This article may be too technical for most readers to understand. Please help improve it to make it understandable to non-experts, without removing the technical details. (April 2010) (Learn how and when to remove this message)
Pathophysiology sample values
Na+ = 140 Cl = 100 BUN = 20 /
Glu = 150
K+ = 4 CO2 = 22 PCr = 1.0
HCO3 = 24 paCO2 = 40 paO2 = 95 pH = 7.40
pACO2 = 36 pAO2 = 105 A-a g = 10
Ca = 9.5 Mg2+ = 2.0 PO4 = 1
CK = 55 BE = −0.36 AG = 16
PMO = 300 PCO = 295 POG = 5 BUN:Cr = 20
UNa+ = 80 UCl = 100 UAG = 5 FENa = 0.95
UK+ = 25 USG = 1.01 UCr = 60 UO = 800
LDH = 100 TP = 7.6 AST = 25 TBIL = 0.7
ALP = 71 Alb = 4.0 ALT = 40 BC = 0.5
AST/ALT = 0.6 BU = 0.2
AF alb = 3.0 SAAG = 1.0 SOG = 60
CSF alb = 30 CSF glu = 60 CSF/S alb = 7.5 CSF/S glu = 0.6

The anion gap[1][2] (AG or AGAP) is a value calculated from the results of multiple individual medical lab tests. It may be reported with the results of an electrolyte panel, which is often performed as part of a comprehensive metabolic panel.[3]

The anion gap is the quantity difference between cations (positively charged ions) and anions (negatively charged ions) in serum, plasma, or urine. The magnitude of this difference (i.e., "gap") in the serum is calculated to identify metabolic acidosis. If the gap is greater than normal, then high anion gap metabolic acidosis is diagnosed.

The term "anion gap" usually implies "serum anion gap", but the urine anion gap is also a clinically useful measure.[4][5][6][7]


The anion gap is a calculated measure. It is computed with a formula that uses the results of several individual lab tests, each of which measures the concentration of a specific anion or cation.

The concentrations are expressed in units of milliequivalents/liter (mEq/L) or in millimoles/litre (mmol/L).

With potassium

The anion gap is calculated by subtracting the serum concentrations of chloride and bicarbonate (anions) from the concentrations of sodium and potassium (cations):

= ([Na+] + [K+]) − ([Cl] + [HCO
]) = 20 mEq/L

Without potassium

Because potassium concentrations are very low, they usually have little effect on the calculated gap. Therefore, omission of potassium has become widely accepted. This leaves the following equation:

= [Na+] - ([Cl] + [HCO

Normal AG = 8-16 mEq/L

Expressed in words, the equation is:

Anion Gap = sodium - (chloride + bicarbonate)
which is logically equivalent to:
Anion Gap = (the most prevalent cation) minus (the sum of the most prevalent anions)

(Bicarbonate may also be referred to as "total CO2" or "carbon dioxide".)[3]


Calculating the anion gap is clinically useful because it helps in the differential diagnosis of a number of disease states.

The total number of cations (positive ions) should be equal to the total number of anions (negative ions), so that the overall electrical charge is neutral. However, routine tests do not measure all types of ions. The anion gap is representative of how many ions are not accounted for by the lab measurements used in the calculation. These "unmeasured" ions are mostly anions, which is why the value is called the "anion gap."[3]

By definition, only the cations sodium (Na+) and potassium (K+) and the anions chloride (Cl) and bicarbonate (HCO
) are used to calculate the anion gap. (As discussed above, potassium may or may not be used, depending on the specific lab.)

The cations calcium (Ca2+) and magnesium (Mg2+) are also commonly measured, but they aren't used to calculate the anion gap. Anions that are generally considered "unmeasured" include a few normally occurring serum proteins, and some pathological proteins (e.g., paraproteins found in multiple myeloma).

