The paediatric ECG is both a friend and foe to the general paediatrician. It can contain a wealth of direct information, but for most, creates an aura of uncertainty. I became interested in ECG interpretation as a junior registrar when I, on a post-take ward round, was told to ‘send it across to cardiology for interpretation’. Now, whilst I do not disagree that ECGs can be tricky, and in those circumstances, paediatric cardiologists are invaluable, at that moment I wondered, why can’t a general paediatrician make a start? After all, in the adult arena ECGs are like paracetamol, you can barely make it over the hospital threshold without having one. I firmly believe we as paediatricians are out of practice and therefore confidence, but that can be changed.
We could be better at interpreting ECGs
A paper published in the Archives of Disease in Childhood highlighted just how bad we are at ECG interpretation.
Crocetti M, Thompson R. Electrocardiogram interpretation skills in pediatric residents. Ann Pediatr Cardiol 2010;3:3–7.
Why is this an issue?
Anyone who has trained in paediatrics will know that most of us receive pretty much no formal ECG training. In fact, unless we work in ED, we actually won’t be asked to interpret ECGs that often at all. But clearly, these are skills that all paediatric trainees need, and being able to identify some common ECG abnormalities is essential.
One study was run by Snyder et al (2005) in the US, where doctors were asked to identify 10 ECGs – junior residents scored 46% correctly; while senior residents faired a bit better at 64%. These results were more or less replicated by Crocetti et al (2010) in a similar assessment of residents.
This study assessed the skills of ECG interpretation in UK paediatric trainees and then gave them some ECG education.
Who did they ask?
The authors contact 8450 members of the Royal College of Paediatrics and Child Health, UK (this means junior doctors in training) identified by searching through the RCPCH’s member list on their website.
There were 764 who agreed to participate (385 completed the study). 266 of the original participants had received either formal ECG training or previously worked in a paediatric cardiology post.
What were they asked to do?
The participants completed a quiz containing 10 ECGs. Each question had five possible answer options.
The 10 ECGs were: normal neonatal ECG; normal paediatric ECG; atrial fibrillation; complete heart block; Wolff-Parkinson-White syndrome; atrioventricular septal defect with the typical superior left axis; right bundle branch block; prolonged QT interval; myocarditis or pericarditis; and hyperkalaemia.
They were then given one page of ECG teaching to read and repeated the test (with different ECGs but the same conditions).
Did they pass?
Participants on the pre-education quiz scored 63.5% and on the post-education quiz scored 73.3%. This demonstrated a significant improvement with the educational intervention (p<0.001).
There was no difference between senior and junior trainees and no difference between those with previous ECG/cardiology training and those without.
What about the ones that didn’t complete the quiz?
There were 493 people who did the first quiz and didn’t complete the second quiz. Their score on average was 61.5%.
What were the study limitations?
The group is self-selecting out of all UK paediatric trainees. It could be that those who were most worried about their ECG skills, or those who were keenest to learn and improve agreed to participate.
Also, improving test scores in such close proximity to educational intervention doesn’t necessarily teach us much. What matters is whether or not we can retain that information for future practice, and this was not tested by the authors.
Understanding the trace
Let’s start at the start, what is an ECG? The electrocardiogram has been in use since the early 1900s and captures electrical activity emitted from the heart. The movement of ions across cell membranes is fundamentally responsible for this activity and it happened in a predictable and coordinated way in a normal heart.
P wave – represents atrial depolarization. The sinoatrial node is sited in the right atrium and as this discharges, electrical activity spreads from there across towards the left atrium. The P wave is therefore made up of two waves (RA and LA depolarization) but in time these are so close together that they become one on the trace
PR interval – a short lag as the electrical activity reaches the atrioventricular node
QRS complex – ventricular depolarization
ST segment – an isoelectric phase immediately prior to the heart resetting itself
T wave – ventricular repolarisation
For the vast majority of children, this process will be repeated 60-150 times a minute, every minute of every day.
Having a good feel for what the trace means and where these waveforms are emitted from is key to figuring out what is going on when things aren’t normal and in what area to focus your attention if there has been a problem but things have resolved at the point of contact.
Take a systematic approach
The heart is systematic in the way it organises activity and the ECG trace produced reflects that, so we need to be systematic in our interpretation.
Information – we always check demographics. Here there is also a scale. The standard ECG should be set at 10mm/mV on the y-axis and 25mm/sec on the x-axis. All of the standard calculations depend on this so it needs to be checked. With those standardised values, one small square becomes 0.04s and a large square 0.2s.
Rate – this can be simply calculated using 300 divided by the number of large boxes between two R waves.
