31 and Table 12–4):

• QRS of at least 0.12 second with midfinal slurring

• V₁: rSR′ with slurred R wave and a negative T wave

• V₆: qRS with evident S wave slurring and positive T wave

• VR: QR with evident R wave slurring and negative T wave

• T wave with its polarity opposed to QRS slurring

Partial or First-Degree Right Bundle Branch Block

In first-degree RBBB, the delay of the activation in the entire ventricle is less important. This explains why the QRS is narrower and the morphology shows less prominent forces to the right (see Fig. 12–32).

The ECG diagnostic criteria are as follows:

• QRS complex duration less than 0.12 second

• Morphology with rsR′ or rsr′ pattern or even rs (RS) in V₁, final r in VR, and s wave in VL and V₆, but with fewer notches and slurring²

A similar pattern may be seen in some cases of RVE, as in atrial septal defect, because of a delay of activation of some parts of the right ventricle as a result of the enlargement (see Fig. 12–26C).

Right Zonal Blocks

Experimentally,^23,39 the injury of the fibers of the superoanterior zone of the right ventricle often causes ECG morphologies of the S_I, S_II, S_III types.²³ Less frequently, the injury of the inferoposterior zone may originate S_I R_II R_III pattern. In clinical practice, these morphologies are difficult to differentiate from normal variants or RVE.² The changes in P and T waves may help in the differential diagnosis. The S_I, R_II, R_III morphology must also be found in the inferoposterior hemiblock.

Advanced or Third-Degree Left Bundle Branch Block

In third-degree left bundle branch block (LBBB), the left ventricle activation is through the septum from the right side and differs completely from normal activation. This transseptal activation causes the formation of four vectors, which are characteristic of this type of block and explain the global change in the QRS loop. The classic ECG morphologies, caused by the loop-hemifield correlation in the frontal and horizontal planes, are shown in Fig. 12–33. The repolarization in case of more advanced LBBB is directed from left to right with the T loop opposite in direction to QRS (discordant LBBB).⁴⁰ This explains the negativity of T wave in leads I, V5, and V6. The T wave is normally located opposite slurring and dominant deflection of the QRS complex. In cases of less advanced LBBB, the T wave is often concordant with the QRS (T-wave positive in leads I, V5, and V6). Also, the concordant repolarization may be explained, in a few cases, by associated IHD.

The morphology of the distal blocks is similar to that of the classic proximal complete left ventricular block but with more significant final slurrings of the QRS complex. Wherever the global block is located (proximal or distal), when the delay is significant, an R-wave morphology in V₆ and a QS complex in V₁, with a QRS of at least 0.12 second, are generated.

The presence of LBBB by itself necessitates the exclusion of heart disease and periodic surveillance. In the acute and chronic phase of IHD, as happens with RBBB,² LBBB is a marker of poor outcomes. The polarity of T wave concordant with QRS is associated with better prognosis and less comorbidity.⁴⁰ The presence of LBBB pattern after acute infarction⁴¹ or that appears with exercise has worse prognosis.⁴² Also associated with a worse prognosis is the LBBB that appears after transaortic valve implantation,⁴³ and in heart failure, especially, if associated with AF.^44,45 Finally, the morphology of QRS in V3 may help to diagnose the etiology of LBBB.⁴⁶

The diagnostic criteria of third-degree LBBB are as follows (see Fig. 12–33 and Table 12–5):

• QRS of at least 0.12 second, sometimes greater than 0.16 second, especially with midportion slurring

• V₁: QS or rS with a tiny r wave and positive T wave

• I and V₆: single R with its peak after the initial 0.06 second

• VR: QS with positive T wave

• T-wave polarity is opposed to the QRS complex slurrings in more than two-thirds of cases (discordant LBBB)

Partial or First-Degree Left Bundle Branch Block

In first-degree LBBB, the activation delay in the entire ventricle is less significant (see Fig. 12–32) but is sufficient to conceal the first vector responsible for formation of r in V₁ and q in V₆. Because of this delay, this first vector is counteracted by the initial activation vector of RV and is not formed. The presence of septal fibrosis, demonstrated by biopsy in patients with aortic valve replacement also explains the lack of first vector (q wave) in left ventricular leads (I, VL, V6).²

The ECG diagnostic criteria are as follows:

• QRS complex duration less than 0.12 second

• A QS complex or a tiny r wave in V₁ and a single R wave in I and V₆

• A similar morphology with disappearance of q in V₆, which may be seen because of the presence of septal fibrosis³¹

• Often present in LVE

Zonal (Divisional) Left Ventricular Block: Hemiblocks

These zonal blocks are much more known and well studied than the zonal blocks of the right ventricle, especially after the studies of Rosenbaum school.^47,48 In zonal (divisional) left ventricular block, the stimulus is blocked in either the superoanterior or inferoposterior division of the left branch (hemiblocks) (Figs. 12–34 and 12–35; see also Fig. 12–13C). Only the ECG criteria of well-established (advanced) superoanterior and inferoposterior division of the left branch (superoanterior or inferoposterior hemiblocks according to Rosenbaum) will be discussed.

A change in left intraventricular activation occurs in both hemiblocks. As a consequence, the blocked area is depolarized with certain delay, which explains the typical ECG changes that can be seen.

Superoanterior Hemiblock

Figure 12–34A shows the activation of the left ventricle in superoanterior hemiblock (SAH) or superoanterior divisional block, and the loop-hemifield correlation in the frontal and horizontal planes. In Fig. 12–34B, a typical example of an SAH is depicted, as well as differences with the S_I, S_II, S_III pattern (see legend for Fig. 12–34). The diagnosis of SAH may be made only using the ECG criteria.

The diagnostic criteria are as follows:

• RS complex duration less than 0.12 second

• ÂQRS deviated to the left (mainly between –45° and –75°). For the differential diagnosis, it is necessary to rule out other diseases with leftward ÂQRS, such as inferior necrosis, type II Wolff-Parkinson-White (WPW) syndrome, pacemaker implant, and S_I, S_II, S_III pattern (see Fig. 12–34B)

• I and VL: qR; in advanced cases with slurring, especially in the descending part of the R wave

• II, III and VF: rS with S_III > S_II and R_II > R_II.

• An S wave seen up to V₆, with intrinsic deflection in V₆ less than VL

Inferoposterior Hemiblock

In inferoposterior hemiblock (IPH) or divisional block, typical ECG morphology and clinical conditions, mainly the absence of RVE and asthenic habit, have to be present. It is also usually considered that evidence of left ventricular abnormalities must exist. Figure 12–35A shows the location of the block and the activation of the left ventricle in the case of an IPH or inferoposterior divisional block. In this figure, the typical ECG morphology, in frontal and horizontal planes, explained by the loop-hemifield correlation, can also be observed. In this case, the diagnosis is assured if the ECG pattern appeared abruptly, in the absence of another possible explanation (see Fig. 12–35B and legend).

The diagnostic criteria are as follows:

• QRS complex duration less than 0.12 second

• ÂQRS shifted to the right (between +90° and 110° or more for some authors and +140° for others)

• I and VL: RS or rS

• II, III, and VF: qR; in advanced cases with slurring, especially in the descending part of the R wave

• Precordial leads: S wave up to V₆, with an intrinsic deflection time in V₆ less than VF

Middle Fibers Block

Block of the middle fibers, called septal block by the Brazilian school, probably produces RS morphology often with slurrings in S in V₂ and a prominent R wave in V₁. To confirm this diagnosis, the ECG pattern has to be transient. However, this intermittent RS pattern may be also recorded in the evolutive pattern of progressive RBBB.²

Bifascicular Blocks

In this section, the ECG criteria of the three most characteristic bifascicular blocks—complete (or advanced) RBBB plus SAH, complete RBBB plus IPH, and bilateral bundle branch block—are discussed.

The diagnostic criteria of advanced RBBB plus SAH are as follows (Fig. 12–36A and B):

• QRS complex duration greater than 0.12 second

• QRS complex morphology: The first portion is directed as in SAH, upward and toward the left, whereas the second portion is directed as in complete global RBBB, anteriorly and toward the right (see Fig. 12–45A). If there is a significant left delay, it can counteract the right forces, giving rise to anterior but left final forces. Therefore, a tall R wave can be seen in V₁, but without an S wave in I and, occasionally, also in V₆. In this case, a complete LBBB appears to exist in the frontal plane and a complete RBBB in the horizontal plane (“masked” block)⁴⁹ (see Fig. 12–36B). Sometimes this pattern may be transient.⁵⁰ A masked bifascicular block usually indicates a worse outcome, even if a pacemaker is implanted.²

The diagnostic criteria of advanced RBBB plus IPH (Fig. 12–36C) are as follows:

• QRS complex duration greater than 0.12 second

• QRS complex morphology: The first portion of the QRS complex is directed downward, as in IPH, whereas the second portion is directed anteriorly and toward the right, as in complete RBBB. This diagnosis requires the presence of some clinical conditions, as occurs in isolated IPH (see Inferoposterior Hemiblock section).

In bilateral bundle branch block, there is an alternation of complete RBBB and LBBB morphologies in the ECG. This diagnosis requires a pacemaker to be implanted.

Trifascicular Blocks

There are several types of trifascicular blocks, but their coverage is beyond the scope of this chapter (see Bayés de Luna²). The most frequent features are as follows:

• RBBB alternating with the block of one of two left bundle branch divisions (Rosenbaum-Elizari syndrome)^2,47 (Fig. 12–37)

• Once the block of three fascicles has been confirmed by an ECG (RBBB alternating with superoanterior and inferoposterior fascicle blocks) (see Fig. 12–37), a pacemaker should be implanted as soon as possible because the patient may suddenly develop a paroxysmal AV block.²

• Bifascicular blocks with a long PR segment. Note that a long PR segment may also be caused by a block at a proximal location (bundle of His); thus, electrophysiologic studies are required to confirm their occurrence.

