Utility and Limitations of the Surface ECG

Chapter 3
Utility and Limitations of the Surface ECG: Present and Future


More than 100 years have passed since the invention of the surface ECG and we now live in an age when new inventions and progress in medicine are seen every day, so it is astonishing that the surface ECG not only still exists but also remains as a “gold standard” technique for the study of electrical disorders of the heart. The following is an overview of the most relevant advantages and disadvantages, as well as the utility and limitations, of the surface ECG.


ECG as the best diagnostic tool (Fiol‐Sala et al. 2020, Bayés de Luna and Baranchuk 2017)

The ECG is the best diagnostic tool for the following:

  • The diagnosis and evaluation of tachy and brady arrhythmias, conduction disorders, at sinoatrial, at atrial, AV junction and ventricular level, and pre‐excitation syndromes.
  • Differential diagnosis in cases of broad QRS tachycardia with one strip of the surface ECG. Correct differential diagnosis (supraventricular vs. ventricular tachycardia) can be achieved in more than 90% of cases.
  • Some ECG patterns may be markers of future tachyarrhythmias, or even of global, CV, or of sudden death such as Wolff–Parkinson–White (WPW) pattern, channelopathies and some P wave indices especially duration P wave, the advanced interatrial block (A‐IAB), the P axis, and P voltage. And, also recently a score of P wave may be more useful that P wave indices alone (Alexander et al. 2019), and the association of P wave indices with CHA2DS2‐Vasc (Maheshwari et al. 2019) (see Chapter 9).

    The ECG also may perform, sometimes with the help of clinical history, the differential diagnosis of some types of ECG pattern such as Brugada phenocopies (Baranchuk 2018).

  • Determining whether or not it is necessary to implant a pacemaker or cardiac resynchronization therapy (CRT) (Chapter 17).
  • The diagnosis of acute ischemic events. In this case, it is essential to have the clinical information, but a good knowledge of ECG patterns may help to perform the diagnosis of acute coronary syndromes (ACS), and to classify them into two types: those with ST segment elevation and those with non‐ST segment elevation (Bayés de Luna et al. 2007) (see Chapters 13 and 20).
  • As said, the detection of a phenotype expression of channelopathies (long QT, short QT, Brugada syndrome). In other inherited heart diseases, such as hypertrophic cardiomyopathy or arrhythmogenic right ventricular dysplasia, the ECG may also help in the diagnosis, but there is no clear genotype–phenotype correlation (Bayés de Luna and Baranchuk 2017) (Chapter 21).
  • The surveillance of different types of pacemakers and implanted defibrillators (ICD).

When the ECG is also important

The ECG is also important for the following:

  • To presume a diagnosis of chronic ischemic heart disease (IHD), especially of chronic Q wave myocardial infarction (MI).
  • The evaluation and follow‐up of patients following MI or cardiac surgery.
  • The diagnosis and study of the evolution of other heart diseases, such as valvular heart diseases or pericarditis (Spodick 2003), and in special circumstances, such as electrolyte imbalance (Surawicz 1967) or drug administration.
  • The evaluation of athletes and check‐ups in general, and as a pre‐operative assessment for cardiac and non‐cardiac surgery.

The correlation of ECG patterns with the clinical setting

The ECG is currently of great utility, not only from a diagnostic point of view, but also from a prognostic one, in addition to its use in the management of heart disease. This is clearly illustrated in the acute phase of IHD. ACS are classified into two groups to decide treatment: with ST elevation (STEMI) and without ST elevation (NSTEMI) (see above and Chapter 20).

This classification, although challenged by some authors (Phibbs 2010), is very helpful in stratifying the management of ACS and is used worldwide and accepted in all guidelines (ACC/AHA/HRS 2007, 2009).

However, it is important to note that even experts in ECG diagnosis experience problems in correctly interpreting ST changes without good clinical information (see Bayés de Luna et al. 2007; Jayroe et al. 2009; Nikus et al. 2010; Fiol‐Sala et al. 2020, and Chapter 19) (see Section “Limitations”).

Correlation with coronarography and imaging techniques

The correlation between ECG patterns and coronarography in cases of ACS is very valuable to better locate the occlusion site and area at risk (Sclarowsky 1999; Wellens et al. 2003; Fiol et al. 2004, 2009; Fiol‐Sala et al. 2020).

