Link to Statistical Validation of Model
Link to Validation against Independent Data
Link to Albumin Titration Curve
Link to Online Model Application
Return to Home Page
Introduction: The Albumin-Adjusted Anion Gap.
The Figge-Fencl Quantitative Physicochemical Model of Human Acid-Base Physiology (version 2.0) and
the Figge-Mydosh-Fencl Model [1] may be used to
develop tools for the detection of increased concentrations of ions in various pathologic states.
This is of particular importance for the detection of anions other than chloride, often unmeasured, in
acutely ill patients presenting with metabolic acidosis. The simplest tool for this purpose is
the classic Anion Gap [2]. Figge, Jabor, Kazda and Fencl described the albumin-adjusted Anion Gap [3,4].
This tool is a very simple, approximate adjustment that corrects the anion gap for fluctuations in albumin concentrations.
Classic Anion Gap with [ K+ ] or without [ K+ ].
The classic Anion Gap (AG) can be calculated with or without the inclusion of [ K+ ].
AGK = [ Na+ ] + [ K+ ] - [ Cl- ] - [ HCO3- ]
AG = [ Na+ ] - [ Cl- ] - [ HCO3- ]
All quantities are expressed in mEq / L.
The normal range for AGK and AG each depends upon the analytic methods employed to measure the component
electrolytes. When [ Na+ ] and [ K+ ] are measured with flame photometry
and [ Cl- ] is measured with colorimetry, the normal range for AGK is approximately 16
+/- 4 mEq / L; when [ K+ ] is not included, the normal range for AG is approximately 12 +/- 4 mEq / L
[2]. For example, in nine normal subjects, Figge and colleagues [3,4] found that AGK was 16 +/- 2
mEq / L and AG was 12 +/- 2 mEq / L. However, in most modern clinical applications, the electrolytes are measured
with ion-selective electrodes and the calculated anion gap is lower. Hence, the common reference range for
AGK is 7 to 17 mEq / L, and the reference range for AG is approximately 4 mEq / L lower, i.e., 3 to 13 mEq / L
[5]. The exact
reference range must be established for each clinical laboratory. For example, Winter and colleagues described
a reference range of 3 to 11 mEq / L for AG in one of their clinical laboratories [6].
An elevated AGK or AG usually signals the presence of increased concentrations of anions other than chloride.
Such anions are often unmeasured in routine clinical settings. However, this scanning tool is non-specific in that no
information is provided regarding the identity of the anion(s). Refer to the chapter by Morgan [5] for an excellent
discussion of diagnostic possibilities.
Thus, a prompt and thorough workup is indicated to elucidate the underlying
etiology of an elevated AGK or AG and/or suspected metabolic acidosis. Given the lack of sensitivity
and specificity of the anion gap, direct measurements of lactate, as well as other suspected anions (e.g.,
beta-hydroxybutyrate, salicylate, formate and phosphate) and toxic screens as clinically indicated, should be routinely
obtained in all critically ill patients [5,7].
A normal AGK or AG cannot be used to definitively rule out increased concentrations of unmeasured anions,
because the presence of such anions can be masked by other conditions such as hypoalbuminemia, the simultaneous presence
of unmeasured cations such as a positively charged paraprotein, or toxic concentrations of a cationic drug such as
lithium [2,3,5].
Albumin-Adjusted Anion Gap.
The albumin-adjusted anion gap was described by Figge and colleagues [3,4]. This tool provides an approximate
correction for fluctuations in albumin concentrations. Calculations using this tool are most reliable when
all measurements, including [ Albumin ], are derived contemporaneously from the same specimen [7]. In ICU patients,
use of an arterial blood sample is preferred [7]. The usual reference ranges for
AGK and AG are used to interpret the results. The parameters in the equation are identical when
based on the Figge-Fencl model version 2.0, or when based on the Figge-Mydosh-Fencl model [1]. The albumin-corrected
anion gap demonstrates increased sensitivity over the classic anion gap for the detection of lactate [7]; nevertheless,
direct measurements of lactate, as well as other suspected anions (e.g.,
beta-hydroxybutyrate, salicylate, formate and phosphate) and toxic screens as clinically indicated, should be
routinely obtained in all critically ill patients [5,7].
The Figge equations for the albumin-adjusted anion gap, with and without [ K+ ], are:
AGK,adjusted = AGK + 2.5 x (4.4 - [ Albumin ] )
AGadjusted = AG + 2.5 x ( 4.4 - [ Albumin ] )
where [ Albumin ] is in mg / dL, and AGK,adjusted, AGK, AGadjusted and AG are expressed in mEq / L.
Example.
Given the following data from a post-operative patient with multiple organ failure [4]:
[ Na+ ] = 137 mEq / L (via flame photometer)
[ K+ ] = 4.9 mEq / L (via flame photometer)
[ Cl- ] = 102 mEq / L (via chloride titrator)
[ Albumin ] = 0.6 mg / dL
pH = 7.369
pCO2 = 42 mmHg
[ HCO3- ] = 24 mEq / L
One can calculate:
AGK = 137 + 4.9 - 102 - 24 = 15.9 mEq / L
This value is within the normal range (16 +/- 4 mEq / L, given the assays employed for electrolyte measurement).
However, the patient has severe hypoalbuminemia. Calculating the albumin-adjusted anion gap yields:
AGK,adjusted = 15.9 + 2.5 x (4.4 - 0.6) = 25.4 mEq / L
This calculation reveals that the patient has an elevated level of undetermined anions that is masked by
hypoalbuminemia.
References.
1. Figge J, Mydosh T, Fencl V. Serum proteins and acid-base equilibria: a follow-up. J Lab Clin Med. 1992;
120: 713-719.[
Abstract on PubMed ].
2. Emmett M, Narins RG: Clinical use of the anion gap. Medicine (Baltimore). 1977;
56:38-54. [Abstract on PubMed ].
3. Figge J, Jabor A, Kazda A, and Fencl V. Anion gap and hypoalbuminemia. Critical Care Medicine. 1998;
26:1807-1810. [
Abstract on PubMed ].
4. Fencl V, Jabor A, Kazda A, Figge J. Diagnosis of metabolic acid-base disturbances
in critically ill patients. Am J Respir Crit Care Med. 2000; 162: 2246-2251.
[ Full text is available online at
http://ajrccm.atsjournals.org/cgi/content/full/162/6/2246 ].
[ See also the online supplement, which is accessible at
http://ajrccm.atsjournals.org/cgi/content/full/162/6/2246/DC1 ].
5. Morgan TJ. Unmeasured ions and the strong ion gap. In Stewart's Textbook of Acid-Base.
Kellum JA and Elbers PWG, editors. Amsterdam: AcidBase.org. 2009. Chapter 18, pages 323 - 337.
6. Winter SD, Pearson JR, Gabow PA, Schultz AL, Lepoff RB. The fall of the serum anion gap.
Arch Intern Med. 1990 Feb;150(2):311-313. [
Abstract on PubMed ].
7. Chawla LS, Shih S, Davison D, Junker C, and Seneff MG.
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.
J Intensive Care Med. 2008 Mar-Apr;23(2):122-127.
[ Full text is available online at
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC2644323/pdf/1471-227X-8-18.pdf ].