Ventricular Preexcitation

Ventricular preexcitation is considered to exist when the electric stimulus reaches the ventricles earlier (early excitation) than via the normal conduction system.² Early excitation can be explained by the presence of Kent bundles, which are accessory pathways with accelerated conduction that connect the atria with the ventricles (WPW-type preexcitation).⁵¹ The conduction through the Kent bundles may be anterograde, retrograde, or in both directions. In rare cases, there are also other anomalous bundles (atypical preexcitation). In more than 80% of these cases, they consist of long slowly conducting atriofascicular or AV bypass tracts. Other rare anomalous bundles are the fasciculoventricular and nodofascicular tracts, which include the formerly named Mahaim fiber.⁵² In addition, the presence of an accelerated AV conduction or the rare occurrence of an atrio-His bundle may cause a short PR-type preexcitation.⁵³

The clinical importance of preexcitation lies in its association with supraventricular tachycardias; its potential to trigger malignant ventricular arrhythmia; and the risk of being confounded with other processes, especially in case of a WPW preexcitation (see the following section).

Wolff-Parkinson-White–Type Preexcitation

WPW-type preexcitation is diagnosed by an ECG that shows a short PR interval plus QRS-T abnormalities, primarily caused by a slurred upstroke at the beginning of the QRS complex known as the delta wave (Fig. 12–38).

• Short PR interval: The PR interval generally lasts between 0.08 and 0.11 second. However, a WPW-type preexcitation may also occur with a normal PR interval in the following instances: (1) conduction block in the anomalous pathway; (2) preexcitation far from the sinus node (left side), frequently with a long anomalous pathway; and (3) atypical preexcitation presenting normal PR intervals. The only way to corroborate that the PR interval is shorter, in order to confirm the diagnosis, is by making a comparison with the baseline ECG tracing without preexcitation.

• QRS-T abnormalities (Figs. 12–39 and 12–40): The QRS complexes show an abnormal morphology with a width greater than the baseline QRS complex (often > 0.11 second) and a characteristic initial slurring (delta wave). These features are secondary to the initial activation through the contractile myocardium where there are few Purkinje fibers. Different degrees of preexcitation (or delta wave) may be observed (see Fig. 12–38A–C).

The QRS complex morphology in the different surface ECG leads depends on the location of the epicardial zone of earlier excitation. The vector of the first 20 milliseconds on the ECG (first vector of the delta wave, which can be measured in the ECG) is located at different sites on the frontal plane according to where the earlier ventricular epicardial excitation occurred first (see Fig. 12–39). As a result, the WPW-type preexcitation may be divided into four types (see Fig. 12–39A–D).² Examples of these four types are displayed in Fig. 12–40.

Different algorithms have been proposed to predict the location of the anomalous pathway.⁵⁴ One of the most popular was published by Milstein et al.⁵⁵

Repolarization is altered except in cases with minor preexcitation. Repolarization abnormalities are secondary to the alteration in depolarization, and the pathology is more severe when the T-wave polarity opposes that of the preexcited R wave, as when the preexcitation is greater (see Fig. 12–38A–C).

Changes in the degree of preexcitation are frequent, and they may be abrupt or progressive (Fig. 12–41). Preexcitation can increase if the conduction of the stimulus through the AV node is depressed (eg, vagal maneuvers, drugs), or in contrast, preexcitation can decrease if the AV node conduction is enhanced (eg, physical exercise).

Regarding the differential diagnosis of WPW-type preexcitation, type-A and type-B preexcitations (see Fig. 12–39) should be differentiated with LBBB (see Fig. 12–40A and B); type-C preexcitation (see Fig. 12–39) with inferolateral infarction, RBBB, or RVE (see Fig. 12–40C); and type-D preexcitation (see Fig. 12–39) with lateral infarction or RVE (see Fig. 12–40D). In all of these cases, the short PR interval and the presence of a delta wave are decisive data for the diagnosis of WPW-type preexcitation.

Paroxysmal arrhythmias in patients with a WPW ECG pattern constitute the WPW syndrome. Patients with WPW preexcitation frequently present macro-reentrant paroxysmal tachycardia. They show a narrow QRS because the stimulus is activating the ventricles via normal AV conduction and a QRS-P′ less than P′-QRS (orthodromic tachycardia) (Fig. 12–42D). In cases of paroxysmal reentrant tachycardia exclusively resulting from the intranodal (junctional) circuit, the P′ mimics the end of QRS or is hidden within it (see Chap. 84). Indeed, approximately 40% to 50% of paroxysmal tachycardias are attributed to a reentry, including an accessory pathway. In a few cases, the anterograde arm of the circuit is via the anomalous Kent bundle, and the QRS complex is wide during the tachycardia (antidromic tachycardia) (see Chap. 84).

AF and flutter episodes are also more frequent than in the general population. This is attributed to the rapid retrograde conduction, via the abnormal pathway, of a ventricular premature complex that may reach the atria during a vulnerable atrial period. Another possibility is that paroxysmal tachycardia triggers other arrhythmias, such as AF/flutter. The risk of these arrhythmias is two-fold. First, AF/flutter in patients with an ECG pattern of WPW can be mistaken for sustained ventricular tachycardia, with the associated consequences (Fig. 12–42A and B). Indeed, the differential diagnosis by surface ECG between atrial flutter in WPW syndrome and ventricular tachycardia is especially difficult (see Fig. 12–42B) The second potential risk, in the presence of AF/flutter, is that the accessory pathway may transmit to the ventricle more stimuli than normal, facilitating the possibility of reaching the ventricle in its vulnerable period. Consequently, this can result in ventricular fibrillation and sudden death (Fig. 12–42C).^56,57

Atypical Preexcitation

In general, the atypical preexcitation that presents reentrant arrhythmias occurs through a long atriofascicular or AV pathway. The other bundles, which include the classic Mahaim fiber, are general bystanders and do not participate in the circuit in reentrant tachycardia.

The ECG in normal sinus rhythm is normal in the majority of cases or presents minimal preexcitation (absence of q wave in leads I and V₆ and rS morphology in lead III).⁵⁸

The arrhythmias are less frequent and usually of reentrant junctional type. During preexcited tachycardia, the anterograde conduction is through an atypical bundle (antidromic tachycardia) with retrograde conduction by right bundle branch. Therefore, the morphology is usually of LBBB with left ÂQRS. Generally, there is a late R/S precordial transition zone (at V₄ to V₆), and in the cases of antidromic tachycardia through Kent bundle and LBBB pattern, the transition zone occurs earlier (R/S in V₂ to V₃) (see Chap. 84).

Short PR-Type Preexcitation

Short PR-type preexcitation, described by Lown et al,⁵³ is evidenced by a short PR interval without changes in QRS morphology (see Fig. 12–38).

By a surface ECG, it is impossible to differentiate between a hyperconductive AV node, the most frequent, and preexcitation occurring via an atrio-His bundle, which bypasses the AV node slow conduction area and, therefore, does not modify the QRS complex morphology. However, it is easy to differentiate them by intracavitary ECG because this will show a short AV interval in case of a hyperconductive AV node and a short HV interval or even an absent H deflection in the presence of an atrio-hisian tract.

The association with arrhythmias and sudden death is less frequent in short PR-type preexcitation than in the WPW-type preexcitation. Nevertheless, in case of AF, the ventricular response may be very fast because of the presence of a short PR-type preexcitation, and this can be a problematic situation.

Ischemia

From the electrophysiologic point of view, according to the importance and duration of ischemia, the TAPs of myocardial cells present the following changes: (1) delay of the repolarization that corresponds to primary T-wave changes, (2) formation of “low-quality” TAP (change in shape) that corresponds to changes of ST, and (3) lack of formation of TAP that explains the presence of the Q wave.

From the surface ECG point of view, the most typical ECG changes resulting from ischemia are as follows^{2,59,60,61,62,63,64,65,66}: (1) primary T-wave changes that include T waves that are wider, peaked, and/or taller and T-wave flat or negative; (2) primary ST deviations, including ST segment shifts (up and down) from baseline; and (3) changes of QRS (Q wave of necrosis and reciprocal patterns) and changes in the mid-late part of QRS (fractionated QRS).

However, ischemia may induce many other changes, including the following:

1. Prolongation of QT interval that may be considered a part of the changes in the T wave. It is present in the first moments of acute transmural ischemia when there is an increase in duration of TAP of the endocardium, making the T wave wider and taller and lengthening the QT interval.⁶⁷

2. Changes in P wave, especially in atrial infarction⁶⁰

3. Distortion of QRS complex, especially in very severe ischemia^{59,60,61,68,69}

4. Appearance of different types of bradyarrhythmias (sinus bradycardia and AV block); tachyarrhythmias (especially AF and premature ventricular contractions, which may lead to ventricular tachycardia and unfortunately to ventricular fibrillation); and intraventricular block (eg, RBBB when the LAD occlusion is proximal to the first septal branch that perfuses the right bundle)⁶⁰

First, we will explain how the different models of experimental occlusion of the coronary artery produce different ECG changes. Then, discussion will focus on the changes in T wave, ST segment, and QRS complex that are induced by experimental and clinical ischemia and their electrophysiologic mechanisms. Finally, the ECG changes that appear during the evolution of different clinical settings of IHD, their diagnostic criteria, and the most important clinical and prognostic implications of these ECG changes will be discussed.

Changes in Electrocardiography under Different Models of Experimental Ischemia Caused by Coronary Occlusion

After experimental coronary occlusion of the LAD in dogs, performed with an open chest and with the electrodes located in the pericardial sac, Bailey and La Due⁷⁰ found (Fig. 12–43, 1) that the first ECG change induced by ischemia was negativity of the T wave, which was called ECG pattern of ischemia. This was followed by an ST-segment elevation that he called ECG pattern of injury, and finally, a Q wave appeared, which was called ECG pattern of necrosis (see Fig. 12–43, 1). This terminology was accepted by several groups.¹³ Thus, it was considered that in patients with IHD, the presence of a negative T wave represented a grade of acute ischemia that was less severe than an ST-segment elevation.