Contrast‐enhanced cardiac magnetic resonance (CE‐CMR) is very useful in determining the presence and size of MI (Moon et al. 2004) and in demonstrating that the first area of myocardium to suffer the effects of ischemia is the subendocardium (Mahrholdt et al. 2005). In patients with chronic MI in particular, the correlation of ECG patterns and CE‐CMR has been very useful in demonstrating and properly locating the presence of necrotic areas according to the new terminology used for heart walls (Cerqueira et al. 2002; Bayés de Luna et al. 2006a; Fiol‐Sala et al. 2020). For example, it has been demonstrated that in post‐MI patients, the R wave in V1 is caused by lateral, not posterior, MI and that “q” in aVL is caused by first diagonal occlusion (midanterior MI) rather than circumflex occlusion (high lateral MI) (Bayés de Luna et al. 2006b, 2008; Rovai et al. 2007; Van der Weg et al. 2009) (see Figure 13.68).

The ECG as a research tool

The ECG is used in epidemiological studies and in the safety evaluation of new drugswith recognized or potential cardiac effects (QT interval, arrhythmias, etc.).

Other benefits

ECG recording is an inexpensive, easy, and quick technique to register the electrical activity of the heart.


  • Despite the usefulness of the ECG in acute MI, many patients (≈30–50%) with old MI or chronic IHD present with a normal or non‐definitively abnormal ECG at rest and often even during exercise. If the physician is not well trained, small changes may not be correctly perceived.
  • Some severe heart diseases such as pulmonary embolism, cardiac tamponade, acute aneurysmal dissection, or even as acute ischemic attack may present not ECG changes at all, or only non‐specific, or subtle ECG changes that may only be interpreted correctly by those with very good ECG training, who are able to make a good correlation with the clinical setting (see Chapters 20 and 21).
  • Following the emergence of new imaging techniques (echocardiography and especially cardiovascular magnetic resonance), the diagnosis of chamber enlargement and congenital heart disease is more precise than with electrocardiography alone. However, the usefulness of ECG recording is still important in some aspects, not only from a diagnostic point of view, but also from a prognostic one (see Chapter 10). Furthermore, although no full correlation between ECG changes and these imaging techniques has been performed, the results obtained to date have been promising (see Chapters 9 and 10).
  • The ECG can be a misleading diagnostic tool if the existence or absence of underlying heart disease is identified based only on the presence of a normal or abnormal ECG pattern. Unfortunately, the ECG does not have a talisman’s power. When faced with a patient presenting with precordial pain of unknown origin, it is a great mistake to proclaim “The ECG will settle this.” While it is true that the ECG may give crucial information that may be decisive for the diagnosis, on some occasions even in the very acute phase of MI, the ECG may remain unmodified or presenting changes (f.i. positive and symmetrical T wave in right precordial leads) that may be considered normal. Therefore, it would be very dangerous to fail to integrate the ECG information with the clinical setting. On the other hand, healthy individuals sometimes present with some alterations of the ECG, which may lead to a wrong diagnosis if used alone (f.i. do not perform a new recording with deep inspiration in the presence of Q in lead III) (see Chapter 25). Therefore, the ECG must always be interpreted in the light of the clinical context. At present, a manual interpretation of the ECG is still advised. Automatic interpretation, which may produce a more accurate and useful computer‐generated report, will likely be used more often in the future, but currently computer interpretation needs physician overreading (see later) (Kligfield et al. 2007).
  • Clearly, a normal ECG should not be considered a “warranty” and in fact does not exclude the possibility of cardiac death resulting from an electrical disturbance (i.e. ventricular fibrillation or bradyarrhythmias), even on the same day that the ECG is recorded. If there is some clinical suspicion of IHD, serial ECGs should be recorded and an attempt should be made to ascertain the cause of the pain, regardless of the presence of a normal or nearly normal ECG at entrance.
  • Normal variants of the ECG may be related to body constitution, thoracic malformations, race, or gender. The ECG may also show transient changes caused by metabolic or other disorders (hyperventilation, hypothermia, glucose or alcohol intake, electrolyte disorders, drug effects, etc.).
  • There are limitations inherent in the technology used. Until recently, the recording process was based on analog techniques. These techniques present limitations because the quality of the recording decreases with time, storage become a problem, and it is impossible to transmit the recording immediately. It also has to be considered the possible presence of technical errors in the recording process (Chapter 6). Therefore, digitalized ECG is very much recommended.
  • Probably the most important limitation of the ECG is not the technique itself but rather the physician’s lack of ECG expertise. Currently, the majority of hospitals do not provide a period devoted to training by experts in ECG diagnosis. It is important that the ECG is studied with an updated mentor in a systematic and sequential way, which is what we attempt to do in this book. It is also extremely important as we have already said to interpret the ECG bearing in mind the clinical context. Once the training period is finished, periodic updated programmes are necessary. Universities, Hospitals, and Scientific Societies must organize intramural and extramural meetings, including internet courses, designed for trainees and also their mentors. This will allow the weaknesses to be identified and the knowledge to be reinforced. An adequate traineeship approach would be of great advantage for the healthcare system because it would allow for: (i) a better and quicker diagnosis, prognosis and management in many situations (acute ischemic events, patients at risk of sudden death, the need for pacemaker implant, etc.) and (ii) the avoidance of more sophisticated techniques, resulting in saved money, time, and discomfort.
  • Some limitations of the surface ECG may be overcome by other electrocardiological techniques. It is clear that the study of the electrical activity of the heart using exercise and long‐term recording (monitoring and Holter technology), intracardiac electrophysiological studies, filtering systems, body mapping, etc. (see Chapter 25), may help in some specific ways. This adds to the usefulness of the surface ECG and reduces its limitations.

