AcidoBaseDrKellum PDF

253
Available online http://ccforum.com/content/8/4/253

Introduction
What is an acid? The first step to understand the evolution of
ideas in acidbase physiology since the beginning of the
twentieth century is to examine the definitions of an acid. There
are many current definitions of an acid used in chemistry [13].
The word acid is derived from the Latin word acidus [2,3],
meaning sour. For most of history sourness has been the
defining feature of acids as well as the method for detecting
the presence of acids. Other definitions of acids include the
ability to produce colour changes in litmus and to negate the
effects of an alkali. Several solution-based definitions have
been described since the late nineteenth century.
Arrhenius [4,5] developed a definition in the 1880s that, in its
generalized form, defines an acid as a substance that, when
dissolved in water, produces an increased concentration of
hydrogen ions [2]. Arrhenius also used a more specific
definition of acids as hydrogen salts [6]. By 1900, Naunyn [5]
and other workers [7] had adopted an acid definition that
appears to combine the generalized Arrhenius definition with
Faradays earlier description of anions such as chloride as
acid forming and of metal cations such as sodium as base
forming. Naunyn proposed that the acidbase status was
partly determined by electrolytes, particularly sodium and
chloride. This definition was embraced around 1920 by Van
Slyke [8] and is now known as the Van Slyke definition [9].
After World War I, Bronsted and Lowry simultaneously, but
separately, developed an identical definition for acids. Under
the BronstedLowry definition an acid is a substance that
could donate a proton (a hydrogen ion) [1,2]. An acid HA will
donate a proton to the solution when it dissociates into a
hydrogen ion and the conjugate anion A
:
HA H
+
+ A
Lewis developed a further definition in the 1920s, to

embrace a wider range of chemical scenarios. Lewis defined
an acid as a substance that can accept a pair of electrons to
form a covalent bond [1,2]. Chemicals such as boron
trifluoride are Lewis acids. Organic chemists often use this
definition [2].
The validity of a given acid definition [1,2] depends on the
given situation, be it cooking (sourness) or organic chemistry
(Lewis) [2]. In biological solutions such as plasma, the given
situation is a water-based solution with tightly controlled
solute concentrations. Both the Van Slyke and Bronsted
Lowry definitions (as well as those of Arrhenius and Lewis)
are valid for plasma [9] because water can supply hydrogen
and hydroxyl ions:
H
2
O H
+
+ OH
Review
Bench-to-bedside review: A brief history of clinical acidbase
David A Story
Associate Professor, The University of Melbourne, Austin Health, Melbourne, Victoria, Australia
Corresponding author: David A Story, David.Story@austin.org.au
Published online: 30 April 2004 Critical Care 2004, 8:253-258 (DOI 10.1186/cc2861)
This article is online at http://ccforum.com/content/8/4/253
2004 BioMed Central Ltd
Abstract
The history of assessing the acidbase equilibrium and associated disorders is intertwined with the
evolution of the definition of an acid. In the 1950s clinical chemists combined the Henderson
Hasselbalch equation and the BronstedLowry definition of an acid to produce the current bicarbonate
ion-centred approach to metabolic acidbase disorders. Stewart repackaged pre-1950 ideas of
acidbase in the late 1970s, including the Van Slyke definition of an acid. Stewart also used laws of
physical chemistry to produce a new acidbase approach. This approach, using the strong ion
difference (particularly the sodium chloride difference) and the concentration of weak acids (particularly
albumin), pushes bicarbonate into a minor role as an acidbase indicator rather than as an important
mechanism. The Stewart approach may offer new insights into acidbase disorders and therapies.
Keywords acidbase equilibrium, acids, bicarbonate ions, sodium chloride
254
Critical Care August 2004 Vol 8 No 4 Story
The Arrhenius, BronstedLowry, and Lewis definitions are all
currently used in chemistry and provide increasing generaliza-
bility when moving from Arrhenius to BronstedLowry to
Lewis [2,9]. The BronstedLowry definition has been the
most popular among acidbase physiologists since about
1955 [5,9]. However, some of the most important develop-
ments in the evolution of acidbase physiology occurred
before the publication of the BronstedLowry theory in 1923.
Before 1923 the Van Slyke definition (although not so-called
at the time) was the leading definition of an acid for
physiologists [9]. These physiologists included Henderson in
1908 [10] and Hasselbalch in 1916 [4] while developing the
HendersonHasselbalch equation (Fig. 1).
The original HendersonHasselbalch equation mathematically
links the variables of pH, partial pressure of carbon dioxide
(carbonic acid), and bicarbonate concentration (Fig. 1) [11].
This equation relates pH with the ratio of the concentration of
undissociated acid HA to the concentration of the conjugate
anion A
[9,12]. However, all weak acids in a given solution

can be inserted into HendersonHasselbalch-type equations
to calculate the pH (Fig. 2).
The second dissociation of phosphate can be used in plasma
in a similar way to bicarbonate as Henderson showed in
1908 [10]. The reason for this phenomenon is that for a
single solution of several weak acids, all the weak acids are in
equilibrium with a single pool of hydrogen ions; the isohydric
principle (Fig. 2) [11]. In physiology, the importance of the
isohydric principle is that, while the ratio of carbon dioxide to
bicarbonate can describe the acidbase status, it is not
necessarily the primary underlying mechanism for both the
respiratory and metabolic components [12].
The rise of bicarbonate
The Van Slyke definition of an acid [9] was the dominant
approach until about 1955 [1315]. The shift in thinking in
the mid-1950s appears to have followed a desire for clinical
chemistry to embrace the modern BronstedLowry definition
of an acid [15]. Using the BronstedLowry definition, many
physiologists confined their thinking on control of hydrogen
ions to weak acids and their conjugate anions, particularly
bicarbonate ions.
Proponents of the BronstedLowry approach [7,15,16]
downplayed that the Van Slyke definition was, in parallel with
the BronstedLowry definition, valid for aqueous biological
solutions [9]. The proponents dismissed defining chloride as
an acid and sodium as a base because these definitions were
old-fashioned and confusing [7,15,16]. The proponents felt
that the older approach did not pay due recognition to the
central, direct role of hydrogen ions. They felt that there was
an insufficient link between electrolytes such as sodium and
chloride and subsequent changes in hydrogen ions. None,
however, examined the role of water as a hydrogen ion
source, as Stewart later did [17].
The proponents of the new BronstedLowry-related
approach could not prove, however, that the previous
approach was wrong. Subsequently, while accepting only
BronstedLowry acids while looking for factors controlling
the nonrespiratory component of acidbase physiology (and
looking for simplicity in a time before calculators), many
researchers focused on the plasma bicarbonate concentration
and the HendersonHasselbalch equation (Fig. 1) [4]. This
was the beginning of the still dominant concept that plasma
bicarbonate is not only an indicator of acidbase status, but
also a principal determinant [11]. The issue of which came
first, a change in electrolytes or a change in bicarbonate,
went to the heart of the debate in the 1950s [15]. It is now at
the heart of the debate over the Stewart approach to
acidbase physiology almost half a century later [18,19].
All have agreed for decades that changes in the partial
pressure of carbon dioxide directly lead to changes in a
patients acidbase status [8]; either respiratory acidosis or
respiratory alkalosis. Debate has, however, centred on the
nonrespiratory (metabolic) component of a patients
acidbase status and the role of bicarbonate [1820]. By the
1930s it was recognized that an increase in the partial
pressure of carbon dioxide also led physiological (rather than
physical chemical) mechanisms to increase the plasma
bicarbonate concentration [8].
More than 20 years later, at the beginning of the polio
epidemic, Danish physicians used the plasma bicarbonate
Figure 1
The HendersonHasselbalch equation. pH, plasma pH; pKa, negative
log to base 10 of the apparent, overall dissociation constant of
carbonic acid; [HCO
3
], plasma bicarbonate concentration; ,

solubility of carbon dioxide in blood at 37C; pCO
2
, partial pressure of
carbon dioxide in blood.
pH = pKa + log
10
[HCO
3
]
pCO
2
Figure 2
The isohydric principal expressed in (a) the law of mass action form
and (b) the HendersonHasselbalch form. Because all weak acids in a
solution are in equilibrium with a single pool of hydrogen ions, the ratio
of any of the conjugate anion and its undissociated acid will be able to
describe the pH.
