Friday, September 20, 2013

Will we use the electrocardiogram for all it's worth?

The authors of a Special Report in Circulation are concerned about the field of electrocardiography. The beginning of the paper is inflammatory:

The ECG is at a crossroads as to its future integration into modern medical practice. Those most interested in electrocardiography remain the old guard, whose careers evolved with this technology. They remain as enamored by the experiential mythology as by the experimental science of the ECG.

That not only sets an unfortunate tone for the rest of the paper (which makes some good points) but it's also inaccurate. I have pointed out before (e.g. here and here) the emergence of the new generation of leaders in the field. Striving to think beyond the old rigid rules and make electrocardiography more evidence based, they are anything but old guard.

With that aside on to the main points of the paper. (I don't think these were meant to be all inclusive. They are merely a list of examples of how the test has been misunderstood and misapplied).


Students and house staff are inadequately trained in electrocardiography.

As a result many practitioners lack the skills to utilize the electrocardiogram effectively. The authors give the following example:

Their understanding of ST elevation myocardial infarction criteria could be easily exposed by asking them to name the contiguous pairs of standard ECG leads. A disappointing number would refer to pairs of leads that are contiguous on the ECG display such as II and III or V1 and V4, rather than the leads separated by 30° going around the surface of the heart as specified in the guidelines.

That’s a reminder that the array of limb leads on the electrocardiograph tracing is not contiguous!


Traditional rules about contiguous leads are not evidence based and may be obsolete.

For example:

Examples of the experiential mythology that continue to haunt electrocardiography include the requirement for contiguous or adjacent leads instead of a single lead for fulfilling diagnostic criteria. The contiguous or adjacent lead constraint is a residual from the thick, noisy tracings from the early days of electrocardiography before high-impedance amplifiers, DC coupling, and digital processing produced the high-resolution tracings of today (Figure 1). Applying the criteria to a single digitally processed ECG lead would avoid the confusion previously discussed without affecting the diagnostic characteristics of the ECG.

The referenced figure is here. Today's leaders in electrocardiography, particularly those in emergency medicine, are well aware of the various types of “STEMI equivalent”---electrocardiographic patterns that signify acute coronary occlusion but do not meet the STEMI criteria of ST elevation in two or more contiguous leads. Unfortunately the door-to-balloon performance incentive may have encouraged rigid use of simplistic criteria as a substitute for thought and nuanced analysis of the electrocardiogram.


Use of the TP segment as the baseline is open to question.

According to the authors:

The T-P baseline remains from vectorcardiography, whereas the PR segment has many reasons to be set as the baseline, as explained in the Common Standards for Quantitative Electrocardiography statement.

The PR segment as baseline is problematic, however, as it is displaced in pericarditis and sometimes by the wave of atrial repolarization (Ta wave).


The electrocardiographic criteria for left ventricular hypertrophy are of questionable value.

It has long been known that the test characteristics for LVH are less than ideal. On the other hand we have more recently learned the potential value of the electrocardiogram for the assessment of left ventricular systolic function.

The dynamic nature of electrocardiographic patterns complicates the diagnosis of channelopathies.

The electrocardiogram can be very useful in the diagnosis of channelopathies such as Brugada syndrome and the long QT provided this limitation is kept in mind. In the case of the QT interval misunderstanding is widespread. Assessment of repolarization is complex and goes beyond a simple measurement of the corrected QT interval.


Computer interpretations are unreliable.

This remains true despite the fact that we've gone through several generations of machines since the technology was introduced decades ago.


New insights about J waves and early repolarization complicate electrocardiographic interpretation.

Specifically, we now know that there is benign and not so benign early repolarization. Background here, here, here and here.

The question is not whether electrocardiography remains clinically useful. Indeed the power of the electrocardiogram is evident as never before. The real question is whether we will use it for all it's worth.


No comments: