In more detail:
Electroneutrality is always preserved within the body. That is, the number positive ions must equal the number of negative ions. Although there are a billion minor exceptions, the major cations (positive ions) are in the plasma are sodium (Na+) and potassium (K+). In turn, the major anions (negative ions) are chloride, bicarbonate and the plasma proteins.
Because plasma proteins are not routinely measured, and because even when they are, establishing how many negatively charged groups they contain is practically impossible, they are usually left out of the equation. Which equation? This one:
Anion gap = (Na+ + K+) - (Cl- + HCO3-)
Since we've left out a major source of anions (the proteins), the equation usually comes out in favour of the cations. That is, the anion gap is positive under normal conditions - the range is between 10 and 20 mmol/L where I work, but the values depend on the way the ions are measured in each particular lab. As I've made clear, the actual normal value of the anion gap isn't important, since it's quite an artificially constructed equation. What is important is whether or not this value changes when a patient has a metabolic acidosis.
The first step in creating a metabolic acidosis is for the H+ concentration to go up. Initially bicarbonate is consumed in compensation - this is the body's bicarbonate-carbonic acid buffering system in action. When this system is overwhelmed, and acidosis ensues. Whether you view the primary problem as too many H+ ions or too few HCO3- ions, the problem of maintaining electroneutrality can only be solved by the retention or creation of more anions. In many cases, the problem is usually solved by the very process that caused the acidosis in the first place, which seems only fitting! For example, in lactic acidosis, lactate (negatively charged) is produced. Similarly, in diabetic ketoacidosis, acetoacetate and β-hydroxybutyrate are formed (but you can safely forget their names!).
Some other causes of a metabolic acidosis aren't quite so considerate though. In diarrhoea and renal tubular acidosis, for example, no anions are simultaneously produced. In this case, the body deals with the electroneutrality problem by ordering the kidney to retain more chloride ions.
So what happens when you calculate the anion gap in each type of metabolic acidosis? In both cases, the bicarbonate component falls, but in the first instance the chloride level is not adjusted by the body. Run the numbers and you will see that the anion gap will increase in this case.
On the other hand, when chloride ions are retained, their plasma level obviously increases. From the point of view of the equation, this will offset the bicarbonate loss, and so the anion gap is unchanged.
And that's the use of the anion gap - it helps to identify the cause of a metabolic acidosis. A useful mnemonic to remember the causes of an increased anion gap acidosis is:
- K - ketoacidosis (diabetic, alcoholic)
- U - uraemia
- L - lactic acidosis
- T - toxins (e.g. salycilates, methanol, ethylene glycol)