Similarly, tests do often measure the anion phosphate (PO3−
) specifically, but it isn't used to calculate that "gap," even if it is measured. Commonly 'unmeasured' anions include sulfates and a number of serum proteins.

In normal health there are more measurable cations than measurable anions in the serum; therefore, the anion gap is usually positive. Because we know that plasma is electro-neutral (uncharged), we can conclude that the anion gap calculation represents the concentration of unmeasured anions. The anion gap varies in response to changes in the concentrations of the above-mentioned serum components that contribute to the acid-base balance.

Normal value ranges

Different labs use different formulas and procedures to calculate the anion gap, so the reference range (or "normal" range) from one lab isn't directly interchangeable with the range from another. The reference range provided by the particular lab that performed the testing should always be used to interpret the results.[3] Also, some healthy people may have values outside of the "normal" range provided by any lab.

Modern analyzers use ion-selective electrodes which give a normal anion gap as <11 mEq/L. Therefore, according to the new classification system, a high anion gap is anything above 11 mEq/L. A normal anion gap is often defined as being within the prediction interval of 3–11 mEq/L,[8] with an average estimated at 6 mEq/L.[9]

In the past, methods for the measurement of the anion gap consisted of colorimetry for [HCO
] and [Cl] as well as flame photometry for [Na+] and [K+]. Thus normal reference values ranged from 8 to 16 mEq/L plasma when not including [K+] and from 10 to 20 mEq/L plasma when including [K+]. Some specific sources use 15[10] and 8–16 mEq/L.[11][12]

Interpretation and causes

Anion gap can be classified as either high, normal or, in rare cases, low. Laboratory errors need to be ruled out whenever anion gap calculations lead to results that do not fit the clinical picture. Methods used to determine the concentrations of some of the ions used to calculate the anion gap may be susceptible to very specific errors. For example, if the blood sample is not processed immediately after it is collected, continued cellular metabolism by leukocytes (also known as white blood cells) may result in an increase in the HCO
concentration, and result in a corresponding mild reduction in the anion gap. In many situations, alterations in renal function (even if mild, e.g., as that caused by dehydration in a patient with diarrhea) may modify the anion gap that may be expected to arise in a particular pathological condition.

A high anion gap indicates increased concentrations of unmeasured anions by proxy. Elevated concentrations of unmeasured anions like lactate, beta-hydroxybutyrate, acetoacetate, PO3−
, and SO2−
, which rise with disease or intoxication, cause loss of HCO
due to bicarbonate's activity as a buffer (without a concurrent increase in Cl). Thus, finding a high anion gap should result in a search for conditions that lead to excesses of the unmeasured anions listed above.

High anion gap

Main article: High anion gap metabolic acidosis

The anion gap is affected by changes in unmeasured ions. In uncontrolled diabetes, there is an increase in ketoacids due to metabolism of ketones. Raised levels of acid bind to bicarbonate to form carbon dioxide through the Henderson-Hasselbalch equation resulting in metabolic acidosis. In these conditions, bicarbonate concentrations decrease by acting as a buffer against the increased presence of acids (as a result of the underlying condition). The bicarbonate is consumed by the unmeasured cation(H+) (via its action as a buffer) resulting in a high anion gap.

Causes of high anion gap metabolic acidosis (HAGMA):

Note: a useful mnemonic to remember this is MUDPILES – Methanol, Uremia, Diabetic Ketoacidosis, Paraldehyde, Infection, Lactic Acidosis, Ethylene Glycol and Salicylates

Normal anion gap

Main article: Normal anion gap acidosis

In patients with a normal anion gap the drop in HCO
is the primary pathology. Since there is only one other major buffering anion, it must be compensated for almost completely by an increase in Cl. This is therefore also known as hyperchloremic acidosis.

lost is replaced by a chloride anion, and thus there is a normal anion gap.

