The ongoing debate about whether the Tpeak-Tend interval of the standard ECG is a valid marker of transmural dispersion of ventricular repolarisation (DVR) recalls a similar debate about the so-called “QT dispersion” and its relation to DVR which began in the early 1990s and lasted about 20 years. In my opinion, these electrocardiographic debates do – in many cases – blur the important distinction that should exist between clinical electrocardiography and theory of electrocardiography.
The general goal of theory of electrocardiography is to define the exact (i.e. mathematical) relation between the electrical events in the heart and the surface ECG. Clinical electrocardiography, on the other hand, has a very different goal. Every non-artifactual wave, wavelet, notch, bump, interval, amplitude, etc. (or combinations thereof) observed on the surface ECG can be a valid clinical ECG parameter (irrespective of their electrophysiological mechanism!) if, a) it can be measured reliably (reproducibly) and b) it is statistically linked to important clinical events (e.g. risk of sudden death) or findings from other cardiovascular tests (e.g. chamber dimension, genetic data, etc.). The electrophysiological basis of an ECG component, index, etc. cannot prove or disprove its validity as a clinical ECG parameter.
Clinical ECG interpretation looks to me very much like a semantic exercise – that is, an attempt to find the “meaning” of the ECG waves (which are treated as symbols), rather than an analytic process i.e. electrophysiological analysis – i.e. to characterise the electrical events that take place in the heart assessing the waves, notches, bumps and intervals in the surface ECG (which in this case is analysed as a signal). That is why we often (probably intuitively) say “reading an ECG” instead of “analysing an ECG”.
In clinical practice we often use ECG parameters of proven or supposed clinical value which clearly violate elementary theoretical principles. A good example is the “total 12- lead QRS amplitude” as a marker of ventricular hypertrophy.[1] Another example is the clinically useful Cornell voltage[2] (S in V3 plus R in AVL) as a measure of left ventricular hypertrophy, which adds up the voltages of two leads with different “strength” (something like assuming that 5 inches + 3 millimetres = 8 what?). From the point of view of the theory of electrocardiography we should not be even comparing the ST segment elevation occurring in two ECG leads (any two) without applying some sort of correction, because no two leads have the same strength. For example, ST segment elevation of 1 mm (0.1 mV) in lead I generally reflects a greater electrical potential generated by the heart than the same degree of ST elevation in lead II, because lead II is “stronger”.
In my opinion, the clinical significance of an ECG marker should be proved or disproved on the basis of its reproducibility and its link to important clinical events or other clinical variables (or lack of it). Attempts to bolster its clinical significance by speculating about underlying electrophysiological mechanism (for example in manuscripts reporting purely clinical ECG studies) should be discouraged.
References
[1] Siegel RJ, William C. Robert WC. Electrocardiographic observations in severe aortic valve stenosis: Correlative necropsy study to clinical, hemodynamic and ECG variables demonstrating relation of 12-lead QRS amplitude to peak systolic transaortic pressure gradient. Am Heart J 1982; 103:210.
[2] Casale PN, Devereux RB, Kligfield P, et al: Electrocardiographic detection of left ventricular hypertrophy: development and prospective validation of improved criteria. J Am Coll Cardiol 1985; 6:572.
I read, with interest, Dr. Batchvarov’s Cardionote. I find myself in total agreement both in respect of the specific items referred to (“Tpeak-Tend” and “total 12-lead QRS amplitude”) and also in relation to the broader point concerning the artificial alignment of incompletely developed scientific theories with empirical but established relationships between specific ECG findings and clinical events or situations. His illustrative point about the protracted and sterile debate on the topic of “QT dispersion” is particularly apposite.
It is reassuring, stimulating and informative when, in a given area, the theory of electrocardiography develops sufficiently to provide a scientific explanation for one or more ECG findings of proven clinical significance. That process should follow the natural development and maturation of scientific understanding, rather than being “shoe-horned” into an explanation at the first opportunity and whilst still scientifically incomplete.
Dr. Batchvarov hits another bull’s-eye with his comment about the occasional total abrogation of scientific principles in some recommended procedures, when he refers to the use of “total 12-lead QRS amplitude” as an index of left ventricular hypertrophy. I would cite, as a further example, the use of the “criterion” that an axis shift of “>400” (between that in sinus rhythm and that in an episode of broad QRS tachycardia) favours a diagnosis of ventricular tachycardia1. There is no scientific justification for claiming a resolution greater than ± 150 in measurement of the frontal plane QRS axis.
References
- Clinical Arrhythmology and Electrophysiology. A companion to Braunwald’s Heart disease. 2nd edition 2012, Table 21-2 and page 502.
The cardio note by Dr. Velislav N. Batchvarov attitudinizes the relative importance of the theory of electrocardiography in clinical electrocardiography for ECG interpretation. There is no doubt that the standard 12-lead scalar ECG remains and continues to be an inexpensive, simple and highly reliable diagnostic tool when used together with a correct clinical examination and evaluation for the recognition of acute or chronic diseases to guide initial treatment.
In clinical electrocardiography the initial diagnosis is usually made de visu recognizing patterns each one corresponding to a determined type of myocardial alteration. When needed, this should be corroborated with measurements and reasoning out the curves based on the vector analysis and a thorough knowledge of the sequence of cardiac activation to explain the QRS complex and ST-T waves that represent the potential variations of the surface ECG. The scientific basis of the theory of electrocardiography usually creates difficulties for the physician: the physicomathematical laws drawn from the membrane theories, the dipole theory, the application of the solid angle, the concept of ventricular gradient, etc. Somebody able to reason out a complex pattern in this way may also untangle whether a certain image is classical or has never been described.
In clinical electrocardiography there are many examples that can not be interpreted without their clinical counterpart and the helpful assistance of other cardiologic information bringing together basic science, genetic studies, a thoracic X ray, echocardiographic exploration and nuclear studies to reach a more refined judgment of the ECG. Unrecognized tracings analyzed with all this information may be extensively and easily understood in a better and more accurate manner. Consequently, I totally agree with Dr. Batchvarov that purely clinical ECG studies just speculating about the underlying mechanisms should be discouraged”.