However, when the experiment to occlude the artery was performed in awake dogs with an unopened chest,⁷¹ the changes in the ECG were similar to the ECG changes that occurred in coronary spasm or transmural MI in humans (Fig. 12–43, 2). The negative T wave only appeared as a late manifestation, when the acute ischemia was already vanishing or had already disappeared. Therefore, a flat or negative T wave represents postischemic changes rather than the first phase of acute ischemia. Thus, it is a frequent mistake to consider the presence of a negative T wave as an indicator of active ischemia.

Changes in T Wave

Morphology of T-Wave Changes and Electrophysiologic Experimental Mechanisms

It has been demonstrated by different experimental open chest methods that when ischemia or similar effect is induced (eg, by cooling the tissue)^7,8,72 in the epicardium or in a part of tissue close to the recording electrode, the TAP of this zone presents a delay of repolarization without changing its shape, and a flat or negative T wave appears (Fig. 12–44B). When the same procedure is performed in the endocardium or in a part of tissue far from the electrode, a delay of repolarization is induced in this zone, and a peaked and usually taller T wave appears (Fig. 12–44A).

The electrophysiologic explanation of these changes in T wave (more positive T wave and flat or negative T wave) caused by experimental subendocardial and subepicardial ischemia, that are found in these areas that present delay of repolarization (prolonged TAP), may be explained (see Fig. 12–44) by two concepts: the summation of TAPs of subendocardium and subepicardium (the longer is located in the ischemic zone), and the concept of ischemic vector. According to these concepts, the area with delayed TAP (named ischemic area), either in the subendocardium or in the subepicardium, was not fully repolarized at the time when the other area was already repolarized. Therefore, it still presents negative charges. As repolarization starts in the zone that is more perfused, the sense of repolarization (see Fig. 12–44) will face the ischemic zone. However, as this zone presents more negative changes, a flow of current having a vectorial expression is generated, going from more ischemic to less ischemic zones. Then, the vector of ischemia with the positive change in the head is directed away from the ischemic area and originates a more positive T wave in experimental subendocardial ischemia and a flat or negative T wave in experimental subepicardial ischemia. Note that the vector of necrosis is also directed away from the necrosis area (see Fig. 12–46), but on the contrary, the vector of the area with greater ischemia (injury area) is directed toward this area (see Figs. 12-49 and 12–50).

T-Wave Changes in Patients with Ischemic Heart Disease

In clinical settings, the explanation of T-wave changes caused by ischemia is in some aspects different (see Electrocardiography in Ischemic Heart Disease).^72,73

• Peaked, symmetrical, wider, and/or taller T wave. This morphology, accompanied by a lengthening of the QT interval, appears in the beginning of acute transmural ischemia resulting from total coronary occlusion, sometimes accompanied by slight ST depression. As a consequence of the occlusion, there is first subendocardial involvement. Different factors explain the susceptibility of the subendocardium to the development of ischemia. These include the greater dependence of this region on diastole perfusion and the greater degree of energy expenditure of this area during systole.⁶⁰

This initial subendocardial ischemia, also demonstrated by contrast-enhanced cardiovascular magnetic resonance (CE-CMR),⁷⁴ produces a delay of repolarization in this area and, as a result, a lengthening of the TAP of the subendocardium, explaining the increase of the QT interval and usually also the presence of wider, peaked, symmetric, and/or taller T waves (see Fig. 12–44A; Fig. 12–45).

As mentioned earlier, the results of animal experiments performed with unopened chest⁷¹ coincide with the ECG changes that often appear in different clinical settings of transmural or near-transmural ischemia that are the consequence of total or near-total coronary occlusion, such as coronary spasm (see Fig. 12–45),^75,76 hyperacute phase of ST-segment elevation ACS (STE-ACS; see Fig. 12–60A), and percutaneous coronary intervention (PCI).⁶³ In the majority of cases, after a short period of peaked T wave with QT prolongation, an evident ST-segment elevation appears (see Fig. 12–60A). Later on, if ischemia persists, the ECG pattern may evolve to Q-wave MI.

In addition, a tall, positive, and symmetrical T wave is frequently seen in a chronic phase of Q-wave MI. This is the expression of the mirror image of a negative T wave. The most frequent leads where this is seen are I, VL, and V₁ to V₃ as mirror images of inferolateral involvement (Fig. 12–46, 2).

• Flat or negative T wave. This morphology may be seen in different clinical situations of IHD, including the following:

  • A negative T wave, often deep, appears in the resolution process of Q-wave MI. Characteristically, this presents a mirror pattern in the frontal plane (ie, if negative in II, III, and VF, then positive in I and VL). This is important to perform differential diagnosis with the negative T wave of pericarditis that usually does not present mirror pattern (see Fig. 12–77C and D). The T wave in these cases may be very deep (see Fig. 12–52, 1) but usually decreases or even disappears with time (see Fig. 12–56). The appearance of a negative T wave in Q-wave MI is at least partially secondary to changes of depolarization (Q wave).

  • A negative T wave can be the manifestation of an open artery if it appears after fibrinolytic treatment or PCI. It may also be deep (see Fig. 12–52, 2).

  • A negative T wave, frequently very deep, may appear in V₁ to V₄ in the evolutive phase of STE-ACS because of LAD subocclusion, which is usually tight. This situation is considered a marker of impending MI,⁷⁷ and if the patient has had angina in the previous few hours, it may evolve to complete occlusion (see Fig. 12–60B). However, currently with the new treatment of acute phase, this pattern often does not evolve and is the expression of an at least partially open artery, or if it is more closed, then there is very good collateral circulation. Therefore, it is necessary to perform PCI but not emergently.

  • A negative flat T wave may also be seen in the evolutive phase of non–ST-segment elevation ACS (NSTE-ACS). The morphology of the T wave is usually symmetric and not deep. In fact, patients with NSTE-ACS and negative T waves have better a prognosis, probably because this indicates that the clinical acute phase of ischemia is over (see Fig. 12–62C). However, relatively often still coronary subocclusion even tight exists. Therefore, currently, we also recommend nonemergent PCI in these cases.

  • Finally, a flat/negative T wave may be seen as a residual pattern in chronic IHD.

None of these clinical situations represents acute active ischemia. Rather, they indicate postischemic changes.

The involvement of the left ventricle wall is usually transmural in the presence of a flat/negative T wave. No clinical situation caused by ischemia presents exclusively with subepicardial involvement. Therefore, the theories that explain the presence of flat/negative T waves in experimental clinical ischemia (see Fig. 12–44B) are not useful in the clinical setting. Probably, the negative T wave seen in these clinical cases, associated with acute transmural ischemia (ST↑) that has subsided, may be explained by the mechanisms shown in Fig. 12–47.

Differential Diagnosis

A long list of conditions, apart from IHD, may present with changes in the T wave (symmetric-wide and/or taller than normal, flat, or negative).^2,60 Tables 12–6 and 12–7 summarize the most important conditions, and Fig. 12–48 presents some of the most striking examples.

Notably, taller T waves of ischemic origin are very transient (see Figs. 12–45 and 12–60A). The persistent high-voltage T wave, seen in V₁ to V₂ in cases of chronic lateral infarction (see Fig. 12–46B), is an indirect (mirror) pattern, rather than a direct pattern, of tall T wave caused by ischemia.

In cases with a negative T wave, chronic pericarditis is the most important condition for a clinical differential diagnosis because this disease also presents with pain in the acute phase. The negative T wave of pericarditis (see Fig. 12–77) is usually more extensive, does not present mirror pattern in the frontal plane, and is less intensive (negative) than in cases of IHD, especially in post–Q-wave MI (see Fig. 12–46).

Changes in ST Segment

Morphologies of ST Deviations and Electrophysiologic Experimental Mechanism

The area of the myocardium with significant and persistent ischemia (called the area of injury) presents an evident diastolic depolarization. From an experimental point of view, two injury currents are generated as a consequence of an important ischemia—one occurring during systole and the other during diastole. However, the injury current generated during systole is the only one considered because the ECG equipment is adjusted by alternating current amplifiers to maintain a stable isoelectric baseline during diastole, despite the existence of diastolic depolarization that would change the QT interval. The area with systolic injury current presents a “low-quality” TAP (eg, slower upstroke, lower voltage, smaller area).

The ECG pattern of the ST segment during an important diastolic depolarization caused by an experimental injury induced in open chest animals varies depending on the location of the injury. Thus, an ST-segment depression is observed if the injury affects the subendocardium, whereas an ST-segment elevation is recorded if the experimental injury is transmural. These two situations may be explained considering the concepts of TAP summation and the vector of severe ischemia.^2,14,72,73

1. Figure 12–49 shows how the sum of the TAPs of the subendocardium and the subepicardium with a TAP of subendocardium with slower upstroke and smaller size may explain the presence of ST-segment depression following a subendocardial injury (ECG pattern of subendocardial injury; see Fig. 12–49A) or of ST-segment elevation after transmural injury (ECG pattern of transmural ischemia that is known as ECG pattern of injury).

2. On the other hand, according to the concept of severe ischemia (injury vector) when the zone with low-quality TAP (injured zone) is on the subendocardium (Fig. 12–50A), at the end of depolarization (end of systole), this zone presents fewer negative charges. Consequently, a flow of current exists from the zone with more negative charges to the low-quality TAP zone with fewer negative charges. This originates an injury vector with fewer negative charges (relatively positive) in the head, which points to the zone with low-quality TAP, or the injured zone. If this zone is limited to subendocardium, therefore, the injury vector points toward the subendocardium (injured zone) and is expressed as an ST-segment depression in the surface leads (see Fig. 12–50A). Likewise, when the injured zone is more important and becomes transmural, this zone is relatively less negative (relatively positive), and the injury vector that starts in the subendocardium points toward this zone and is recorded in the surface lead as an ST-segment elevation (Fig. 12–50B).