There are three types.
1. Low renin may be due to diabetic nephropathy or NSAIDS (and other causes).
2. Low aldosterone may be due to adrenal disorders or ACE inhibitors.
3. Low response to aldosterone maybe due to potassium-sparing diuretics, trimethoprim/sulfamethoxazole, or diabetes (and other causes).[13]

Note: a useful mnemonic to remember this is FUSEDCARS – fistula (pancreatic), uretero-enterostomy, saline administration, endocrine (hyperparathyroidism), diarrhea, carbonic anhydrase inhibitors (acetazolamide), ammonium chloride, renal tubular acidosis, spironolactone.

Low anion gap

A low anion gap is often due to hypoalbuminemia. Albumin is an anionic protein and its loss results in the retention of other negatively charged ions such as chloride and bicarbonate. As bicarbonate and chloride anions are used to calculate the anion gap, there is a subsequent decrease.

The anion gap is sometimes reduced in multiple myeloma, where there is an increase in plasma IgG (paraproteinaemia).[14]

Correcting the anion gap for the albumin concentration

The calculated value of the anion gap should always be adjusted for variations in the serum albumin concentration.[15] For example, in cases of hypoalbuminemia the calculated value of the anion gap should be increased by 2.3 to 2.5 mEq/L per each 1 g/dL decrease in serum albumin concentration (refer to Sample calculations, below).[9][16][17] Common conditions that reduce serum albumin in the clinical setting are hemorrhage, nephrotic syndrome, intestinal obstruction and liver cirrhosis. Hypoalbuminemia is common in critically ill patients.

The anion gap is often employed as a simple scanning tool by clinicians at the bedside to detect the presence of anions such as lactate, which can accumulate in critically ill patients. Hypoalbuminemia can mask a mild elevation of the anion gap, resulting in failure to detect an accumulation of unmeasured anions. In the largest study published to date, featuring over 12,000 data sets, Figge, Bellomo and Egi[18] demonstrated that the anion gap, when used to detect critical levels of lactate (greater than 4 mEq/L), exhibited a sensitivity of only 70.4%. In contrast, the albumin-corrected anion gap demonstrated a sensitivity of 93.0%. Therefore, it is important to correct the calculated value of the anion gap for the concentration of albumin, particularly in critically ill patients.[18][19][20] Corrections can be made for the albumin concentration using the Figge-Jabor-Kazda-Fencl equation to give an accurate anion gap calculation as exemplified below.[17]

Sample calculations

Given the following data from a patient with severe hypoalbuminemia suffering from postoperative multiple organ failure,[21] calculate the anion gap and the albumin-corrected anion gap.



In this example, the albumin-corrected anion gap reveals the presence of a significant quantity of unmeasured anions.[21]

