Wide anion gap metabolic acidosis

A recent run through on this topic at ERCAST has some nice pearls on the nuances of wide gap acidosis.

We're all familiar with the mnemonic MUDPILES for the differential diagnosis. So, first, what does it stand for? (There are a few variations on this).

M-methanol (I would add medications, e.g. metformin).
U-Uremia
D-DKA (which should also remind you of AKA, alcoholic ketoacidosis).
P-propylene glycol
I-infection, iron, INH, inborn errors of metabolism
L-lactate
E-ethylene glycol
S-salicylates

I haven't found it all that helpful but there it is. And as the discussants at ERCAST point out it's not all that simple. Here are a few pearls from the discussion:

Toxic alcohols
We routinely check the osmolar gap but it's not as straightforward as we'd like. It's elevated early in the course but goes away as the parent compound is metabolized. Toxic acids are generated by this metabolism and by the time metabolic acidosis is seen the osmolar gap may have closed.

Check an ethanol level because it figures into the osmolar gap calculation. Also, an ethanol level over 100 makes it less likely (though doesn't rule out) that a toxic alcohol ingestion is responsible for the acidosis. The common sense reason for this is that significant ethanol on board makes it less likely that the patient would have sought an alcohol substitute. Moreover, ethanol inhibits the metabolism of toxic alcohols (it was an antidote before we had fomepizole). So if a significant amount of ETOH was ingested before the toxic alcohol (enough to give an ETOH level of over 100) it would protect against the development of acidosis.

The metabolic acidosis induced by toxic alcohol ingestion can be profound, so think of toxic alcohol when you see single digit bicarbs. This is a soft clue, not an absolute rule.

Don't forget to consider propylene glycol as a cause of toxic alcohol related acidosis. Though we are used to thinking of it only in hospitalized patients getting large doses of certain medications (see here) remember that it is contained in “pet friendly” antifreeze.

Hypocalcemia (or a rapid, unexplained downward trend in the serum calcium) is a surrogate marker and may offer a clue for ethylene glycol intoxication.


Ketoacidosis: is it diabetic ketoacidosis (DKA) or alcoholic ketoacidosis (AKA)?
The history and clinical circumstances usually make this distinction clear. When they don't, there are clues. First, if your hospital lab can measure both ketone bodies the ratio of beta hydroxy butyrate to acetoacetate may be helpful, as it tends to be higher in AKA due to the profound redox shift that prevails in that condition. In addition, simple volume repletion and dextrose (or food) will generally correct AKA quickly, in a matter of a few hours. (What the podcast didn't mention is the need to give thiamine before food or dextrose in alcoholic patients).


Lactic acidosis: remember the confounders
Elevated lactate levels may accompany other disturbances and not be the principal (or original) cause of the acidosis. For example, one of the metabolites of ethylene glycol causes a falsely elevated lactate due to a lab artifact. Propylene glycol is actually metabolized to lactate, leading to a real lactic acidosis. Lactic acidosis accompanies salicylate intoxication since salicylate in toxic doses is a mitochondrial poison, uncoupling oxidative phosphorylation.