ST-Segment Deviations in Patients with Ischemic Heart Disease

See also Electrocardiography in Ischemic Heart Disease. ST-segment depression appears in the following circumstances:

1. In patients usually with preexistent IHD who present with NSTE-ACS without total occlusion of the artery. In these cases, there is important involvement of subendocardial flow, as a consequence of sudden but not total cessation of blood flow. The ST-segment depression tends to normalize from the acute to the subacute phases (Fig. 12–51A; see legend).

2. In patients with IHD who, as a consequence of exercise or stress, have an increased demand for flow, but where it cannot be supplied because there is a fixed stenosis and a decrease in subendocardial flow. In these cases, different types of ST-segment depression may appear (Fig. 12–51B). As the stenosis is fixed, there is no total occlusion and thus no transmural ischemia.

ST-segment depression in clinical practice usually appears because the subendocardium is more vulnerable to ischemic damage than the rest of the ventricular wall, and thus, it may be affected without transmural involvement if the occlusion is subtotal. When the subendocardial/subepicardial ratio is reduced to approximately 40%, the flow ratio decreases dramatically. This occurs during exercise and mental stress and also when there is an increased left ventricle end-diastolic pressure.⁶³ In these cases, the TAP of the subendocardium resulting from extensive and already existent ischemia shows a change in its shape but not a prolongation of the TAP duration, as happens in the first minutes of total occlusion in patients usually without previous ischemia. This explains why, in one case, a taller T wave appears (see Fig. 12–44B), whereas in the other case, an ST-segment depression occurs (see Fig. 12–49B).

ST-segment elevation appears in transmural ischemia, often after a period of taller and wider T wave (STE-ACS). Usually, these patients do not present with a previous important ischemia. The ST-segment elevation tends to normalize from the acute to the chronic phases (Figs. 12–52 and 12–53) (see Acute Coronary Syndrome with Narrow QRS: Evolutionary Changes).

To explain the appearance of this pattern, we have to consider that even the injured zone is transmural and the surface electrodes located closer to the epicardium are recording the ECG pattern as in the case of the experimental subepicardial injury that presented a change of the TAP shape in the subepicardial layers and a less negative charge (relatively positive) (see Fig. 12–50B, right).

Because the injury patterns develop at the end of the depolarization, which corresponds to the end of generation of QRS at the beginning of repolarization, the ECG expression of injury starts during the first part of the ST segment. Therefore, in cases of relevant injury, there is an elevation of the S wave, which persists along the whole ST segment and usually encompasses a T wave that becomes negative (see Electrocardiogram in Acute Coronary Syndrome with ST-Segment Elevation).

An ST-segment elevation may also be recorded in other clinical settings that are not related to atherothrombosis, such as isolated coronary spasm^75,76 and Takotsubo syndrome (TTS)^60,78,79 (see Electrocardiographic Changes in Ischemia Not Related to Atherothrombosis).

The leads that face the head of the vector of injury in transmural acute ischemia record an ST-segment elevation, whereas the leads facing the tail of the injury vector record an ST-segment depression. The global study of direct and mirror images of the ST deviations in different leads is crucial to correctly diagnosis the location of the injury and to understand the corresponding prognostic implications (see Fig. 12–53 and Electrocardiogram in Acute Coronary Syndrome with ST-Segment Elevation).

Differential Diagnosis

The list of conditions that, apart from IHD, may present an ST-segment elevation or depression is large.^2,60

Tables 12–8 and 12–9 summarize the most important conditions of ST-segment depression and elevation not caused by IHD. On the other hand, Fig. 12–54 shows the most typical variants that present with ST-segment elevation or depression. In the first phase of pericarditis, a disease that also presents with chest pain, the most striking change is an elevation of the ST segment. Many conditions can also present with ST-segment depression, although usually small. The most important factor to take into account is the correlation of the clinical symptoms with the deviations of the ST segment. The diagnostic criteria for considering the deviations (ups and downs) of ST suggestive of ACS and all their prognostic implications will be discussed later (see Acute Coronary Syndrome with Narrow QRS: Evolutionary Changes).

Changes in QRS

Q Wave of Necrosis: Electrophysiologic Mechanisms

The activation of the subendocardial area is very fast because the density of Purkinje fibers is so great that the electrodes located in this area do not record any positive deflection (1 and 2 in Fig. 12–55A). Later on, the subsequent activation of the other part of the left ventricle wall records progressively greater R waves until a unique R wave is recorded from the epicardium (5 in Fig. 12–55A).

In the presence of MI, the diastolic depolarization in the infarcted area is so important that it cannot be excited and does not originate a TAP. Therefore, a Q wave is generated when the infarcted area surpasses the subendocardium and affects part of the myocardium that is depolarized within the first 40 to 50 milliseconds⁶⁴ (Fig. 12–55B and C).

The Q wave of necrosis may be explained by two theories. The first theory is the theory of the electrical window of Wilson. According to this theory, the transmural infarcted area acts as an electrical window and, consequently, the electrode that faces that area records the negativity of the intracavitary QRS (Q wave of necrosis) (see Fig. 12–55B).

The second theory is the theory of the infarction vector. This theory proposes that the infarcted area generates a vector (infarction vector) that has the same magnitude, but opposite direction, to the one that would have been generated at the same zone in the absence of infarction. Therefore, the infarction vector moves away from the infarcted area (see Fig. 12–46). As we have stated, the ventricular depolarization changes that originate the Q wave occur when the infarction area is depolarized within the first 40 milliseconds of ventricular activation.

Fractioned QRS

Anomalies in the mid-to-late part of QRS, such as slurring, rsr′, and low voltage in the left precordial leads (Fig. 12–55D), are called fractioned QRS.^65,66 Fractioned QRS may occur as a consequence of alterations of ventricular activation after 40 to 50 milliseconds (basal segments according to standardized myocardial segmentation nomenclature by Cerqueira et al⁸⁰). The fractioned QRS may appear either isolated or with a Q wave. It may be caused by a necrotic area that does not involve the whole wall (see Fig. 12–55D).

The MI with fractioned QRS forms part of the non–Q-wave MI, which will be discussed later (see Myocardial Infarction Without Q Wave or Equivalent).

Changes in QRS in Patients with Ischemic Heart Disease

The appearance of the Q wave of necrosis in the evolutionary changes of STE-ACS is evidence that the patient has evolved to MI, involving an area that surpasses the subendocardium (see Fig. 12–55B). However, this is not an irreversible pattern. In fact, in a few cases, such as aborted MI and Prinzmetal angina, the presence of Q waves may be transient (see Electrocardiography in Ischemic Heart Disease).

The criteria to diagnose a Q wave as abnormal and the location of MI, according to the presence of the Q wave in different leads, are discussed later in Q-Wave Myocardial Infarction (see Table 12–11).

As with changes in the T-wave and ST-segment deviations, the presence of a mirror image of the Q wave of necrosis is also very important for diagnostic purposes. It is especially evident that, in patients with IHD, the presence of a prominent R wave in V₁ may be the only manifestation of abnormality. It is currently known that this R wave result from a lateral, not a posterior, MI (see Q-Wave Myocardial Infarction).^81,82

Recently, it has been reported that the presence of fractioned QRS provides a more accurate criterion for the diagnosis of necrosis than the existence of a Q wave.⁶⁵ However, it is necessary to match these morphologies with the clinical setting because they may also be seen in normal subjects.

The fractioned QRS also arises in other diseases or circumstances, such as ventricular aneurysm⁸³ and Brugada syndrome.⁸⁴

Differential Diagnosis

The specificity of the pathologic Q wave (Table 12–10) for diagnosing myocardial necrosis is relatively high. However, it must be noted that similar “Q” waves are seen in other processes. The diagnosis of acute MI is based not only on ECG findings, but also on the clinical setting, as well as on enzymatic changes. The presence of ST-T alterations accompanying a pathologic Q wave gives support to an IHD as the cause for this ECG pattern. However, as indicated later, in 5% to 25% of infarctions (higher incidence in inferior infarction), the Q wave disappears with time; thus, the sensitivity of the ECG for detecting old infarction is not very high (Fig. 12–56).

The principal causes of the pathologic Q wave, other than myocardial necrosis, are listed in Table 12–10.

Other Changes

As already stated, there are many other changes in the ECG parameters that are induced by ischemia. Some of them, such as lengthening of the QT interval, have already been discussed. For more information on other ECG changes, such as for instance those of the P wave in case of atrial infarction consult Bayés de Luna,² Surawicz,¹² and Bayés de Luna and Fiol-Sala.⁶⁰

Electrocardiography in Ischemic Heart Disease

Types of Myocardial Ischemia

Myocardial ischemia may occur by two different pathophysiologic mechanisms⁸⁵: (1) when there is an abrupt decrease in the blood flow, usually in patients with coronary atherothrombosis, after the rupture or erosion of a vulnerable plaque; or (2) when there is an increased myocardial demand, such as in exercise or other form of stress, while the supply from the coronary tree is insufficient, usually because of a fixed atherothrombotic stenosis. In both cases, there are other causes besides atherothrombosis that may explain the presence of ischemia.

In the following sections, ECG changes in different clinical settings related to a decreased supply, usually caused by atherothrombosis, will be discussed, including ACS and its evolutionary changes up to the chronic phase (Q-wave and non–Q-wave MI). In addition, the ECG changes found in clinical situations with an increased demand, such as in classic exercise angina, and in exercise stress test will be described. Finally, ECG changes induced by ischemic damage that does not have an atherothrombotic etiology will be discussed.