See also


  1. ^ Oh MS, Carroll HJ (1977). "The anion gap". N. Engl. J. Med. 297 (15): 814–7. doi:10.1056/NEJM197710132971507. PMID 895822.
  2. ^ Gabow PA, Kaehny WD, Fennessey PV, Goodman SI, Gross PA, Schrier RW (1980). "Diagnostic importance of an increased serum anion gap". N. Engl. J. Med. 303 (15): 854–8. doi:10.1056/NEJM198010093031505. PMID 6774247.
  3. ^ a b c d "Electrolytes: Common Questions: What is anion gap?". Lab Tests Online. American Association for Clinical Chemistry. 24 February 2015. Retrieved 10 November 2015.
  4. ^ Emmett M.; Narins R.G. (1977). "Clinical use of the anion gap". Medicine. 56 (1): 38–54. doi:10.1097/00005792-197701000-00002. PMID 401925.
  5. ^ "Urine Anion Gap: Acid Base Tutorial, University of Connecticut Health Center". Archived from the original on 21 November 2008. Retrieved 14 November 2008.
  6. ^ "Urine anion and osmolal gaps in metabolic acidosis". Retrieved 14 November 2008.
  7. ^ Kirschbaum B, Sica D, Anderson FP (June 1999). "Urine electrolytes and the urine anion and osmolar gaps". The Journal of Laboratory and Clinical Medicine. 133 (6): 597–604. doi:10.1016/S0022-2143(99)90190-7. PMID 10360635.
  8. ^ Winter SD, Pearson JR, Gabow PA, Schultz AL, Lepoff RB (February 1990). "The fall of the serum anion gap". Archives of Internal Medicine. 150 (2): 311–3. doi:10.1001/archinte.150.2.311. PMID 2302006.
  9. ^ a b Kraut JA, Madias NE (2006). "Serum Anion Gap: Its Uses and Limitations in Clinical Medicine". Clinical Journal of the American Society of Nephrology. 2 (1): 162–174. doi:10.2215/CJN.03020906. PMID 17699401.
  10. ^ Nosek, Thomas M. "Section 7/7ch12/7ch12p51". Essentials of Human Physiology. Archived from the original on 23 May 2016.
  11. ^ "The Anion Gap". Retrieved 4 October 2008.
  12. ^ "Anion Gap: Acid Base Tutorial, University of Connecticut Health Center". Archived from the original on 21 November 2008. Retrieved 4 October 2008.
  13. ^ Sabatine, Mark (2011). Pocket Medicine. Lippincott Williams Wilkens. p. 4–3. ISBN 978-1-60831-905-3.
  14. ^ Lolekha PH, Lolekha S (1 February 1983). "Value of the anion gap in clinical diagnosis and laboratory evaluation". Clinical Chemistry. 29 (2): 279–83. doi:10.1093/clinchem/29.2.279. PMID 6821931.
  15. ^ Berend K, de Vries A, Gans R (9 October 2014). "Physiological approach to assessment of acid-base disturbances". The New England Journal of Medicine. 371 (15): 1434–45. doi:10.1056/NEJMra1003327. PMID 25295502. S2CID 1675324.
  16. ^ Feldman M, Soni N, Dickson B (December 2005). "Influence of hypoalbuminemia or hyperalbuminemia on the serum anion gap". The Journal of Laboratory and Clinical Medicine. 146 (6): 317–20. doi:10.1016/j.lab.2005.07.008. PMID 16310513.
  17. ^ a b Figge J, Jabor A, Kazda A, Fencl V (November 1998). "Anion gap and hypoalbuminemia". Critical Care Medicine. 26 (11): 1807–10. doi:10.1097/00003246-199811000-00019. PMID 9824071.
  18. ^ a b Figge J, Bellomo R, Egi M (13 October 2017). "Quantitative relationships among plasma lactate, inorganic phosphorus, albumin, unmeasured anions and the anion gap in lactic acidosis". Journal of Critical Care. 44: 101–10 [Epub ahead of print]. doi:10.1016/j.jcrc.2017.10.007. hdl:20.500.14094/90004761. PMID 29080515. Open access icon
  19. ^ Chawla L, Shih S, Davison D, Junker C, Seneff M (16 December 2008). "Anion gap, anion gap corrected for albumin, base deficit and unmeasured anions in critically ill patients: implications on the assessment of metabolic acidosis and the diagnosis of hyperlactatemia". BMC Emergency Medicine. 8 (18): 18. doi:10.1186/1471-227X-8-18. PMC 2644323. PMID 19087326. Open access icon
  20. ^ Mallat J, Michel D, Salaun P, Thevenin D, Tronchon L (March 2012). "Defining metabolic acidosis in patients with septic shock using Stewart approach". American Journal of Emergency Medicine. 30 (3): 391–8. doi:10.1016/j.ajem.2010.11.039. PMID 21277142.
  21. ^ a b Fencl V, Kazda A, Jabor A, Figge J (December 2000). "Diagnosis of metabolic acid-base disturbances in critically ill patients". American Journal of Respiratory and Critical Care Medicine. 162 (6): 2246–51. CiteSeerX doi:10.1164/ajrccm.162.6.9904099. PMID 11112147. Open access icon