Tuesday 13 June 2017

Think Sepsis - What does that mean?

You may have noticed that there are a lot more paediatric sepsis guidelines flying around these days.  When people write guidelines, they are trying to be helpful but it is always worth knowing why they decided to be helpful.  For example, guidelines for diagnosis of asthma in childhood are strongly motivated by the desire to reduce the overdiagnosis of asthma in childhood.  We know this because the guideline writers tell us.  They're quite good like that.

So what about sepsis?  Well there are two genuine problems that keep coming up around sepsis:
  • Early diagnosis of sepsis
  • Early and aggressive treatment of sepsis
There's probably more to it than that but that is the main thrust of what most sepsis guidelines are trying to achieve.

I think that the guidelines that have come out over the past few years have done a good job in guiding our management of sepsis.  Once you have decided a child has enough evidence of being septic to be treated, crack on and don't spare the horses.  There is no doubt that as a profession, we are getting our act together in this respect.

The first part is more tricky. diagnosing sepsis is difficult.  Sorry, let's be honest, it is really, really difficult.  Sepsis is missed all the time, and I am not talking about the overdiagnosis of missed sepsis which goes like this:

There is a two part truth which guideline writers and readers need to accept.  Sepsis is often missed because it is often easy to miss it.

So, back to the guideline writing - in order to help us diagnoses sepsis, guidelines have been written to help us to recognise sepsis.  As a colleague of mine recently pointed out, that only works if you know to look at the sepsis guideline.  If you are already looking at the sepsis guideline then the battle is already won, because if you are worried enough to look at the sepsis guideline, it's usually time to phone a friend.

So why is it easy to miss sepsis?  There are several reasons;
  • The diagnosis of sepsis is subjective.  There is no mathematical equation (Fever + Tachycardia ≠ Sepsis), test or even definition that gives anyone the answer to the question does this child have sepsis. 2016 saw the third meeting of the Society of Critical Care Medicine and the European Society of Intensive Care Medicine in an attempt to achieve a consensus definition of sepsis.  They will need to meet up again in 2018 if they are going to find true consensus about what sepsis is and what it looks like, since they couldn't quite decide the first three times.  Perhaps they just miss each other that much between conferences.
  • Sepsis doesn't appear, it develops.  There is a reason that we don't have an issue with clinicians missing a diagnosis of croup.  Croup announces its arrival most helpfully.  If only sepsis did the same.  
  • There is almost always another diagnosis to distract the clinician.  As mentioned above, before a child is diagnosable with sepsis, they usually have a prodromal illness.  A classic example of this is secondary sepsis in children with chickenpox.  It is completely understandable that when a child presents with fever and being miserable, having the typical chickenpox rash seems to make the diagnosis obvious.  However, some of these children have sepsis, and it is important to know when that is a strong possibility.
  • All of the features of sepsis are non-specific and can belong to another diagnosis.  Features such as tachycardia are frequently difficult to interpret as a fast heart rate may be due to pain, fear or pyrexia - all of which occur in children who do not have sepsis.  In any case, there is no definition of tachycardia, so we rely on guideline figures.  
  • You can't rely on any one feature to be present all of the time.  Even pyrexia may be absent.  Recognising severe sepsis is relatively easy, but we are being asked to recognise sepsis earlier, before it becomes severe.  That is much more of a challenge.

So, in summary, early sepsis is vague and it is easy to miss because it often hides behind a more obvious diagnosis.  The result is that guidelines are inherently too vague or too prescriptive when it comes to the recognition of sepsis.  Furthermore, if sepsis is not considered, the guideline is of no use whatsoever.

There is relatively simplistic way to deal with all of this.  Since the challenges are mainly about awareness and decision making, I think that a sepsis guideline could simply look like this:

Here are some footnotes on this:
  • Recognising the unwell child starts with recognising the well child.  All sorts of factors are taken into consideration.  While many guidelines emphasise physiological values (heart rate etc.), the behaviour and activity of a child are very important.  The gut feel assessment of the parent and the clinician are also valid.
  • The trajectory of the illness is not always treated with the importance that it deserves.  Children who are not septic often have periods of being subdued but then pick up and have a time where they look and behave as though they are much improved.  This is the "I can't believe how well my child looks now doctor!" effect.
  • Recognising sepsis comes with experience but any clinician can think about the possibility of sepsis.  If you are unsure, get a further assessment.

So, thinking about sepsis is the crucial first step.  It's the deciding that takes the most skill.  Then, when it comes to acting, we should be getting on with with doing whatever we need to do without unnecessary delay.  Hopefully that has made it sound a lot more simple than it really is.

Edward Snelson
Simple is what I need
@sailordoctor

Disclaimer - Simplification is a huge cop out for medical writers, but it's also a lot of fun.  Try it sometime.