About Looking at an Electrocardiograph: Part 2

This blog is the second part of a talk given to Emergency Department Registrars in August 2016. The theme was "Looking at an Electrocardiograph".

The first part of the blog had three main topics:

  • Learning
  • Seeing
  • Looking

In this second part the emphasis will be on:

  • Narration (description)
  • Structured approach (to description or analysis or both)
  • Diagnosis: a necessary but not sufficient process


Ways of Seeing ECGs

We will use the writings of John Berger (b. 1926) to introduce this second part of the blog. Berger is an English art critic, novelist, painter and poet. He has has written subtle and insightful articles about photography, art, politics, and memory. The two works of Berger that are relevant to ECG evaluation are "Ways of Seeing" and "About Looking". He is perhaps best known for Ways of Seeing (WOS).

"Ways of Seeing was first broadcast by BBC in 1972, as a four-part TV series. In the series, Berger, long-haired, his shirt open at the neck, set against a blue background void of books and other symbols of knowledge, acts as the author, presenter and iconoclast. He is a charismatic speaker who talks directly to his audience, patiently explaining his ideas, mercilessly demolishing the bourgeois idea of art.  ......... The book version, published also in 1972, extends the four-part TV series into seven numbered essays, three of which use only images.......

More than anything else, Berger is critical of art’s aura of unassailability and mystery. Berger dismisses the cultural authorities, elitism, and the false illusion of cultivation that accompanies an interest in art." Source: Peter Bilak. https://www.typotheque.com/articles/ways_of_seeing_book_review


There are two quotes from the book version of WOS that are relevant to this blog:

"The relation between what we see and what we know is never settled" (Ways of Seeing John Berger 1972)

"The way we see things is affected by what we know or what we believe" (Ways of Seeing John Berger 1972)


About Looking at ECGs

About Looking is a collection of essays by John Berger, first published in 1980, in which he explores our role as observers to reveal new meaning in what we see. One of the essays is "Why Look at Animal?". This dealt with the cultural and symbolic links between animals and man, the economic links, and the practical value of domestic (or domesticated) animals. This long established relationship between man and beast has been reduced in industrialised societies in the twentieth century to looking at each other in zoos (Figure 1). This change has also resulted in an increasing number of household pets, animal farms and "cute" You Tube videos of pets.

Figure 1 Who are you? At the Melbourne Zoo

"The eyes of an animal when they consider man are attentive and wary....... The animal scrutinises him across a narrow abyss on non-comprehension.......The man too is looking across a similar, but not identical, abyss of non-comprehension.....And yet the animal is distinct, and can never be confused with man"  (Why Look at Animals? John Berger) 

Berger's essays and ideas discuss cultural and aesthetic issues that seem unrelated to the (apparently) objective process of ECG evaluation. However we can link the title of two works to summarise the "nuts and bolts" of ECG evaluation: 'Ways of Seeing' ⇔ "About Looking'.


I Dream of (EC)Geannie

Figure 2: I Dream of (EC)Geannie

Figure 2 has a photograph taken by the American photographer Lee Friedlander, part of a series of photographs of television screens that he took during the 1960s. This picture was taken in 1962. The series was called The Little Screens, and first appeared as a 1963 picture essay in Harper’s Bazaar, with commentary by Walker Evans. Six untitled photographs showed television screens broadcasting glowing images of faces and figures into unoccupied rooms in homes and motels across America. Between 1963 and 1969 the series grew but was not brought together in full until a 2001 exhibition at the Fraenkel Gallery, San Francisco.

The cover of the Penguin edition of WOS had a painting called the Key to Dreams, one of a number of works with the same title by Rene Magritte (1898-1967). Rene Magritte was born in Belgium and came from a petty bourgeois milieu in which keeping up appearances was of prime importance. A crucial childhood experience was the suicide of his mother, who threw herself into the River Sambre in 1912 and whose body was not found until weeks later.

His always dapper attire, his bowler hat, and the small, humble apartment with studio in the living room provided a facade of normality that jarred with the enigmatic and unsettling paintings of ordinary scenes or objects.  

