Albumin-Corrected Anion Gap Calculator

How to Use This Calculator

1. Enter measured anion gap and serum albumin for instant corrected AG.
2. From Raw Labs tab: enter Na, Cl, HCO3, and albumin to compute both measured and corrected AG from scratch.
3. Hypoalbumin Effect tab: see whether low albumin is masking a true HAGMA.
4. Professional tier adds delta-delta ratio and full acid-base summary.

Formula

Example

Frequently Asked Questions

• Why correct the anion gap for albumin? ▼
  Albumin is a negatively charged protein that constitutes the largest component of the normal anion gap, contributing approximately 2.0–2.5 mEq/L of unmeasured anion for each 1 g/dL of serum albumin. The standard anion gap formula (Na − Cl − HCO3) assumes a normal albumin of ~4.4 g/dL. When albumin falls below 4.4 g/dL — which is extremely common in critically ill, malnourished, liver disease, or nephrotic syndrome patients — the anion gap is artifactually lowered. If a hypoalbuminemic patient with albumin 2.0 g/dL (a drop of 2.4 g/dL from normal) develops DKA or lactic acidosis that would normally raise the AG by 10 mEq/L, the measured AG only rises by 4 mEq/L (10 − 6 correction), appearing as a mildly elevated value of about 14 mEq/L rather than a more alarming 20 mEq/L. Without albumin correction, this HAGMA could be missed or its severity underestimated. Figge et al. (1998, Crit Care Med) demonstrated this problem extensively and quantified the correction at 2.5 mEq/L per 1 g/dL albumin decrease. The formula: Corrected AG = Measured AG + 2.5 × (4.4 − albumin). This correction should be applied routinely in any patient with documented hypoalbuminemia or in settings where hypoalbuminemia is common.
• When does albumin correction matter most? ▼
  Albumin correction has the greatest clinical impact in three scenarios. First, the ICU and critically ill patients: serum albumin is almost universally low in ICU patients due to inflammation (albumin is a negative acute-phase reactant), redistribution, third-spacing, and reduced hepatic synthesis. Mean albumin in ICU patients is often 2.0–2.5 g/dL, meaning the correction adds 4.8–6.0 mEq/L to the measured AG. A study by Dubin et al. (Crit Care Med 2007) found that in ICU patients with hypoalbuminemia, uncorrected AG failed to detect HAGMA in up to 43% of cases compared to corrected AG. Second, hepatic disease: cirrhosis and liver failure cause profound hypoalbuminemia (reduced synthetic function) alongside multiple competing acid-base disturbances (respiratory alkalosis from hyperammonemia, metabolic alkalosis from diuretics, metabolic acidosis from renal impairment or lactic acidosis). Correcting the AG is essential to detecting the acidotic component. Third, nephrotic syndrome: massive urinary albumin losses cause hypoalbuminemia, and these patients are also at risk for renal insufficiency-related metabolic acidosis. In contrast, albumin correction matters minimally in outpatients with normal or near-normal albumin, where the correction is less than 1 mEq/L and does not change the clinical interpretation.
• What is the formula for corrected anion gap and how should it be interpreted? ▼
  The albumin-corrected anion gap formula (Figge et al.) is: Corrected AG = Measured AG + 2.5 × (4.4 − albumin g/dL). The 2.5 factor represents the approximate charge contribution of albumin per g/dL, and 4.4 g/dL is the reference albumin value at which the normal AG of 6–12 mEq/L was established. Example: A patient has Na 140, Cl 102, HCO3 22, albumin 2.0 g/dL. Measured AG = 140 − (102 + 22) = 16 mEq/L (appears elevated but barely). Corrected AG = 16 + 2.5 × (4.4 − 2.0) = 16 + 6 = 22 mEq/L — clearly elevated, indicating significant HAGMA. The corrected AG should be interpreted using the same thresholds as the measured AG: corrected AG >12 mEq/L is elevated (HAGMA), 6–12 is normal, and <6 is low (consider paraproteinemia). After establishing an elevated corrected AG, the next step is to calculate the delta-delta ratio (ΔAG / ΔHCO3) to detect concurrent mixed acid-base disorders: a ratio of 1–2 indicates pure HAGMA, <0.4 suggests additional non-AG metabolic acidosis, and >2 suggests concurrent metabolic alkalosis. Always check both the measured and corrected AG together — the difference between them tells you how much albumin correction was necessary and flags patients where hypoalbuminemia was masking the severity of their acid-base disturbance.
• Should I always use the corrected anion gap? ▼
  In clinical practice, routinely correcting the anion gap for albumin is recommended as the default approach by most critical care and nephrology guidelines, because hypoalbuminemia is so prevalent in hospitalized patients. Adrogue and Madias's NEJM review series on acid-base emphasizes that failing to correct for albumin is a common source of diagnostic error, particularly in the ICU. However, there are practical considerations. Albumin is not always available simultaneously with other electrolytes. If albumin is normal (3.5–4.5 g/dL), the correction is small (<2.5 mEq/L) and does not change clinical interpretation. If no albumin is available, assume the AG is underestimated in any patient with risk factors for hypoalbuminemia. In outpatient or low-acuity settings, uncorrected AG is usually sufficient. Some centers have implemented automated AG reporting that includes the albumin-corrected value alongside the standard AG whenever albumin is measured on the same blood draw, which is the ideal workflow. Note that the Figge correction is not perfect — it is derived from a physicochemical model and validated primarily in ICU populations. Alternative corrections (e.g., using 2.0 mEq/L per g/dL) exist in some references. The widely cited 2.5 factor is most commonly used and recommended in MDCalc and most textbooks.
• How does hypoalbuminemia affect overall acid-base interpretation? ▼
  Hypoalbuminemia does not just affect the anion gap — it influences the entire acid-base picture through what Stewart's physicochemical approach calls the strong ion difference and total weak acid concentration. In Stewart's framework, albumin is a weak acid (A-total) whose reduction shifts the equilibrium toward alkalosis. This means hypoalbuminemia itself generates a mild metabolic alkalosis (or attenuates an existing acidosis) by reducing the total weak acid buffer load in plasma. Clinical implication: a patient with albumin 2.0 g/dL who has arterial pH 7.38 and HCO3 27 mEq/L might appear to have metabolic alkalosis, but when the hypoalbuminemia contribution is accounted for, they may have a base excess that is actually lower than expected — representing a true underlying acidosis that is partially masked by the alkalotic effect of low albumin. While this level of analysis is beyond routine clinical practice, it is relevant in complex ICU cases where the acid-base picture does not fit a simple explanation. For practical purposes: whenever interpreting acid-base in hypoalbuminemic patients, (1) always correct the AG, (2) be aware that pH and HCO3 may appear more normal than the underlying pathology warrants, and (3) use the delta-delta ratio on the corrected AG to identify mixed disturbances. These steps will detect most clinically significant acid-base problems even without full physicochemical modeling.

Related Calculators