(a) H
2
CO
3
HCO
3
+ H
+
+ HPO
4
2
H
2
PO
4
(b) pKa1 + log = pH = pKa2 + log

H
2
CO
3
HCO
3
H
2
PO
4
HPO
4
2
255
concentration (total carbon dioxide content) alone, and as a
consequence incorrectly diagnosed metabolic alkalosis rather
than respiratory acidosis [4]. These errors were quickly
detected and led Danish researchers to aggressively pursue a
clinically useful method to determine at least two of the three
components of the HendersonHasselbalch equation (the pH
and the partial pressure of carbon dioxide) to calculate the
third (the plasma bicarbonate concentration) [4].
Several groups searched for methods to better assess the
metabolic component of acidbase status [4]. Singer and
Hastings [14] introduced the buffer base in 1948 in an
attempt to identify acidbase changes independent of carbon
dioxide. The buffer base is the sum of weak acid (buffer)
anions in plasma including albumin anions and bicarbonate.
Singer and Hastings defined fixed acids as nonbuffer anions,
chloride being one. More than 10 years later, other workers
pursued bicarbonate-centred assessments of the metabolic
component. Base excess and bicarbonate rules of thumb
were developed in an attempt to isolate primary changes in
the intimately linked variables of carbon dioxide and
bicarbonate [12]. Both base excess and bicarbonate rules of
thumb used a bicarbonate-centred approach [4,5,9].
The great trans-Atlantic debate
Siggaard-Andersen, from Copenhagen, developed base
excess in the late 1950s after examining titrations of human
blood. Base excess can be defined as the amount of strong
acid (in mmol/l) that must be added to the blood sample to
return the sample to pH 7.40 after equilibration while
maintaining the partial pressure of carbon dioxide at
40 mmHg [21]. If blood has a pH of 7.40 and a partial
pressure of carbon dioxide of 40 mmHg, therefore, the base
excess will be 0 mmol/l. Siggaard-Andersen developed a
nomogram [9] to determine base excess in the clinical
setting. This nomogram has been mathematically transcribed
(the Van Slyke equation) to allow calculation by blood gas
machines [22].
Schwartz and Relman from Boston argued that deriving
plasma base excess from blood in vitro was inaccurate [23].
First, plasma in vivo is in continuity with interstitial fluid that
has less buffer capacity. Siggaard-Andersen dealt with this
argument by assuming a haemoglobin concentration of
50 g/l, thus reducing the apparent buffer capacity of the
blood in vitro. The subsequent base excess estimate is
known as standard base excess [20]. The second problem
was that in patients with chronic elevation of the partial
pressure of carbon dioxide, the base excess approach
diagnosed a coexisting alkalinizing metabolic process
decreasing the acidity. One approach to this problem was
modifying the base excess nomogram [9]. Another approach
was to develop a correction factor [24].
Disagreement between Americans and Danes over the
usefulness of base excess led to the Great Trans-Atlantic
AcidBase Debate [25]. Instead of base excess the
Americans offered six rules-of-thumb to correct changes in
the partial pressure of carbon dioxide or bicarbonate
concentration for changes in the other [11,20,26]. These
rules described the physiological compensation to acidbase
changes to optimize acidbase homeostasis. Having allowed
for expected physiological compensation, residual changes in
carbon dioxide or bicarbonate are then seen as the
mechanisms for changes in acidbase status. Reflecting the
strength of the debate, some current texts [11,26] do not
mention base excess despite its apparent advantage of
simplicity [20].
Stewart
In the late 1970s and early 1980s, Peter Stewart proposed
that the generalized Arrhenius definition of an acid, with
Naunyns ideas, is more useful to acidbase physiology than
the BronstedLowry definition [17,27,28].
Stewart introduced an approach to acidbase physiology
and disorders with elements of previous ideas but packaged
in a new way [17]. Stewarts main reason for exploring
acidbase physiology was that he found the bicarbonate-
centred approach confusing and inadequate.
Using several principles of physical chemistry (particularly
elctroneutrality, conservation of mass, and dissociation of
electrolytes), Stewart produced an approach to acidbase
physiology with a strong relationship to the approaches of
Van Slyke [8] and of Singer and Hastings [14]. Stewarts
model has three independent controlling variables: the partial
pressure of carbon dioxide, the strong ion difference, and the
total weak-acid concentration [17,28]. The concentrations of
bicarbonate and hydrogen ions are dependent on these three
factors, in association with the (temperature-dependent)
dissociation constants of the weak acids and water (Fig. 3).