One of Magritte's well known images is The False Mirror Le Faux Miroir (Figure 2). The picture shows an enormous lash-less eye with a luminous cloud-swept blue sky filling the iris and an opaque, dead-black disc for a pupil. The title, provided by the Belgian Surrealist writer Paul Nougé, suggests there are limits to optical vision: a mirror provides a mechanical reflection, but the eye is selective and subjective.

Magritte also challenged the traditional way of linking (or sharing) words and images in the same frame. This link is not as linear or concrete or straightforward as we assume. The point is well made in his painting Key To Dreams 1935 (Figure 2). In this picture there are four frames, one image in each frame, and a word in each frame. The word and the object only (seem to) correspond in the lower right hand frame. 

The black background suggests night, and the frames form a window looking into the night. Linking "night" with each word produces:

  • Night-wind (poem by Emily Bronte, also the alias of Bingham Harvard, the main character in a 1913 story about a fugitive falsely accused of a crime who is trying to prove his innocence)
  • Night-mare
  • Night-bird (Night Bird Flying was a song written by Jimi Hendrix and released on his posthumous album The Cry of Love in 1971)
  • Night-case or Overnight bag

The ECG has (superficial) similarities to the Key to Dreams series of paintings: there are different frames with different images that can be described by various terms. The interpretation of ECGs and the works of Magritte also deal (in different ways) with semiotics: "the study of signs and symbols and their use or interpretation"

Another famous work by Magritte is a painting of a pipe titled "This is not a pipe" (Ceci n'est pas une pipe 1948). In an interview Magritte said: 

"It's quite simple. Who would dare to say that the PICTURE of a pipe is a pipe? Who could smoke the pipe in my picture? No one. So IT IS NOT A PIPE". (Interview with Louis Quiévreux. René Magritte Selected Writings Translated by Jo Levy Alma Books 2016) 

Note: I Dream of Jeannie was a fantasy situation comedy (sitcom) that described the relationship between a 2000 year old female "genie" (a supernatural entity [spirit] that had free will, could become invisible, could take on a physical form, and had aspects of a guardian angel) and an American Astronaut, who become her master. It was shown on American TV between 1965 and 1970, with the [predictable] outcome that the 'genie' falls in love with the Astronaut. 


Ways of Recording: The Mechanical Eye

We assume that the recording of an ECG is a abstract or precise process, like that famously described by the Soviet film director Ziga Vertov in 1923 about the movie camera: 

"I’m an eye. A mechanical eye. I, the machine, show you a world the way only I can see it."

While the ECG "camera" may be mechanical the "image" it records is constantly changing; to adapt a quote by John Berger: 

"An ECG* is a image** which has been recreated or reproduced, it is an appearance, or a set of appearances, which has been detached from the place and time in which it first made its appearance and [was] preserved." (WOS John Berger 1972)

(* "image" in the original; ** "sight" in the original)


Signal vs Noise

A standard ECG has 12 leads (13 if you include the rhythm strip). Each complex in each lead has at least 7 named parts (P-Q-R-J-ST-T-U) which may have up to 4 variables (size - shape - position - duration). At a heart rate of 100 beats per minute there will be about 4 complexes in each lead. The data that is potentially available for analysis in a 12 lead ECG is (12 x 7 x 4 x 4) items i.e. 1344 items. 

These items create a data noise that can drown out the signal (the main diagnosis/diagnoses). It is impossible to assess or analyse every squiggle or contour in the ECG: a systematic approach is essential.


You are Not a Computer

Looking at an ECG is not a "passive" or "neutral" process - it is complex and often stressful. It is an active process. The way we look at an ECG is influenced by learnt assumptions and previous experience. 

You are given an ECG with a lot of "noise" (data), often in a challenging clinical situation, and have to make a decision (diagnosis or course of action or both) in a short interval. 

