The anion gap is a quick screening calculation, but it has a blind spot: it depends heavily on serum albumin, which is never directly included in the formula. The corrected (albumin-adjusted) anion gap fixes that blind spot by adding back the contribution that a low albumin level would otherwise hide, so unmeasured-anion accumulation isn't missed in patients with hypoalbuminemia.

How the correction works

The raw anion gap is AG = Na − (Cl + HCO3). Albumin — a negatively charged plasma protein — is the single largest unmeasured anion behind that calculation, even though it never appears in the formula directly. Figge, Jabor, Kazda, and Fencl (Crit Care Med, 1998) quantified how much a low albumin suppresses the gap: for every 1 g/dL that albumin sits below the normal reference value of 4 g/dL, the anion gap is understated by roughly 2.5 mEq/L. The corrected anion gap simply adds that amount back: Corrected AG = AG + 2.5 × (4 − Albumin).

When albumin is exactly 4 g/dL, the correction term is zero and the corrected value equals the raw value. As albumin falls, the correction grows, reaching a maximum addition of 10 mEq/L at an albumin of 0 g/dL.

Inputs and what they mean

Sodium, chloride, and bicarbonate (all in mEq/L) feed the standard anion gap formula, exactly as in the uncorrected calculation. If you already know the raw anion gap, you can skip these three fields entirely and enter the anion gap directly using the "Enter AG Directly" toggle.

Serum albumin (in g/dL) is the value the correction is built around. Because hypoalbuminemia is extremely common in hospitalized and critically ill patients — from malnutrition, cirrhosis, nephrotic syndrome, burns, or acute illness itself — it is worth checking a current albumin whenever the anion gap is being used to screen for metabolic acidosis in a sick patient.

Limits and edge cases

The Figge correction accounts only for albumin. It does not adjust for other unmeasured cations or anions — lithium, IgG paraproteins in multiple myeloma, or bromide can all distort the true gap independent of albumin. Some sources use a coefficient of 2.3 instead of 2.5 mEq/L per g/dL of albumin deficit; this calculator uses the more widely cited 2.5 coefficient from Figge et al. Like the raw anion gap, the corrected value is a screening calculation, not a diagnosis — it should always be interpreted alongside the full electrolyte panel, blood gas, and clinical picture.