Rhythm – most ECGs will be normal sinus rhythm (NSR) so it important to be able to confidently characterise that. The standard criteria are:
- The normal rate for age
- Normal P wave axis (it should be upright in lead II, III and aVF). The axis of the wave gives you information about where in the RA it is situated.
- Normal P wave morphology (shape). This fundamentally means that the current is moving away from the sinus node in a normal pattern and over a normal time period. If the P wave looks broad and contains a notch it takes up a greater part of the x-axis/time suggesting left atrial enlargement. If it has a peaked/spiky appearance then the voltage is high ie right atrial enlargement
- Normal PR interval
- P wave preceding every QRS complex
Note that most of the criteria for NSR, perhaps not unusually, focus on and around the P wave. If criteria 2-5 are met but the rate is abnormal then you would have sinus….tachycardia, bradycardia or arrhythmia.
Axis – here we are looking at where the bulk of the cardiac impulse is heading. Outside of early infancy, this should be towards the muscle bulk of the left ventricle. The axis is calculated by looking at leads I and aVF (but any two perpendicular leads could be used) and looking for the net deflection. This is then plotted along lead I (either in the positive or negative direction depending on whether the deflection is up or down), added to the end of the line along lead one is a similar line representing the overall deflection of aVF. A further line is then drawn from the base of the first line to the tip of the line along aVF the angle created is the cardiac axis.
Waveforms – the shape of each wave is important and the more you see the more familiar this will become, however, for me even more important than this is the return to basics. If you know WHAT the wave represents then you have identified where the issue is…
Intervals – there are reference tables for paediatrics so remembering them is unnecessary. PR and QRS durations are straight calculations from the trace. Simply counting the number of boxes will suffice. The QT interval needs to be corrected for rate so takes a little more work.
QTc = QT/√ (RR) The machine will give you a reading, however, I advocate manual calculation for greater accuracy based on where the T wave truly ends.
Summary – I like to think of this as a case presentation, but about the child’s investigation. Given you have systematically looked through everything it should be a short process to pull this into a summary and if there are abnormalities being discussed with cardiology you will be able to succinctly highlight these, even if the diagnosis remains elusive.
Why are you looking at an ECG?
There are, of course, many reasons for performing an ECG but I’d argue this boils down to just a few important ones in acute general paediatrics. Consider what you are really looking for as some changes can be subtle and the primary complaint may have resolved to leave you with only a shadow to guide you to the underlying issue.
- Chest pain – a common reason for ECG, but finding a cardiac problem is very unlikely – around the magnitude of 1%. How suspicious you are will depend a lot on the history but things such as pericarditis, myocarditis and myopathies might be in your differential. If they are in there then look HARD! These conditions irritate the ventricular pericardium and myocardium so changes will be seen in the latter part of the ECG.
Anomalous coronaries can also create exertional cardiac chest pain and may not have any ECG findings so again the history is key, is it convincing? Yes? Is there a better alternative explanation? No? Then you need an echocardiogram, and tell the operator what you are worried about!
- Palpitations – you may find the arrhythmia and be in resus giving exciting drugs on a cardiac monitor. It is far more likely, and less frightening, to have a child back in normal sinus rhythm but describing something that might be an arrhythmia. So what could you look for? There might be nothing, but if the rhythm was SVT you could see an accessory pathway on the ECG – a delta wave leading from PR-QRS – it’s easy to miss. Thankfully, ventricular arrhythmias are far less common in children, especially those with structurally normal hearts, but there might be evidence of arrhythmic disorders such as LQTS. These disorders put children at higher risk of ventricular arrhythmias and fittingly, any clues are therefore found in this area so be suspicious.
- Syncope – many adolescents have syncopal episodes due to autonomic BP control issues. The reason for the ECG is to weed out those where syncope has in fact been due to a short-lived cardiac arrhythmia.
So there you have my approach to ECGs. With a comprehensive system and one eye on what you might find they don’t have to be feared but should become part of your investigation battery and a fruitful discussion with cardiology when require.
Understand what the different areas of the trace represent
Be systematic and look at each area individually
In combination with the history be super critical of the areas where you could find a small clue to something bigger
Crocetti M, Thompson R. Electrocradiogram interpretation skills in pediatric residents. Ann Pediatr Cardiol 2010;3:3–7.
Jheeta JS, Narayan O, Krasemann T. 215 The Accuracy in Interpreting the Paediatric Electrocardiogram and the Need for Further Training. Archives of Disease in Childhood. 2012 Oct 1;97(Suppl 2):A62-.
Snyder CS, Bricker JT, Fenrich AL, et al. Can pediatric residents interpret electrocardiograms? Pediatr Cardiol 2005;26:396–9.