Over the years I have had various reactions in this situation:

  1. Panic ("Why me?")
  2. Diagnostic capitulation ("I'm not a cardiologist")
  3. Aetiolgical confabulation - select a diagnosis that is the least stressful ("aberrant conduction" comes to mind)
  4. Desperate search for some pattern or change that is familiar - this is the"Clutching at Straws" situation). This can result in aRush to Judgement (aka Instant [Mis]Diagnosis)
  5. The Sin of Diagnostic Greed (aka the Over-analysis Syndrome [OAS]).
  6. "Decision Block" - when the noise overwhelms the signal


Walk and Talk and Chew Gum at the Same Time

A popular quote by Lyndon Baines Johnson (LBJ), the 36th president of the United States of America, was:  "Gerald Ford [the 38th president of the United States of America] can't walk and chew gum at the same time". 

This was a sanitised version of LBJs actual words: "Jerry Ford is so dumb he can't fart and chew gum at the same time."

This quote highlights the importance of multi-tasking; in the context of looking at ECGs we must be able to "look & think & describe" (Figure 3).

Figure 3

The first step in ECG assessment is pattern recognition - with experience you will identify the main finding/abnormality most of the time. Pattern recognition is associated with a "One in Five (1in5)" effect: there is a "one in five" rate of overlooking an abnormality [The 1in5 value is an 'guesstimate' based on my experience].

The 1in5 effect may apply to each ECG, but as your ECG analysis experience increases it applies to one in five ECG tracings

The next steps in ECG assessment are analysis (including measurement) and deduction. This sequence is shown as a logical progression in Figure 3, culminating in the diagnosis/diagnoses

This can be simplified into two stages: 

  1. assessment
  2. diagnosis


Words Matter - Importance of Narrative

The acronym "ECG" has two meanings (electrocardiograph or electrocardiogram) that I often use interchangeably, i.e.  'electrocardiograph' is a synonym of 'electrocardiogram'. I know that the terms are different, and my intuitive definitions of the terms are: 

  • electrocardiograph refers to the grid and graphical aspects of the tracing
  • electrocardiogram refers to the output i.e. the tracing. This is a message ("cardiac telegram") that has been sent to the observer (you/me). 

The actual definition of the terms are: 

  • An electrocardiograph is the measurement of electrical activity in the heart and its recording as a visual trace (on paper or on an oscilloscope screen), using electrodes placed on the skin of the limbs and chest (Oxford Dictionary).
  • An electrocardiogram is a record or display of a person's heartbeat produced by electrocardiography(Oxford Dictionary). 

So we look at an image (the electrocardiogram)  that has been produced by image-taking device (the electrocardiograph). The analogy with photography is obvious. 

“Photographs are valued because they give information”
”A photograph is not only an image ........ it is also a trace, something directly stencilled off the real, like a footprint or a death mask.”
”Photographs, which cannot themselves explain anything, are inexhaustible invitations to deduction, speculation, and fantasy”
— Quotes from On Photography Susan Sontag Penguin 1977

One difference between a photograph and an ECG is that we can look at a photograph passively, while with an ECG we actively assess the tracing and provide an interpretation of our observations.

Let us contrast the assessment of a patient with the evaluation of their ECG. The bedside evaluation of a patient is narrative based, with a significant verbal component: history (talking to the patient), examination, synthesis of information, diagnosis (often written, with the writing including the history and examination findings) and talking to another health professional. 

The evaluation of an ECG is different: we look at the tracing, but we usually do not describe our step by step observations or analysis. Our initial ECG diagnosis is usually verbal, brief and sometimes no more than an 'ECG emoji'.

A "look & think & describe [talk] as you look and think" approach to ECG evaluation is an essential skill for those of you sitting exams. It is also a good way of teaching ECG evaluation at the bedside. 

“And functioning takes place in time, and must be explained in time. Only that which narrates can make us understand.”
— On Photography Susan Sontag Penguin 1977

The following case is an example of this approach.  The patient is a healthy 70 year old man with a history of hypertension and diet controlled diabetes who developed mild central chest discomfort associated with hypotension and "flash" pulmonary oedema. 