Finally, the correlation with imaging techniques, especially echocardiography, cardiac multislice scanner, and cardiac magnetic resonance, complements the capacity of ECG for the diagnosis, prognosis, and management of heart diseases. We will deal with each of these aspects later along the book.

The future of electrocardiography

Undoubtedly, what is most important for the future is to try to overcome the limitations that still exist in electrocardiography. Fortunately, we are already obtaining great results in this respect. The ECG is an evolving, living science and its future is very promising, particularly if the advancements described below are achieved (Guidelines AHA/ACC/HRS 2007–2009) (see Chapter 26). Due to that, the standards for ECG interpretation require periodic review and revision (Gettes 2009).

The ECG in the digital era (see Chapter 26)

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. Nearly all current ECG machines convert the analog ECG signal to digital form. The initial sampling rate during analog‐to‐digital conversion is higher than the sampling rate that is used for further processing. This oversampling was originally introduced to detect and represent pacemaker stimulus outputs which are generally <0.5 ms in duration.

Ideally, ECG machines should be small and compact, and must work in an integrated fashion in various health institutional settings, independently of the type of technology used. These systems allow us to work online and may be incorporated into telemedical services. Compression of ECGs is recommended for transmission and data storage. In an ideal world, the diagnosis could be performed by an expert in real time through the Internet and telephone, and this could then be applied to multiple scenarios (ambulance, isolated areas, ships, etc.) with poor access to assistance because of their geographic location or associated economic costs (see Figure 6.19).

Improvements in automated ECG interpretation

In spite of all, new improvements in automatic interpretation currently, all automatic recordings need to be reviewed by a physician (American College of Cardiology/American Heart Association 2007). By entering more data about the patient (sex, age, body habitus, and clinical history) and facilitating a link to ECG‐pedia—a corpus containing information taken from many ECG books and reports that can assist in the correct diagnosis—it may be possible, although it has to be proved along this century, that the need that all automatic interpretation to be reviewed will not be necessary. However, according to Surawicz (2010), a comparison between a diagnosis performed by a computer linked to an updated ECG‐pedia and an ECG expert would be as interesting as a chess match between a computer and a chess master. For the moment, the ECG expert would win. Probably in the future, the match will be more equal (see Chapter 26).

New advances in correlation with imaging techniques

As previously explained, CE‐CMR is useful in detecting the presence of necrosis and the size of MI, as well as identifying, in correlation with the ECG, the location of the necrosis. However, while the correlation between scoring ECG system and CE‐CMR for estimating infarction size and measuring the left ventricular ejection fraction (LVEF) is reasonable, it is not consistent enough because the results are always larger with CE‐CMR (Weir et al. 2010). Therefore, to obtain an accurate and reproducible estimation of infarct size and LVEF by surface ECG in survivors of acute MI using a 12‐lead ECG would be a great success and something that would allow a significantly better correlation between ECG scoring and CE‐CMR (Weir et al. 2010).

In addition, a better definition of the diagnosis of chamber enlargement performed by ECG criteria could be made using CMR correlation. Echocardiographic comparison underestimates the volume size of atrial chambers (Whitlock et al. 2010). Diagnosis would probably be more accurate if a correlation between ECG and CMR detecting chamber enlargement were performed (see Chapter 9).

Recently has been demonstrated (Lacalzada et al. 2018) that the presence of A‐IAB in surface ECG is a surrogate of reduced LA strain measured by echo speckle‐tracking, and also A‐IAB is a surrogate of atrial fibrosis and LA dyssynchrony detected by CMR (Ciuffo et al. 2020).