Acute Coronary Syndrome with Narrow QRS: Evolutionary Changes

The majority of these ACSs are caused by coronary atherothrombosis.^{86,87,88,89,90,91,92,93,94,95,96} They may be classified into two types, either with ST-segment elevation (STE-ACS) or without ST-segment elevation (non–STE-ACS). Angiographically, STE-ACS frequently (in > 90%) presents with a red clot that usually provokes a total occlusion of the vessel. In non–STE-ACS, however, the incidence of an angiographic white clot is lower (50%–70%) and does not produce a total occlusion. Even though this classification has clinical and therapeutic implications, not all cases of ACS that present STE-ACS show an ST-segment elevation as the most important ECG change, and another ECG pattern may be seen in the follow-up.^60,63 Furthermore, NSTE-ACS includes cases with ST-segment depression, negative T wave, and even normal ECG (Fig. 12–57).

Electrocardiography in Acute Coronary Syndrome with ST-Segment Elevation

In many cases, STE-ACS will evolve to an MI, usually of Q-wave type (see Figs. 12–52, 12–53, 12–56, and 12–57). However, thanks to modern treatments, more MIs can now be aborted (unstable angina) (see Fig. 12–52).⁶⁰

Electrocardiogram Patterns

The typical ECG pattern is the ST-segment elevation. However, as already stated, others patterns may be seen.

1. Typical ECG pattern: The new occurrence of an ST-segment elevation greater than 2 mm at 60 milliseconds after the J point in V₂ to V₃ and greater than 1 mm in other leads, present in at least two consecutive leads, is considered abnormal and evidence of acute ischemia in the clinical setting of ACS (presence of precordial pain or equivalents). Whenever possible, it is important to compare this pattern with previous ECGs.

   The morphology of the ST-segment elevation is diverse, but the most characteristic morphology is a concave shape with respect to the isoelectric line, especially in QRS complexes with rS morphology. The height of the ST-segment elevation may be very high, even more than 20 mm, and then it decreases along the acute-subacute phase. However, it may remain slightly elevated even in the chronic phase. Meanwhile, if the infarction is not aborted, the typical Q wave and negative T wave appear (see Fig. 12–52). In this case, the deepness of the T wave may be important (≥ 8 mm) and often disappears in the evolution (see Fig. 12–56).

   The ST-segment elevation prevails in the leads facing the affected zone as a direct ECG pattern. In general, in some opposed leads, the ST-segment depression is seen as an indirect ECG pattern (see Fig. 12–53). These same features also occur in the chronic phase, with a Q wave as the direct pattern and an R wave as the indirect pattern (see Fig. 12–47).

   In STE-ACS, the global evaluation of direct and reciprocal changes (ups and downs of ST) is important for detecting which artery is the culprit (the right coronary artery [RCA] or the left circumflex artery [LCX]) in the case of ST-segment elevation in II, III, and VF (Fig. 12–58; see also Fig. 12–53B), as well as to find the location of the LAD occlusion in the case of an ST-segment elevation in the precordial leads (Fig. 12–59; see also Fig. 12–53A).^87,88 Table 12-10 presents the more common pitfalls in the ECG interpretation of ACS.^89,90

2. Atypical ECG patterns (Fig. 12–60): The following patterns occur without ST-segment elevation as the most striking ECG change.^90,91,92

   • Presence of ST-segment depression in V₁ to V₃ that is more evident than the ST-segment elevation in II, III, VF, and/or V₅ to V₆ can occur. Nevertheless, the ST-segment elevation is usually present at least in some leads. This is a clear pattern equivalent to the ST-segment elevation that represents the mirror pattern of injury in the lateral zone. In this case, the occluded vessel is the LCX, and the ST-segment elevation may be seen in the back leads (see Fig. 12–60C). These patients must be treated as for STE-ACS.

   • In many occasions, but usually only for a very brief period, the first recorded ECG change during hyperacute ischemia is a change in the T-wave morphology (wider and peaked and sometimes taller) accompanied by an increase in the Q interval. Often, a mild ST-segment depression may be seen (see Fig. 12–60A). This pattern is not usually encountered in the emergency department because it is very transient and quickly followed by an ST-segment elevation (see Fig. 12–60A).

   It has been described (see the following section) that STE-ACS with grade 1 ischemia may only present stable changes of the T wave (wider, peaked, and tall). In our experience, this rarely occurs.

   • Relatively often, during the evolution of an STE-ACS caused by an LAD proximal occlusion, when the artery has been spontaneously opened or after treatment and once the pain has vanished, a negative T wave, usually striking (> 4–5 mm), may appear⁶⁰ (see T-Wave Changes in Patients with Ischemic Heart Disease). This is a sign of an open artery or, at least, that a very good collateral circulation exists, but it is not a guarantee that the problem is resolved. Sometimes, the artery may reocclude (see Fig. 12–60B) with the appearance of pain; changes in the ECG may occur; first pseudonormalization of the T wave may occur; and often, but not always, the ST-segment elevation can reappear. Therefore, patients with deep negative T waves in the right precordial leads who have had anginal pain in the previous hours should be submitted to coronary angiography as soon as possible. Nonetheless, this situation is not considered an emergency, as in the case of a clear ST-segment elevation.

Clinical and Prognostic Implications

The characteristics of the ST-segment elevation and mirror images have a great relevance for localizing the occlusion and area at risk, quantifying the ischemic burden, and detecting the grade of severity of ischemia.

• Location of occlusion and area at risk. Figure 12–59 shows in detail (see legend) the algorithm that allows one to predict the site of the LAD occlusion in the case of an ACS with STE in the precordial leads.⁸⁷ Figure 12–58 represents the algorithm to follow in cases of ACS with STE in the inferior leads, to distinguish between RCA or LCX occlusions.⁸⁸ In cases of RCA occlusion, there are increased chances of a proximal occlusion and right ventricle involvement if an ST-segment elevation in the right precordial leads (V₃R, V₄R) is present.⁹¹ We have found that the presence of isoelectric or elevated ST segment in V₁ is also useful. Furthermore, in case of a dominant RCA occlusion, an ST-segment elevation in V₅ to V₆ is usually seen.⁹²

• Quantification of ischemic burden. Despite the existence of some limitations, as occurs in cases of a proximal RCA occlusion that may mask the extension of ischemia because of present isodiphasic ST in V₁, the sum of the ST changes in all leads may estimate the myocardium at risk. More than 15 mm of ST-segment deviation usually represents a great area at risk.⁹⁶

• Grade of severity of ischemia. The morphology of QRS-ST may suggest the grade of ischemia. According to the Birnbaum-Sclarovsky ischemia grading system,^94,95 grade 1 ischemia presents a tall permanent T wave; grade 2 presents an ST-segment elevation without distortion of the QRS complex; and grade 3 presents an ST-segment elevation that sweeps the QRS complex upward, provoking a distortion of it (Fig. 12–61)^95,96

Electrocardiography in Acute Coronary Syndrome with Non–ST-Segment Elevation

NSTE-ACS includes cases of ACS that present new ST-segment depression (≥ 0.5 mm) and/or new flattened or negative T wave or negative U wave, in two or more consecutive leads, as the most prominent ECG changes. Obviously, all atypical cases of STE-ACS have to be excluded (see Fig. 12–60). Usually the deepness of the T wave in NSTE-ACS is not very important (< 3–4 mm). The deepness of the negative T wave in an atypical ECG pattern in case of LAD occlusion caused by ST-segment elevation MI (STEMI) is often more important (see Fig. 12–60B).

Electrocardiography Patterns and Clinical and Prognostic Implications

NSTE-ACS with ST-segment depression may present two patterns:

1. ST-segment depression in seven or more leads with more striking feature in V₃ to V₅, usually without positive T wave and with evident positive ST-segment elevation in VR as a mirror pattern (Fig. 12–62A). This ECG pattern represents a left main coronary artery subocclusion or equivalent (2–3 proximal vessels disease). These cases correspond to a circumferential involvement and have a worse prognosis. Therefore, they need urgent catheterization and administration of the appropriate treatment.

2. ST-segment depression in less than seven leads (regional involvement) usually with final positive T wave. These cases have a better prognosis. However, it is difficult to ascertain what artery is affected, although if the LAD is compromised, the ST-segment depression is usually located between V₂ to V₃ and V₄ to V₅. However, three-vessel disease may be present, often with normal basal ECG and only small changes in exercise testing (Fig. 12–62B).^2,60,63

The prognosis becomes worse in proportion with the number of leads involved and the importance of the ST-segment depression.

In NSTE-ACS with only a flat/negative T wave, the negative T wave is usually not deep (usually < 3 mm) and not very diffuse (Fig. 12–62C). As previously mentioned, the prognosis is better than in cases with ST-segment depression. However, a coronary subocclusion often exists and is generally significantly stenotic. Therefore, for these cases, we recommend nonemergent PCI.

Sometimes a normal ECG at entrance later evolves to a clear ACS (see Fig. 12–57). The ECG may remain normal in 5% to 10% of cases. The cases of NSTE-ACS that maintain a normal ECG usually have a good prognosis.

Acute Coronary Syndrome with Confounding Factors

In cases of left ventricular hypertrophy (LVH) with strain pattern and/or wide QRS complex, the diagnosis of ACS is more difficult, especially if concurrent with NSTE-ACS. Performing a study of sequential pattern may be useful.

