Our teaching on ABGs is based on 1940s lab technology, when we could not measure things like albumin or lactate. This led to a high number of unmeasured anions that were grouped as the ‘anion gap’. We are also taught at medical school and on ALS to follow a 5 step approach, which does not take into account some pretty important determinants of blood pH. I recently came across EmCrit’s podcast on Acid Base and it has revolutionised my understanding.
Firstly, plain old bicarbonate is not a very useful value on an ABG. The body has an endless capacity to produce H+ and bicarbonate ions depending on the chemical environment. They are dependent variables. What we actually want to know are the independent variables that determine the levels of H+ and bicarbonate. It turns out that there are four sources, which we will look at in an example.
The base excess is a much more handy tool in the ABG box. The base excess is the amount of acid you need to add to the blood to bring it back to a pH of 7.4, assuming that the PCO2 of the blood were actually 40 mmHg. This means that the respiratory component is taken out, and so the base excess is a pure metabolic indicator of acid/base status.
Once you get your ABG results, look at the base excess. The normal range is -2 to +2. Your task is to account for anything abnormal. Let’s go through an example.
Say the base excess is -10. You need to find why we are lacking 10mmol of base, or in other words why we have 10mmol too much acid. I said earlier there are four sources that explain all acid/base changes, so let’s go through them.
The first is the Strong Ion Difference. This can be simplified as the difference between [Na] and [Cl]. The normal difference is 38 (based on serum [Na] = 140, [Cl] = 102). Any decrease in this difference leads to acidosis. A SID of say 32 explains 6 mmol of acid, as this is 6 less than 38. This is particularly important when it comes to IV fluids, especially saline 0.9% which has a SID of 0 ([Na] = 154, [Cl] = 154). Just 2L saline 0.9% in 24 hours can cause an acidosis.
The second is albumin, which is actually a weak acid. Take 42 g/L as normal, and every 1g increase above that explains 2.5 mmol of acid.
The third is lactate. A value of 3.5 mmol explains 3.5mmol of acid.
The final groups are the things which you normally put in your raised anion gap, like ketones, aspirin, ethylene glycol etc. Note lactate is already accounted for.
The beauty of using this quantitative approach is that you have accounted for every single mmol of acid/base. For example, patients recovering from DKA can remain acidotic not because of ketones, but because of the crazy volumes of saline administered. Armed with your quantitative approach, you can look at the electrolyte levels and prove to yourself (and your stressed registrar) that the acidosis is explained by too much chloride.
(This is a simplification – see http://www.acid-base.com/strongion.php for a fuller story. The strong ion difference should be [Na] + [K] + [Mg] + [Ca] – [Cl] and [other anions], and the other weak acids along with albumin include phosphate. The raised phosphate in CKD may partially explain the acidosis that is typical of the disease, although the failure to excrete the usual 1 meq/kg/day of acid generated by the metabolism of sulphur containing amino acids is probably the dominant mechanism there. However, this method is applicable clinically and sticks in the memory nicely. Try using it to analyse the next few ABGs you see and it quickly becomes second nature).