The two most important strong ions (completely dissociated
ions) in plasma are sodium and chloride [17,28,29]. The most
important weak acid (partly dissociated acid) is albumin, with
a minor effect from phosphate [29]. Stewart felt that the
major use for bicarbonate and base excess was to determine
the extent of a clinical acidbase disorder rather than the
mechanism [17].
Potential clinical implications
The Stewart approach appears to provide more straight-
forward explanations than the bicarbonate-centred approaches
for many acidbase phenomena seen in the critical care
setting [30,31]. This includes explanations for metabolic
alkalosis associated with decreased plasma albumin concen-
trations [32,33], the mechanism of hyperchloremic acidosis
[34], and the role of ammonia in acidbase homeostasis [30].
The Stewart approach has refined detecting unmeasured
ions. Figge and colleagues [35] demonstrated that the
256
traditional calculation of the anion gap does not allow for the
large changes in plasma albumin concentration often seen in
critically ill patients. Subsequently, unless a correction factor
is used, the true incidence of an increased anion gap may go
unrecognized [31,36]. The strong ion gap [37] uses
Stewarts approach to develop a more complete picture of
the anion gap.
Another approach to detecting unmeasured ions is to
examine the base excess effects of the sodium chloride
strong ion difference and the albumin weak acid effect [29].
These effects are then subtracted from the standard base
excess to give the base excess effect of unmeasured ions.
This approach combines the clinical utility of using base
excess with Stewarts insights to underling mechanisms
[38,39]. The bicarbonate rules of thumb are not as easily
amenable to this kind of quantitative analysis.
The Stewart approach may provide a better understanding of
not only the mechanisms of acidbase disorders, but also the
various management strategies including fluid management
[34,40,41], buffer therapy [42], and renal replacement
therapy [43]. With time, the Stewart approach is being
refined [44,45]. Stewarts work, like the great trans-Atlantic
debate, has had its detractors [18]. There is currently no
clear strategy to determine which of the modern approaches,
the Stewart approach [17] or the bicarbonate-centred
approach [16], is the correct one; however, sodium chloride
dilution studies may be one worthwhile area for study [46].
Measuring the unmeasured ions in critically ill patients is
another area of ongoing interest.
A clinical example
An example will allow a review of this acidbase history and
some of the implications for bedside work. A patient returned
to our intensive care unit after a complex liver transplant.
Blood gasses and arterial electrolytes were taken on arrival.
The blood gas results for Siggaard-Andersens approach
were a pH of 7.19, a partial pressure of carbon dioxide of
48 mmHg, and a base excess of 10.1 mmol/l. From this we
may conclude that there is a marked acidemia due to a
respiratory acidosis and a (quantified) metabolic acidosis.
The bicarbonate level was 18 mmol/l. If we use the rules of
thumb, again there is a respiratory acidosis; the partial
pressure of carbon dioxide has increased by 8 mmHg from
40 mmHg, which is almost 10 mmHg. If there were compensa-
tion we would expect the bicarbonate level to increase by
about 1 mmol/l and the expected bicarbonate would be
25 mmol/l [11]. The actual bicarbonate measured was
18 mmol/l; we therefore conclude there is a (unquantified)
metabolic acidosis.
The anion gap assists both the base excess and rules of
thumb approach to assess the source of the acidosis. The
other anion gap variables were: sodium, 145 mmol/l; potassium,
4.5 mmol/l; and chloride, 111 mmol/l. On first inspection the
increased chloride suggests the possibility of an unquantified
hyperchloremic metabolic acidosis. The calculated anion gap
was 20.5 mmol/l, suggesting a possible role for unmeasured
anions. However, the plasma albumin concentration was only
10 g/l which is likely to mask the true size of the anion gap.
Using Figge and colleagues correction [35] the anion gap
becomes 28.5 mmol/l.
The actual anion gap is therefore considerably larger than the
uncorrected anion gap. One component of this gap will be
the lactate of 3.7 mmol/l. Therefore, using either of the bi-
carbonate-centred approaches, we conclude that bicarbonate
has decreased in part through increased lactic acidosis,
through hyperchloremic acidosis, and through other unknown
acids. The relative contributions of these variables to the
acidosis remain unquantified.