In the presence of bundle branch block, the diagnosis is possible if the block is an RBBB. However, in some cases of STE-ACS in the presence of LBBB or other situations with wide QRS (pacemaker or WPW), it is also possible to visualize the ST-segment elevation (Fig. 12–63). Sgarbossa et al⁹⁷ reported that, in cases clinically compatible with acute MI in the presence of LBBB, the diagnosis is supported by the following criteria: (1) ST-segment elevation greater than 1 mm concordant with the QRS complex, (2) ST-segment depression greater than 1 mm concordant with the QRS complex, and (3) ST-segment elevation greater than 5 mm nonconcordant with the QRS complex. Similar criteria are also useful in the presence of a pacemaker.⁹⁸ On the contrary, the diagnosis of concurrent NSTE-ACS may be difficult using the surface ECG because there may be no visible changes or the changes may be confounded with the basal ones.

The scientific guidelines⁹⁹ consider new LBBB to be an ACS equivalent, in the right clinical context. However, when symptoms are atypical, primary PCI protocol activation for LBBB should be considered after a positive point-of-care troponin test 1 to 2 hours after symptom onset.

Acute MI and the need for emergent revascularization are relatively uncommon among patients who present with LBBB, especially when symptoms are atypical.⁹⁹

Electrocardiography in Chronic Myocardial Infarction with Narrow QRS

Before the era of reperfusion, it was relatively easy to predict the final Q-wave infarction pattern according to the acute phase of STE-ACS.⁶⁰ However, with current strategies of treatment, this is impossible because, if the treatment is started on time, the infarction may be aborted or faded.

Therefore, in patients with chronic Q-wave MI, it is presently impossible to know, from the correlation with the Q-wave pattern, the exact degree and location of the occlusion of the culprit artery. Nevertheless, the coronary lesion will be less than expected if reperfusion treatment had been applied. However, the location of the occlusion that produced the Q-wave MI could probably be predicted.

Furthermore in approximately more than 50% of cases with MI, the QRS complex does not show a Q wave of necrosis or equivalent (R in V₁-V₂). The MI without Q wave encompasses the classic non–Q-wave MI (ancient subendocardial MI) and many other types of MI, as shown in Table 12–13. Even in presence of an important ST-segment depression in the evolutionary phase, the ST morphology may completely recovered.

Until relatively recently, it was thought that MIs of exclusive subendocardial location existed and were electrically mute. However, the concepts that MI with Q wave is transmural and non–Q-wave MI only implies subendocardial compromise are false.^60,100 Now, it has been demonstrated that from a pathologic point of view, infarction of exclusive subendocardial location does not exist and that infarctions with predominant subendocardial involvement may or may not develop a Q wave.¹⁰⁰ However, some transmural MIs, such as those involving the basal areas of the left ventricle, do not exhibit a Q wave.¹⁰¹

CE-CMR is the ideal technique for infarct identification, transmural characterization, and infarct location and quantification.¹⁰² CE-CMR has demonstrated that the Q-wave MIs are larger than non–Q-wave MIs, although not always with a deeper transmural extension.^60,103 Therefore, the terms Q-wave MI and non–Q-wave MI are useful in the chronic phase from a prognostic point of view.

Table 12–11 presents the new definition of MI according a committee appointed by the European Society of Cardiology and the American College of Cardiology.¹⁰⁴

Q-Wave Myocardial Infarction

A. The evolution of the ECG pattern. Since the beginning of the ECG, the pattern of established transmural infarction has been classically associated with the presence of pathologic Q wave or equivalents. Figures 12–52 and 12–56 show the evolutionary appearance of the Q wave and negative T wave in cases of Q-wave MI. Now, if quick current treatment is established, the evolution of the ECG may be very different (see Figs. 12–52B and 12–57).

B. Diagnostic criteria of Q-wave MI. Table 12–12 shows the accurate criteria for the diagnosis of Q wave of necrosis.²

C. Location of Q-wave MI. A statement from the International Society for Holter and Noninvasive Electrocardiology presented a new classification of Q-wave MI based on the correlation between the Q wave and the infarcted area as assessed by CE-CMR (Figs. 12–64, 12–65, and 12–66).^{105,106,107,108,109,110} We have based our classification (see Fig. 12–66) on the following facts:

• The heart was divided into two regions (see Fig. 12–64): the anteroseptal, perfused by the LAD; and the inferolateral, perfused by the RCA or the LCX. Evidently, these two regions are not always equal, according to the anatomic variations of the arteries perfusing them (eg, length, dominance).

• The inferobasal segment of the left ventricle (former posterior wall) that corresponds to segment 4 of the statement by Cerqueira et al⁸⁰ (see Fig. 12–64) depolarizes after 40 to 50 milliseconds. Therefore, this cannot originate an R wave in V₁ that would be equivalent to the Q wave in the back of the heart.

• We have demonstrated that the inferobasal segment rarely bends upward, and therefore, a true posterior wall hardly ever exists. Thus, a vector that points to V₁ cannot usually be originated, even in the case that this vector may exist.

• The oblique position of the heart into the thorax explains that the infarction vector of the inferobasal segment faces V₃ and that the infarction vector of the lateral wall faces V₁ (see Fig. 12–65).

• Therefore, it is clear that the posterior wall usually does not exist because the diaphragmatic wall usually does not bend upward and, even when it does, it cannot originate a prominent R wave in V₁ when necrotic, because activation arrives late. Furthermore, even if necrosis vector exists, it would be directed to V₃-V₄ and not V₁-V₂⁸² (see Figs. 12–65 and 12–66; Fig. 12–67).

• The ECG can distinguish well MIs of anteroseptal and inferolateral zones (see Fig. 12–66). The specificity of these criteria is very high, especially in MIs of the inferolateral zone, although the sensitivity is lower.

The most important conclusions of these correlations are as follows:

• R/S of at least 1 in V₁ is never caused by MI of the inferobasal (old posterior) wall, but instead, is always caused by MI of the lateral wall (see Fig. 12–67A–C). Other new criteria for lateral MI (R/S > 0.5) in V₁ that are even more restrictive have been published.⁸¹

• MIs of exclusive inferior wall always present small r in V₁ (see Fig. 12–67D).

• Q wave in VL (and sometimes in I and small q in V₂-V₃) is not the result of high lateral MI but of midanterior MI caused by occlusion of the first diagonal (Fig. 12–68).

• In the presence of MI of anteroseptal zone with Q wave beyond V₂, the presence of Q in I and/or VL suggests that the infarction is more extensive than the apical zone (Fig. 12–69).

• It is difficult to differentiate the types of MI of anteroseptal zone, especially when the location depends on the presence or not of a Q wave in V₃ to V₅. This is because the location of electrodes in the appropriate place in these leads may change if the person who performs the ECG does not always follow the same methodology for location of electrodes. Because of this, the SE and SP of Q wave to differentiate distinct MI of anteroseptal zone may change if a nonstrict methodology of location of electrodes is used.

D. Quantification of necrosis.^111,112,113 Selvester et al¹¹¹ described a 31-point scoring system based on 50 criteria (eg, presence of Q wave in different leads or R wave in V₁ to V₂ as mirror pattern). This scoring quantifies the amount of infarcted tissue (3% of the left ventricular mass for each point). At the individual level, the standard error of myocardial damage quantification using this score is large, such that its clinical usefulness is limited. The most important cause of errors with this scoring system comes from its inability to quantify basal infarcted areas, mainly the septal and lateral areas. Now, CE-CMR has demonstrated great accuracy in estimating the size of the infarcted mass, which makes this technique the “gold standard” for damage quantification. However, it is still necessary to develop standards that could be consistently applied for CE-CMR measurements.¹¹³

Myocardial Infarction without Q Wave or Equivalent

MI without Q wave or equivalent includes all of the MIs listed in Table 12–13.

The presence of an R wave in V1 in post-MI patients may be equivalent to the Q wave as expression of a mirror image of the Q wave of a lateral MI.

The most typical MIs of this type are the non–Q-wave MIs and the MIs with fractioned QRS complex (Fig. 12–70). In these cases, the diagnosis should be based on the presence of clinical symptoms of angina accompanied by enzymatic changes and repolarization alterations (ST segment and/or negative T wave).

According to the new classification of MI developed by the European Society of Cardiology and the American College of Cardiology,¹⁰⁶ the presence of increased enzymes allows one to differentiate a non–Q-wave MI from unstable angina. The masked Q-wave MI in case of wide QRS will be discussed in the following section.

Other characteristics of non–Q-wave MIs and other MIs without Q waves are listed in Table 12–14.

Myocardial Infarction in the Presence of Confounding Factors

We will briefly discuss the diagnosis of MI in chronic phase in the presence of ventricular blocks, pacemakers, and WPW preexcitation (see also Acute Coronary Syndrome with Confounding Factors).

Right Bundle Branch Block

In the chronic phase, because cardiac activation begins normally, the Q-wave MI may be generated as it is in cases with a normal QRS complex.

Left Bundle Branch Block

Because of changes of ventricular activation even if important zones of the left ventricle are necrotic, the overall direction of the depolarization vector continues to point from the right to the left, and the Q wave of necrosis is usually not recorded. However, if the necrosis is extensive, an evident q wave in I, VL, V₅, or V₆ and/or an r in V₁ may be recorded (Fig. 12–71).

Hemiblocks

Generally, in the presence of SAH, a Q wave of necrosis may be diagnosed without problems. The loop-hemifield correlation (Fig. 12–72) explains that in the presence of an SAH in lead II, there is no terminal r wave, whereas it is present in lead VR, which is the opposite pattern as occurs in isolated inferior MI. This is an example of how the ECG may provide the same information as the VCG loop. Some authors¹¹⁴ have considered wrongly that the association of MI and SAH can only be diagnosed with the VCG.

In the presence of IPH, the change of activation may mask or decrease the Q wave in an inferior MI or may mask the Q wave in a small lateral MI.

Preexcitation and Pacemakers

In both cases, it may be difficult to diagnose the presence of a Q-wave MI. In the case of pacemakers, the presence of a spike-qR pattern in V₅ to V₆ confers high specificity but low sensitivity as a sign of necrosis.