Figure 4 

Figure 4 is the ECG on admission, and includes a computer analysis as well as the hand written assessment of the treating doctor (the "pattern recognition" phase).

The "look and think and describe" narrative is:

"Sinus tachycardia at a rate of about 115 bpm [using the computer report]. Left axis deviation is present, the QRS complexes are not widened [using the QRSD value of 110 msec in the computer report], there is probable T wave inversion in Lead 1 and definite T wave inversion in Lead aVL (which also has a small Q wave). Because of these changes in the (frontal) anterolateral leads we check the lateral precordial leads (V4 to V6) - there is slight J point depression and ST depression in Leads V5 and V6. Leads V5 and V6 have prominent S waves, suggesting possible right bundle branch block (RBBB). There is a prominent R wave in Lead V2, but no S wave in Lead I, whIch argues against RBBB. Looking at Leads V1 to V3 we see Q waves in these leads and slight ST elevation in Lead V1 to V2".

After this description we can summarise the key findings (in order of importance):

  • Q wave and ST elevation in Leads V1 and V2
  • Anterolateral ST depression and T wave inversion
  • Sinus tachycardia
  • Left axis deviation

Two other finding (an example of the 1in5 effect) are ST elevation in Lead aVR a prolonged QTc interval [using the computer report].


The Fingerprint is Not the Hand: The Rule of Fives

The above example illustrates a narrative based evaluation of the ECG findings. The case showed that an ECG can have a number changes of varying importance or relevance. We often (try to) reduce multiple finding(s) to a single diagnosis. In the above case this would be "changes of acute cardiac ischaemia".

An approach to ECG evaluation based on the main diagnosis or finding has the advantage of brevity, but involves the loss of other information. A single final diagnosis is like a fingerprint - it identifies the finger (the individual [cause[s]) but does not describe the (shape of the) hand ( Figure 5).

Figure 5.  Primate Fingerprints

We can improve our analysis of an ECG by using a five point approach (a 'full hand' description) where we include five main descriptors that include:

  1. Abnormal & relevant findings
  2. Normal findings or non-specific findings or abnormal but non-relevant findings
  3. Frontal plane axis
  4. QT/QTc interval
  5. Transition zone in the chest leads

We will apply this to a normal ECG (Figure 6)

Figure 6

The five point analysis is:

  1. Sinus rhythm with a heart rate of about 83 beats per minute
  2. The QRS morphology and the ST-T waves are normal 
  3. The frontal plane QRS complex is normal
  4. The QT interval is normal
  5. The transition zone is between V2 and V3

The advantage of this approach [compared to the comment "Within normal limits"] is that it reminds us to look at the frontal leads (to determine the axis), the praecordial leads (to determine the transition zone), to consider the intervals (including the QT interval) and the QRS-ST-T morphology.

In the remainder of the blog the Rule of Fives will be applied to some real world ECGs (unlike the exotica encountered in examinations).


Rule of Fives: Case 1

Figure 7: This is ECG of a 42 year olds homeless man with a severe psychosis

Five point analysis of Case 1:

  1. Sinus rhythm, tall and pointed P waves [see Lead II], with the P waves close to the QRS complex
  2. Rightward frontal plane axis
  3. QTc interval slightly prolonged
  4. Transition zone is in Lead V6
  5. ProminentS waves in all the chest leads

Diagnosis: Possible right ventricular hypertrophy; QTc prolongation noted but not clinically relevant in this case.


Rule of Fives: Case 2

Figure 8: This is the initial ECG of a 85 year old woman with a chest infection

Five point analysis of Case 2:

  1. Sinus rhythm with a ventricular rate of 60 bpm and ventricular bigeminy in the first eight complexes*
  2. The voltages of the sinus beats are consistent with left ventricular hypertrophy
  3. T waves in the infero-lateral leads have a very low amplitude
  4. ST elevation is present in Leads V1 to V4
  5. The frontal plane axis and the QT interval are both normal, and the transition zone is in Lead V5

* Additional observations

  1. The R-R’ interval between each sinus beat and the following ventricular ectopic complex (VEC)  - the coupling interval - is the same i.e. we have a constant coupling interval
  2. The T wave of each VEC has a small upward deflection just before its lowest point. This is a P wave produced by the VEC, which has passed retrograde through the atrioventricular node and stimulated the atria

Diagnosis: Possible antero-septal ischaemia; ventricular bigeminy; left ventricular hypertrophy.


