Tuesday, 8 May 2018

ECG in children - an amuse bouche (and what to do with one)

Food etiquette is a minefield.  I remember the first time I was presented with an amuse bouche.  I had no idea why it was, let alone what I was supposed to do with it.  One minute later, with the tasty morsel in my belly, I realised I had really over-thought the whole thing.  Also, I was still hungry.  Non-acute paediatric ECG is a lot like that.

Some tests are so simple, usually because they are quantitative.  You do a blood sugar and you get... a blood sugar.  Some tests are much more qualitative, such as Chest X-ray.  Show a chest X-ray to half a dozen radiologists and you may be surprised by the range of interpretations.  ECG in children definitely falls into the second category in that it is a test which requires interpretation.  The interpretation of an ECG is fairly standard, but I've never yet met a standard child or a standard clinician.

There are many reasons why an ECG might be done for a child.  I am not talking about during an acute presentation such as a severe tachycardia or other signs suspicious of a cardiac cause for a child to be unwell.  I am talking about ECG in a child who is well, but had a symptom that warranted an ECG.  If you want to know about the kind of critical care ECG interpretation best done wearing a cape and with underpants on the outside of your trousers, you might like to listen to the PEMplaybook.
If your cape is in the cupboard and underwear is wherever you normally keep it, then ECG is a more fickle friend.  When an ECG is normal normal then that's great but it's not the end of the story.  Often it appears abnormal, because paediatric ECGs look different much of the time.  In a BMJ article about ECGs, the authors write, "Chest pain in children is rarely cardiac in origin and is often associated with tenderness in the chest wall. Electrocardiography is not usually helpful in making a diagnosis, although a normal trace can be very reassuring to the family." (1) That is all very well if with a 12 lead ECG you can tell them that all is normal.

So what is the problem with paediatric ECG?  Well actually there are two problems.  The first is the issue of things that look abnormal and are not.
Much of the differences in paediatric ECG are to do with the initial right sided dominance.  The other thing that can be a factor is physics.  The ECG may show up as LVH, RVH, atrial enlargement etc, but this is often because there is just very little in the way of chest wall between the sticker and the myocardium.  You don't get this problem in children with a more substantial chest wall.  Simply put, in a small or skinny child, large waves are usually normal. If something appears big, look at the child and check the axis.  A skinny child and a normal axis means that the "LVH by voltage criteria" is probably a lie.

Knowing these things helps us to be able to say more often, "This is a normal ECG."

Then there is the opposite problem: a 12 lead ECG done when asymptomatic does not rule out significant pathology.  Take this case study as an example:

A 12 year old girl presents having had a collapse while playing tennis.  She had no palpitations or chest pain and simply recalls feeling faint just before she collapsed.  The adult playing tennis with her describes a sudden collapse, while she was walking t pick up a ball.  When they ran over, the young person was unresponsive for only a few seconds before slowly coming around.

Clinical examination is normal.  The only other history of note is in the family history - a sudden unexpected death in infancy of a 9 month old sibling.  

Would you be happy to rule out a cardiac arrhythmia based on a resting, asymptomatic 12 lead ECG?  With that history, I wouldn't recommend it.

So, if an ECG in a child who is currently asymptomatic has a lot of false positives and false negatives, what is it useful for?  The answer to that is that it should mainly be used to answer specific questions.  For example, in the case above, I want to know the corrected QT interval.  A 12 lead ECG will tell me that.  In fact, resting asymptomatic ECGs are mostly useful for checking rhythm and intervals.

Morphology and high voltage account for most of the false positives and normal rhythm can be a false negative.  Both these things are fine, because a 12 lead ECG is simply an amuse bouche.  If you aren't hungry (no red flags), an amuse bouche is simply a tasty mouthful which won't fill you up.  It is debatable whether it is even needed in a child who has had a typical faint with no red flags.

If you are hungry (red flags in the history or examination), don't rely on an amuse bouche, which should just be there to keep you happy until the real food arrives.  What you have for your main course depends on where you work.  Perhaps you have direct access to 24 hr ECG and cardiac echo. Me? I phone a friend to do these things for me.

Edward Snelson
The Gourmand of Child Health

Disclaimer - I may not have had all of the ECGs the right way up.  That could explain some of the abnormal morphology.

  1. Steve Goodacre, Karen McLeod, Paediatric electrocardiography, BMJ 2002;324:1382
  2. PEM playbook EKG killers