HOW COMPLEX CAN AN ABG really BE?

At fellowshipexam.com we've spent a lot of time on ABGs. They remain one of the most feared and poorly understood props used in the fellowship exam.

So we were thrilled when one of our candidates had the confidence to ask a really hard question on the VSG. Keep reading for the question and the commentary.  

'....a derivation of a derivation using small margins....'

The question we were asked was in regard to a really high lactate in the context of a barely evident raised anion gap. 

The patient was a 50 year old male who presented with a 2 week history of weakness and lethargy associated with vomiting and diarrhoea. He had returned from PNG 2 weeks earlier where he had not taken malaria prophylaxis.

He was diagnosed with Falciparum malaria with a parasitised cell count of more than 10%.

Clinically he was dry and looked unwell. His vital signs were:

• HR 115 
• BP 100/80 
• RR 32
• sats 100% RA.

The venous gas showed:

pH           7.34 
pCO2      41.6 
HCO3      21.8 
BE           -3.7 
Na          124 
K             3.4 
Cl             92 
Glu         10.7 
Lactate   9.7
Creat      129 

So, the question is how can it be that he has a lactate of almost 10 and not a metabolic acidosis? Is he just managing to compensate for the massive lactate?

When you think about it, that's a very good question.

The first thing to remember is that the Boston Bedside rules (the basis of interpreting an ABG) are only approximations. There are two areas this is especially relevant: 

• when calculation margins become quite slim the calculation approximations tend to break down a bit
  
• likewise there is a degree of inherent unreliability in measurements, and when we are talking about slim margins a change of a couple of points makes a big difference to the calculations.

This patient has :

• a mild acidaemia 
• a metabolic acidosis (the bicarbonate is decreased from 24) 
• a very mild raised anion gap (13) and a delta ratio of ~.5 
• appropriate respiratory compensation. 

Taking into account my statements above, the delta ratio (a derivation of a derivation using small margins) is likely to be particularly inaccurate. 

It's worth noting that if you correct that sodium for the glucose you get a delta ratio closer to 1. 

BUT, it's important to realise that the delta ratio is only an approximation, and really is only useful for 2 metabolic processes: 

• defining a concurrent (or isolated, or non-extant) NAGMA to go with a RAGMA 
• defining a concurrent metabolic alkalosis with a RAGMA. 

The question that you are really asking is what happens when all 3 processes are present.

Going back to the clinical information you have provided you have a patient with falciparum malaria - which is associated with a lactic acidosis. 

The question really is that in addition to the RAGMA evident (but of lower degree than expected) could the patient have *both* a concurrent NAGMA (because of HCO3 loss from diarrhoea) and a metabolic alkalosis from HCl loss mediated by vomiting? 

If THAT were the case a couple of things would happen... 

1. Chloride would tend to be retained as part of the NAGMA (due to electrical equilibration in the kidneys, or whatever you want to ascribe it to...) 
2. Bicarbonate would be pushed up by the metabolic alkalosis. Normally you'd expect to see a big drop in the Cl associated with vomiting, but if your kidneys are retaining if due to (1) above this may not be apparent. 

THUS, the two anions you would expect to see pushed down by an impressive RAGMA due to lactate are being retained due to the other pathologic processes occurring, and creating the picture have described above. 

We love taking our candidates through this stuff, and we love having the discussion generated on our virtual study group during our course. It helps all of us learn together.