Rule of Fives: Case 3

Figure 9

Figure 9: This is the ECG of a 54 year old man with a perforated duodenal ulcer

Five point analysis of Case 3:

  1. Sinus rhythm is present, with a ventricular rate of about 79 bpm
  2. There is left axis deviation and a [complete] right bundle branch block i.e. a bifasicular block is present. The PR interval is normal
  3. The T waves inLeads III and aVF are flat, and the T waves in Leads V3 toV5 are biphasic (with a −/+ configuration) 
  4. The QT interval is at the upper limit of normal
  5. The transition zone is in Lead V5

Diagnosis: Left axis deviation; right bundle branch block; non-specific T wave changes.


Rule of Fives: Case 4

Figure 10: This is the ECG of a 66 year old man with a possible transient cerebral ischaemic event

Five point analysis of Case 4:

  1. Sinus arrhythmia
  2. Incomplete right bundle branch block
  3. Normal frontal plane axis and normal QT interval
  4. Transition zone is in Lead V5
  5. S1Q3T3 is present - this is a normal finding in this clinical situation

Diagnosis: Sinus arrhythmia; incomplete right bundle branch block


Rule of Fives: Case 5

Figure 11: This is the ECG of a 102 year old woman with a fractured neck of femur

Five point analysis of Case 5:

  1. Atrial fibrillation with a ventricular rate of about 80 bpm
  2. Left axis deviation and intra-ventricular conduction defect
  3. Low amplitude T waves, and T wave inversion in Leads V3 to V6
  4. Voltages in frontal leads suggest left ventricular hypertrophy
  5. QTc is slightly prolonged, and the transition zone is in Lead V6

Diagnosis: Atrial fibrillation with controlled ventricular rate; left axis deviation; intraventricular conduction defect; non-specific T wave changes


Rule of Fives: Case 6

This is a 84 year old man with pulseless electrical activity

Figure 12A - Frontal leads of Case 6

Five point analysis of frontal leads of Case 6:

  1. The QRS complexes are widened, with a low amplitude
  2. P waves are not visible
  3. The rhythm is slightly irregular, with a heart rate of 80 bpm
  4. The frontal plane QRS axis (initial 40 msec) is about + 900
  5. The QT interval is prolonged

Figure 12B - Praecordial leads of Case 6

Five point analysis of praecordial leads of Case 6:

  1. The QRS complexes are widened, with a left bundle branch block morphology
  2. P waves are not visible*
  3. The rhythm is slightly irregular, with a rate of 80-100 bpm
  4. The T waves in Leads V2 -V5 are very peaked, consistent with hyperkalemia**
  5. The QT interval is prolonged

* A small upright notch is seen in the invertedT wave of each complex in Lead V6 - this might be a (retrograde) P wave

 ** The serum potassium concentration was 8.3 mmol/L

Diagnosis:  Hyperkalaemia with absent P waves, widened QRS complexes (with a left bundle branch block morphology), prolonged QT interval, and right axis deviation.


Take Away Points: On Looking at ECGs

  • The opening sentence of WOS is "Seeing comes before words". We all develop our own way of "seeing" ECGs - but we also need to able to describe what we are seeing.
  • The process of ECG analysis is more subjective than we realise (“There are no facts, only interpretations.” Friedrich Nietzsche), and is also a two-way process. To misquote Friedrich Nietzsche: "And if you gaze long enough into an ECG*, the ECG* will gaze back into you" (* "abyss"  in the original quotation).
  • Describing the main features of a ECG is more complex and difficult than just reaching the correct diagnosis. 
  • Competence in interpretation and description takes practice, and a standardised approach is helpful. Sometimes other priorities are more important than ECG assessment. 

Figure 13: First Things First