The use of the ECG in new treatments of heart disease

The use of ECG in new treatments such as cardiac regeneration or the detection of cardiac rejection show promises for the future.

Rapid recording of the ECG means better treatment

Figure 20.1 shows the evolution of the ST elevation acute coronary syndrome (STE‐ACS) from the 1970s to the present. Today, we can take quick decisions based on the prompt recording of the ECG and save lives, or at least reduce cardiac muscle damage, while in the past, we are not able frequently to detect the early changes and identify as abnormal or presumably abnormal some characteristics of especially ECG repolarization, such as pecked positive T wave recorded in right precordial leads (see Figure 20.4).

Improving the capacity of the ECG

Further improvements in the capacity for better diagnosis, risk stratification, prognosis, and management of heart diseases or other processes are possible (Poplack Potter 2010). This will be included in the concept of “expanded surface ECG” or “ECG plus” and may be accomplished by the following potential future developments:

  • Using amplification methods in order to record the ECG (×4 amplification). This will permit the shifts of ST to be better recognized; a 0.5 mm change is sufficient for diagnosing ACS (see Figure 13.24). This will also help us to study better the P wave especially the morphology and duration so important to diagnose interatrial blocks, the initial forces of pre‐excitation, intraventricular blocks, early repolarization pattern, or Brugada pattern, among others.
  • Recording the ECG at a higher speed may be useful to detect atrioventricular (AV) dissociation.
  • Additional leads, including Lewis leads (Bakker et al. 2009), esophageal leads, etc. (see Chapters 6 and 25), may be used.
  • Recording parameters such as late potentials and others to study ANS imbalance (HRV, HRT, T wave alternans, etc.).
  • Being able in the future to record the bundle of His deflection externally, which permits a better evaluation of the long PR interval and intraventricular conduction disorders. This may help in the decision‐making process for pacemaker implants. This was reported to be feasible by signal averaging more than 40 years ago, but has not been implemented in commercial devices (Wajszczuk et al. 1978).
  • Developing the capacity to filter some waves of the ECG to better see other waves using wavelet technology. For instance, by filtering the T wave, a better definition and correct identification of P wave activity may be achieved in cases of arrhythmias. While this was once reported using analog technology, it has now been demonstrated using digital technology (Goldwasser et al. 2011). These techniques may currently perform the correct diagnosis of different types of supraventricular tachycardias, especially in distinguishing between atrioventricular nodal reentrant tachycardia (AVNRT) and atrioventricular reentrant tachycardia (AVRT) (see Chapter 25). The next step would be to identify the presence of AV dissociation, which is crucial to the differential diagnosis of broad QRS tachycardias. Filtering the QRS to identify any hidden P‐wave and to better study the characteristics of atrial fibrillation waves is another possibility (Platonov et al. 2012).
  • Redefining old ECG criteria with new methodology and technology. For example, “notches and slurrings” in the QRS to diagnose some types of necrosis and many other processes, which was already published some decades ago by Horan et al. (1971) and now rediscovered by Das et al. (2006).
  • Synthesizing VCG loops from the 12‐lead ECG recording. This may be done with a conversion matrix (Edenbrandt and Pahlm 1988, Kors et al. 1990), or directly from a 12‐lead ECG (Rautaharju et al. 2007). This can be useful not only for teaching, but also for some diagnostic (for instance, presence of minor pre‐excitation and better study of P wave), prognostic, and research purposes. Different VCG studies have demonstrated that a widened spatial QRS–T angle is associated with subsequent CV death (Kardys et al. 2003) and also recently new possibilities of VCG in stratifying the risk of sudden death have been noted (Lazzara 2010). These studies have mainly been performed taking the VCG loops from the orthogonal leads x, y, and z (leads perpendicular to each other similar to leads I, aVF, and V2, see Chapter 25). As these leads are not recorded in clinical practice, the vectorcardiographic loops may be synthesized from the 12‐lead ECG recording with similar results (Kors et al. 1990). Therefore, it may be interesting to incorporate synthesized VCG loops into the “expanded ECG” concept. In fact, although was published that the ECG frontal plane QRS‐T angle should not be considered an adequate diagnostic substitute for the spatial QRS‐T angle (Brown and Schlegel 2011), recent studies as yet not published seem to demonstrate that the calculation of the QRS–T angle in the frontal plane from the surface ECG will probably be equivalent to the spatial QRS–T angle taken from the VCG synthesized from an ECG.
Oct 9, 2021 | Posted by in CARDIOLOGY | Comments Off on Utility and Limitations of the Surface ECG

Full access? Get Clinical Tree

Get Clinical Tree app for offline access