Clinical Case 042: Cough with a significant negative

Todays case is not too tricky.  The context is a recent epidemic of bronchiolitis in the Kimberley region.  So lots of coughing, sick babies all over the NW.

This 3 month old girl presented to a remote nursing post (just under 1000 km away) with a one day history of cough, increasing dyspnoea and a fever.  She was seen and thought to have bronchiolitis – young enough and unwell enough to need transfer to Broome.

So we saw her at 6 AM after evacuation by RFDS.  And she looked sick – now hypothermic 35.0, RR = 55, grunting / subcostal recession,  desaturated to 85% off of oxygen.  So something is going on….

At this point I was thinking – probably bronchiolitis, looks pretty unwell.  And then I put on the stethoscope – nothing, no wheeze, crackles or anything to suggest bronch.  She had a flow murmur consistent with her degree of unwellness, but the chest was quiet.  This is NOT bronchiolitis!

So my usual teaching to the students is to never do a CXR on a kid you think has bronchiolitis UNLESS they don’t fit or they don’t follow the bronchiolitis plan.

This kid looked sick and I was thinking sepsis – hypothermia, breathing up, clear sounding chest – maybe compensating for septic acidosis?  So IV access, empirical ABs, and a saline bolus in.  Septic screen – urine catch, blood cultures and consider an LP.  I decided to wake up the radiographer and get a CXR

CXR – RUL consolidation

I thought I would share this case as I never get such a nice clear pic on XR, and I think it demonstrates an important principle of diagnostic decision-making.

I am not to proud to say that I think my auscultation is really not that great in small infants to pick local chest signs.  I think we should realise this and accept that a normal exam in no way excludes lung pathology.

If you have a piece of data that doesn’t fit with your provisional diagnosis, then DON”T ignore it. Sometimes it is a significant negative finding – a quiet chest doesn’t fit with bronchiolitis. It is human nature to try and force data to fit into our working diagnosis – a common error or bias. Ignoring a significant negative is easier than leaving out a positive finding.

2 Comments

Add a Comment

Your email address will not be published. Required fields are marked *