Electrocardiographic Changes in Ischemia Caused by Increased Demand

ECG changes resulting from ischemia caused by increased demand are explained by a coronary stenosis, resulting from a fixed-stable plaque, that produces at-rest impairment of subendocardium perfusion that is not usually so prominent as to change the basal ECG.⁶⁰ Because the vasodilatory capability of the subendocardium is low, it is more vulnerable to myocardial ischemia. Therefore, even though the ECG is frequently normal in the basal state, it presents ST-segment depression during exercise, which is the typical ECG change of subendocardial ischemia without total occlusion. These changes appear especially in leads with a predominant R wave.

Rarely, an ST-segment elevation appears during exercise and is generally caused by coronary spasm.

The ECG changes during exercise angina may be detected by Holter monitoring (Fig. 12–73). They may also be reproduced during exercise testing (see Fig. 12–51B and Chap. 13).

Electrocardiographic Changes in Ischemia Not Related to Atherothrombosis

This section will discuss the ECG characteristics found in some of the clinical setting of myocardial ischemia not resulting from atherothrombosis. For more information, consult Bayés de Luna and Fiol-Sala⁶⁰ and Chap. 35.

In many cases, such as in coronary dissection,¹¹⁵ congenital abnormalities of coronary arteries,¹¹⁶ or hypercoagulation states,¹¹⁷ the clinical and ECG changes reflect a clear STE-ACS evolving to a Q-wave MI.

TTS is included in the classification of transitory left ventricular ballooning (Fig. 12–74). This syndrome is difficult to differentiate from STEMI caused by LAD middle occlusion, because it presents in typical cases the same ECG changes found in this STEMI (ST-segment elevation in V₂-V₃ to V₄-V₅, but without ST-segment elevation in VR).^78,79 However, ECG changes in TTS are often more widespread. The syndrome usually occurs in postmenopausal women, and presentation is typically precipitated by emotional or physical stress.^118,119,120 In TTS, the ST-segment elevation evolves to very negative T waves (reperfusion pattern), which often is accompanied by a usually transient Q wave (see Fig. 12–74).

Coronary spasm presents the typical pattern of a very transient ST-segment elevation that is often (> 50% in our experience) preceded by a clear tall T wave that vanishes in a few minutes (see Fig. 12–46). Sometimes, the ECG only presents a small ST-segment depression or minimal changes in the T wave.^75,76

Lastly, some conditions, such as the X syndrome,¹²¹ present an ST-segment depression as the most frequent abnormality, which is often only seen during exercise.

Electrocardiography in Other Heart Diseases

Heart Failure

No clear ECG marker is suggestive of heart failure, but almost any ECG diagnostic entity may be found in its presence. Because heart failure with preserved ejection fraction is often found in the elderly with hypertension, signs of LVH are the most common ECG changes observed in this type of heart failure. Now, we summarize the most interesting points of the ECG occurring in patients with heart failure with reduced ejection fraction.

In heart failure class III or IV of New York Heart Association, a normal ECG is very rare. Actually, a normal ECG would suggest that the heart failure is caused by high cardiac output states, such as beriberi.

The ECG is abnormal in patients with class IV heart failure, showing significant QRS and ST-T alterations, as an expression of chamber enlargement and/or intraventricular blocks or associated IHD ECG changes. Furthermore, sinus tachycardia, interatrial blocks AF, and ventricular arrhythmias are frequent findings (Fig. 12–75).

Characteristically, patients with congestive heart failure often present with attenuated QRS voltage that may be related to pericardial effusion, or to peripheral edema.¹²² However, attenuation of the amplitude of the P wave is only seen in patients with peripheral edema.^122,123 This observation is useful in the differential diagnosis of these two clinical conditions.

Patients with congestive heart failure are at high risk for malignant arrhythmias and death. We have developed a scoring method for the risk of death.⁴⁵ From the ECG point of view, the concomitant presence of LBBB and AF is a marker of a poor outcome.^44,45

In cases of advanced heart failure, a heart transplantation may be the solution. The ECG of the transplanted heart presents the following characteristics:

1. The P wave of both the receptor and the donor (the latter linked to the QRS) can be seen on the surface ECG (Fig. 12–76). Occasionally, it is necessary to use a right atrial lead to make the P wave of the receptor visible.

2. Generally, the donor P wave is of a higher rate than the receptor P wave because there is no vagal control on the donor heart.

3. Other frequent ECG findings are repolarization alterations, RBBB morphology, interatrial blocks, atrial arrhythmias (flutter and fibrillation), and sinus bradycardia, sometimes severe, with evidence of sick sinus syndrome.¹²⁴

Valvular Heart Disease

Usually, the presence of signs of LVE or RVE indicates that the disease is severe. This does not always correlate with the presence of important clinical symptoms.

Mitral stenosis produces a biphasic positive/negative P wave in V₁ and a wide and bimodal P wave in other leads, representing LAE (mitral P wave) (see Fig. 12–22). These patients usually develop AF during the natural history of their disease. Its occurrence is more probable when the P wave is more altered. The presence of a biphasic positive/negative P wave of at least 0.12 second in II, III, and VR is a specific marker of paroxysmal supraventricular arrhythmias at 1 year of follow-up (see Fig. 12–24). ECG signs of LVE are usually seen when there is associated a significant mitral regurgitation or aortic valvular disease.

Patients presenting with tricuspid regurgitation as a consequence of mitral valve disease usually have ECG findings of enlargement of the right chambers. Frequently, the presence of an evident R wave in V₁ but with rS pattern in V₆ is already suggestive of associated RVE (see Fig. 12–26).

Repolarization alterations in leads II, III, and VF and the left precordial leads and ventricular arrhythmias are common in patients with a mitral valve prolapse.

Isolated aortic valve disease, especially in the early evolutionary stages, is usually not accompanied by AF. Its presence should raise the suspicion of associated mitral valve disease.

In the initial stages of LVE caused by aortic valve disease, the QRS complex generally presents a qR with a positive T-wave pattern, which is more evident in aortic regurgitation than in aortic stenosis. Over the course of the disease, both aortic stenosis and aortic regurgitation can show a morphology with ST-segment depression called strain pattern. The qR morphology remains more frequently in aortic regurgitation, although the presence of a q wave in V₅ to V₆ depends more on the degree of septal fibrosis (the greater the fibrosis, the smaller the q wave) than on the lesion type (stenosis or regurgitation) (see Figs. 12–29 and 12–30).

Cardiomyopathies

In hypertrophic cardiomyopathy, the ECG may present different alterations. In only 5% of the cases, the ECG is normal. The most typical findings, which are characteristic but infrequent, are (1) a pseudonecrosis q wave; (2) the pattern of apical hypertrophic cardiomyopathy in which a giant negative T wave is usually seen (see Fig. 12–29D and E; this finding does not indicate a poor prognosis); and (3) often signs of a classical LVE type.

The presence of AF in hypertrophic cardiomyopathy is generally poorly tolerated and, together with the presence of ventricular arrhythmia in Holter recordings, septal hypertrophy greater than 20 mm, mild or no increase in blood pressure during the exercise test, and a family history of syncope or sudden death, are signs of bad outcome.^125,126

In dilated cardiomyopathy, the ECG usually reflects the signs of atrial and/or ventricular enlargement, but the most characteristic finding is the presence of a low-voltage, notched, wide QRS complex, particularly in the frontal plane (atypical morphologies of LBBB), sometimes with a q wave of pseudonecrosis (see Fig. 12–75).

In advanced cases of restrictive cardiomyopathy, striking patterns (eg, q wave of pseudonecrosis, P wave with large negative component in V₁, prominent R in V₁) may be seen.

In arrhythmogenic right ventricular cardiomyopathy, the most frequent ECG findings are as follows.¹ Atypical RBBB morphology presents in general, a single R of low voltage with notches and/or slurrings in the plateau, and a wide QRS complex (≥ 0.12 second) may be seen in 10% to 20% of cases.² Often, there are differences in the QRS complex duration (> 50 milliseconds) between V₁ and V₆, probably caused by an associated right ventricle parietal conduction block.³ A symmetrical negative T wave is also frequently seen (with or without an RBBB pattern) from V₁ to V₃/V₄.⁴ Notches resulting from delayed depolarization caused by parietal block may rarely be observed by the surface ECG at the end of the QRS complex (epsilon wave). These waves may be clearly seen on signal-averaged ECG recordings. Premature ventricular impulses generated in the right ventricle are often seen (Fig. 12–77).⁶ The ECG is normal only in 20% of cases.

In myocarditis, the ECG is variable and nonspecific. Sometimes important repolarization alterations, especially deep negative T waves with slight ST-segment changes or even Q waves, may be seen, especially in precordial leads. It is necessary to make a differential diagnosis with IHD, including the TTS.

Pericardial Disease

Although it has been described that idiopathic pericarditis undergoes four typical phases in the evolutionary ECG,¹²⁷ often not all of them appear (Fig. 12–78). Remarkably, the ST-segment elevation occurring in the acute phase of pericarditis with chest pain must be differentiated from an ACS. Frequently, the elevation or depression of the PR interval in lead VR, may help in the differential diagnosis (see Fig. 12–25).

Likewise, the repolarization abnormalities (flattened or negative T wave) of chronic pericarditis must be differentiated from similar T waves found in other processes (see Fig. 12–78C). Characteristically, the flat/inverted T-wave changes in pericarditis are usually seen in more leads, but the deepness is lower than in the negative T wave after Q-wave MI (see Figs. 12–46 and 12–78). However, the difference is not so visible with the negative T wave seen in NSTE-ACS (see Fig. 12–62C).

Characteristically, the more frequent ECG change found in constrictive pericarditis is the presence of an extensive flat or mildly negative T wave, without a mirror image in the frontal plane. However, the same morphology may also be seen after cardiac surgery or in other processes (eg, drugs, alcohol intake).

