The Anion Gap - What causes it and why should we care?

The anion gap is a calculated measure. It is used to help determine metabolic causes of acidosis and can suggest underlying pathology in an undifferentiated patient.

The full formula for calculating an anion gap is:

([Na+] + [K+]) − ([Cl-] + [HCO3−]) = 16 mEq/L

This is frequently shortened to:

= [Na+] − ([Cl−] + [HCO3−]) =12 mEq/L

A high anion gap acidosis has numerous causes. Before we had the ability to measure lactate directly, a high anion gap would suggest that a patient had a lactic acidosis or one of several other metabolic derangements. The “MUDPILES” mnemonic has been greatly expanded over the years:

https://en.wikipedia.org/wiki/High_anion_gap_metabolic_acidosis#/media/File:Cat_mudpiles_-_causes_of_high_anion-gap_metabolic_acidosis.svg

Some have suggested other mnemonics where drugs that were included in MUDPILES, but rarely seen in practice, were replaced by more frequent causes of a high anion gap acidosis.

From the Lancet for the 21st Century

G — glycols (ethylene & propylene)
O — oxoproline (metabolite of tylenol)
L — L-lactate
D — D-lactate
M — methanol

A — aspirin

R — renal failure

K — ketoacidosis (starvation, alcohol, and diabetic ketoacidosis)

http://www.thelancet.com/journals/lancet/article/PIIS0140-6736(08)61398-7/fulltext

In most contemporary hospitals and laboratories, the lactate level can be measured directly. This removes the need to memorize a subset of lactate-producing diseases and allows a simpler mnemonic:

Lactate vs Non-lactate-generating Causes

K — Ketoacidosis (DKA, AKA)
U — Uremia
P — Production (Lactic acidosis, D-Lactate)
I — Ingestion (Ethylene glycol, methanol)
N — Need-to-Remember Drugs

(eg: aspirin, isoniazid, iron,toluene.)

Note that when dealing with the large anion gap acidosis you can subtract lactate level directly from the anion gap. Be careful to do this only when lactate is calculated in mmol/L (this is the units we use when we set the level of “4” for severe sepsis). If there is a still a significant gap after accounting for lactate, then you need to consider that there may be additional etiologies for their metabolic derangement.

Now, what is the “strong ion gap”? This concept is covered well in Lifeinthefastlane at http://lifeinthefastlane.com/ccc/strong-ion-difference/

Their short summary is that while the strong ion gap is accurate, it is a complex equation and doesn’t appear to have significant added benefit in clinical practice.

The last thing we want to talk about is what tests you should use to calculate your anion gaps. Most hospital labs have basic metabolic panels, comprehensive metabolic panels and blood gases (sometimes call shock panels). The short answer is that you can use any of them to calculate a gap. There are small differences between the tests but they all should give you the same approximate answer with the assumption that your labs are careful in how they handle the samples and are timely in running the tests.

The longer explanation of the difference between the types of tests is as follows:

The measurements of ions (Na⁺, Cl^–, K⁺) that are determined from basic/comprehensive metabolic profiles (BMP/CMP) are from plasma derived from “spun” samples. “Spun” just means that a centrifuge is used to separate the cells from the plasma in the specimen. So, in those spun samples, broken cells will spill their intracellular contents into the plasma that then gets concentrated with the centrifuge. This analytic method matters most for K⁺, as it’s mostly intracellular. At the same time, we often use the abbreviated calculation for anion gap that doesn’t include K^+,so this is a moot point. BMP/CMP are considered the “standard measurement” as these tests have been around for decades.

Conversely, the blood gas or “shock panel” provides measurements from whole blood (“non-spun”) samples. In theory, this should give a more accurate electrolyte concentration and may explain why K⁺ is routinely 0.4-0.5 mmol/L lower in blood gases even in non-hemolyzed samples. However, the downside to the blood gas is that HCO3-is not measured directly. It is calculated from the pH using the good ol’ Henderson–Hasselbalch equation. That equation assumes a number of variables like temperature, concentration, and standard ranges for pH. Variation of those components will alter the “K” constant in the Henderson–Hasselbalch equation and therefore change the HCO₃^– result.

In practice, there is extremely good consistency between BMP/CMP and shock panels especially in routine clinical use. A large anion gap calculated through either method should compel you to start thinking of the numerous causes that lead to an anion-gap-acidosis.