In some occasions, a striking QRS alternans may be seen in cases of cardiac tamponade (see Fig. 12–88A).

Cor Pulmonale

The ECG can be normal in an acute cor pulmonale that is caused by pulmonary embolism, albeit this is not usually the case when the pulmonary embolism is significant. The most valuable diagnostic findings, which are poorly sensitive but highly specific, are as follows: (1) the McGinn-White pattern (S_I, Q_III with negative T_III pattern; see Fig. 12–28); (2) RBBB, which may be seen in cases of massive pulmonary embolism; (3) a right-directed shift of the ÂQRS; (4) repolarization alterations (negative T wave) in right precordial leads; and (5) sinus tachycardia, which is common in pulmonary embolism, except in elderly patients or in those with sinus node dysfunction.

Acute right heart decompensation, caused by respiratory infection or other processes, can lead to morphologies similar to a pulmonary embolism. Among other alterations, the characteristics are (1) a right-sided shift of the ÁQRS and (2) a transient negative T wave in the right precordial leads.

In chronic cor pulmonale, the ECG may or may not exhibit signs of RVE, although an RVE may exist without an evident ECG alteration. The ECG findings that suggest RVE are (1) ÂQRS in the frontal plane with S_I, R_II, R_III or S_I, S_II, S_III patterns or (2) rS morphology in V₁ with an evident S in V₆ (see Fig. 12–26E and F).

In pulmonary emphysema, the ECG may show (1) ÂP between +60° and +90°; (2) a low-voltage P wave in leads I and V₁ and a relatively high-voltage P wave in leads II and III; (3) a QS in lead V₁ and sometimes in V₂ with S in V₆; (4) a low-voltage QRS in the frontal plane; and (5) ÂQRS somewhat to the right or with an S_I, S_II, S_III pattern.

Arterial Hypertension

The ECG changes found in arterial hypertension are caused by LVE. In general, there is a strong correlation between the severity of arterial hypertension and the ECG findings. With treatment, these changes may decrease or revert to normal.

Occasionally, a primary factor (drugs or ischemia) modifies the repolarization that already presented as a secondary repolarization abnormality (strain) due to LVH. Then, a mixed repolarization pattern is seen.

The criteria for the diagnosis of LVE in arterial hypertension that display a higher sensitivity (80%) include RV₅ to RV₆ greater than 26 mm and total QRS voltage in 12 leads greater than 120 mm.³⁶ Their specificity, however, especially for the first criterion (20%), is low. The sensitivity of the ECG criteria increases in proportion with the prevalence of LVH in the study population (Bayés theorem).¹²⁸ The utility of the ECG to diagnose LVE is high in severely hypertensive patients and low in the general population.

Note the following regarding the importance of ECG as a marker of prognosis in patients with hypertension. In patients with hypertension and ECG with LVH, ECG at entry has a worse prognosis if new ST abnormalities develop.¹²⁹ The presence of nonspecific left ventricle repolarization changes in hypertensive patients represents a risk factor for IHD.¹³⁰ The regression of LVH detected by Cornell criteria during antihypertensive treatment is associated with fewer hospitalizations for heart failure.¹³¹ The presence of ventricular arrhythmias is a marker of poor prognosis in hypertensive patients. It has been demonstrated¹³² that both LVE and asymptomatic ST-segment depression (silent ischemia) are independent predictors of ventricular arrhythmias in asymptomatic hypertensive patients.

Congenital Heart Disease

A comprehensive review of ECG alterations in congenital heart disease is beyond the scope of this book. Furthermore, new imaging techniques are more useful for the diagnosis of congenital heart disease than ECG. However, there are some ECG signs that are highly suggestive of specific congenital heart diseases that are presented here.

Atrial Septal Defect

(1) Presence of rsR′ pattern in lead V₁ with QSR < 0.12 second; (2) lack of sinus arrhythmia if the atrial septal defect (ASD) is large; (3) in ostium primum–type ASD, ÂQRS is shifted sharply to the left; and (4) AF in adults.

Ventricular Septal Defect

(1) Normal or almost normal ECG in small ventricular septal defects (VSDs); (2) signs of biventricular hypertrophy in large VSDs with hyperkinetic pulmonary hypertension, sometimes with high voltages in the intermediate precordial leads (Katz and Watchel morphology); and (3) tall R in lead V₁ when significant pulmonary hypertension is present (Eisenmenger syndrome).

Tricuspid Atresia

(1) ÂQRS sharply to the left; and (2) P wave of RAE (Fig. 12–79A).

Ebstein Disease

(1) High-voltage P wave, sometimes with large negative mode in V₁; (2) atypical right ventricular block morphology often with many deflections (rsr′s′) and sometimes with slight ST-segment elevation in V₁ to V₂; (3) relatively often WPW-type preexcitation; (4) and occasionally, long PR interval (Fig. 12–79B).

Stenotic Lesions

Stenotic lesions of the right heart (pulmonary stenosis and tetralogy of Fallot) and the left heart (aortic stenosis and coarctation of aorta) produce enlargement of the respective chambers when they are severe enough. Tight pulmonary stenosis presents R morphology in V₁ to V₂/V₃ (see Fig. 12–27A), and the tetralogy of Fallot pattern shows an R wave in V₁ and rS morphology in V₂. In newborns, RVE is observed in severe pulmonary stenosis and may be accompanied by a positive T wave. After surgery for the tetralogy of Fallot and other congenital heart diseases, an advanced RBBB pattern frequently appears as a result of a distal lesion of the specialized conduction system caused by the surgery. In this case, the distance from the onset of ventricular depolarization to the right ventricular apex (V-apex) is not prolonged, in contrast to what occurs in cases of proximal block. Congenital aortic stenosis, even if it is severe, does not usually result in a typical LVE pattern within the first years of follow-up. However, over time, it may advance to the so-called strain pattern (see Fig. 12–30A).

Mirror-Image Dextrocardia

The ECG is typical (Fig. 12–80). The same morphology may be recorded with the inverted limb leads and precordial leads located on the right side. This explains the negative P wave in lead I and the decreasing wave voltages of the precordial leads.

Electrocardiogram of the Newborn: Suspicion of Congenital Heart Disease

The most interesting features of the newborn ECG with possible congenital cardiac malformation are as follows:

• Many neonates with congenital cardiac malformation exhibit a normal ECG on the first day of life. Thus, it is necessary to record the ECG sequentially.

• In a newborn with a left-sided QRS (beyond +30°), AV cushion defect, tricuspid atresia, and a single ventricle should be considered as probable causes.

• AV block may appear in VSD and corrected transposition. However, advanced congenital block may be an isolated finding without any associated heart disease.

• The presence of a delta wave requires Ebstein disease to be ruled out.

• A positive T wave in leads V₁ and V₂ from day 2 of life may suggest RVE.

• The presence of an R wave (not qR), possibly tall (> 15 mm), may be seen in newborns without congenital heart disease.

Channelopathies

The concept of channelopathies includes heart diseases without apparent structural involvement. They are caused by mutations in certain genes that generate defects in the ionic transport through the channels. The first manifestation of the disease may be syncope or even sudden death resulting from ventricular tachycardia/ventricular fibrillation.

Characteristically, in the best-known channelopathies (long and short QT syndrome and Brugada syndrome), there is an important phenotype ECG expression. Figure 12–81 shows examples of the typical ECG pattern in cases of long QT syndrome,¹⁶ short QT syndrome,¹⁸ and Brugada syndrome type 1 and 2^{133,134,135,136}. In Fig. 12–81C, we may see how the descendent arm of r’ allow to perform the differential diagnosis between Brugada pattern type 2 and many other cases with r’ such as athletes. Figure 12–82 shows an algorithm (Baranchuk algorithm)¹³⁶ that allows the differential diagnosis of all cases with r’ in V1.

Electrocardiography in Specific Situations

Other Diseases

In this section, the most striking ECG findings that may be found in the following diseases are briefly described.

Cerebral Disease

Repolarization alterations appear with certain frequency, particularly in subarachnoid hemorrhage, and may include very negative T waves or wide-based positive T waves (Fig. 12–83) or ST-segment changes (depression or elevation).

Endocrine Disease

ECG signs typical of myxedema are a low-voltage QRS complex, bradycardia, and flattening or inversion of the T wave. Some common features found in hyperthyroidism are sinus tachycardia and supraventricular arrhythmias, including AF.

Diabetes

Diabetes frequently causes repolarization alterations. It has been demonstrated that these are more common in patients with coronary heart disease with type 2 diabetes than in patients with coronary heart disease without diabetes.

Miscellaneous

There are many other pathologic processes in which ECG abnormalities range from minor repolarization alterations (eg, hepatic cirrhosis, anemia) to bundle branch block and/or necrosis patterns of some systemic diseases (eg, amyloidosis, hemochromatosis) and pheochromocytoma. Characteristically, the ECG pattern of restrictive cardiomyopathy is infiltrative, including amyloid, sarcoid, and Gaucher disease, or noninfiltrative, including hemochromatosis and Fabry disease. It presents in the advanced phase, with some striking ECG patterns that include (1) pseudonecrosis “q” wave; (2) low or normal QRS voltage despite LVH, (3) P wave with clear negative component in V1, of LAE, and (4) sometimes clear repolarization abnormalities.

In some neuromuscular diseases, such as Duchenne disease and Friedrich ataxia, patients may present ventricular enlargement as a negative T wave.

Electrolyte Imbalance

The most remarkable ECG alterations in electrolyte imbalance are caused by changes in serum potassium levels. In Fig. 12–84, typical examples of hyperkalemia and hypokalemia are seen.¹³⁷

In hypocalcemia, the QT interval may be prolonged at the expense of the ST segment, and in hypercalcemia, the QT interval may be short.