Many clinicians using the Stewart approach would integrate
the Siggaard-Anderson approach and conclude that there is
a respiratory acidosis and a quantified metabolic acidosis of
10 mmol/l. The difference between the principal plasma
strong ions, sodium and chloride, is 34 mmol/l, which has an
acidifying base excess effect of 4 mmol/l assuming the
reference value is 38 mmol/l [38]. This acidosis is offset by an
alkalinizing albumin base excess effect of 8 mmol/l assuming
a normal albumin value of 42 g/l [38]. This leaves an
unmeasured ion effect on base excess of 14.5 mmol/l.
Lactate, another strong anion, will have a base excess effect
of 3.7 mmol/l. Phosphate is a weak acid. The plasma
phosphate concentration was 1.7 mmol/l and will have a base
excess effect of 3.1 mmol/l [30]. Confirming an important
effect of unmeasured ions, the strong ion gap [37] was
8.6 mEq/l given that the plasma magnesium concentration
was 0.57 mmol/l and the plasma ionized calcium
concentration was 1.17 mmol/l.
If one chose to treat the acidemia, the respiratory acidosis
can be dealt with through greater ventilation. On the meta-
bolic side, there is a decreased strong ion difference acidosis.
The strong ion difference can be widened (alkalizing) by
maintaining a high normal sodium, possibly by using sodium
bicarbonate, while decreasing the chloride through careful
Figure 3
Important factors in the control of hydrogen and bicarbonate ions using
the Stewart approach.
H
+
/ HCO
3
Strong-ion-difference

Weak acids
pCO
2
Dissociation
constant of water
and weak acids
257
use of intravenous fluids [34] and possibly furosemide to
increase chloride excretion [47]. If we chose to, we could
limit further acidosis by limiting the use of albumin, a weak
acid; particularly if the supporting solution is sodium chloride.
Alkalosis will occur as the liver removes lactate from the
plasma as a direct effect [30], not because of bicarbonate
formation [48].
In the present patient, unmeasured anions included gelatin as
a weak acid, from intravenous colloid therapy [49], as well as
acetate and gluconate, strong anions, from Plasmalyte [34].
Removal of these substances from the plasma by the kidney
or the liver will be directly alkalizing because less gelatin will
decrease the amount of weak acid in plasma, and less
acetate and gluconate will widen the strong ion difference.
The Stewart approach closely integrates acidbase physio-
logy and clinical chemistry, providing detailed clinical strategies
to manage acidbase disorders. The bicarbonate-centred
strategies are less integrated with general plasma chemistry
and appear less able to pinpoint specific components of
acidbase disorders.
Competing interests
None declared.
References
1. Oxford Dictionary of Chemistry, 3rd edition. Oxford: Oxford
University Press; 1996.
2. Dorlands Illustrated Medical Dictionary, 30th edition. Philadel-
phia, PA: Saunders; 2003.
3. Shorter Oxford Dictionary. Oxford: Oxford University Press; 1992.
4. Astrup P, Severinghaus JW: The History of Blood Gasses, Acids
and Bases. Copenhagen: Munksgaard; 1986.
5. Kassirer J: Historical perspective. In AcidBase. Edited by Cohen
J, Kassirer J. Boston, MA: Little, Brown and Co; 1982:449-464.
6. Arrhenius S: Theories of Solutions. New Haven, CT: Yale Univer-
sity Press; 1913.
7. Relman A: What are acids and bases? Am J Med 1954, 17:
435-437.
8. Peters J, Van Slyke D: Quantitive Clinical Chemistry. Baltimore,
MD: Williams and Wilkins; 1931.
9. Siggaard-Anderson O: The AcidBase Status of the Blood, 4th
edition. Copenhagen: Munksgaard; 1974.
10. Henderson LJ: The theory of neutrality regulation in the animal
organism. Am J Physiol 1908, 21:427-448.
11. Abelow B: Understanding AcidBase. Baltimore, MD: Williams
and Wilkins; 1998:52-54.
12. Fencl V, Leith DE: Stewarts quantitative acidbase chemistry:
applications in biology and medicine. Respir Physiol 1993, 91:
1-16.
13. Gamble J: Chemical Anatomy, Physiology and Pathology of
Extracellular Fluid. A Lecture Syllabus. Cambridge, MA: Harvard
University Press; 1954.
14. Singer RB, Hastings AB: An improved clinical method for the
estimation of disturbances of the acidbase balance of
human blood. Medicine 1948, 27:223-224.
15. Christensen H: Anions versus cations. Am J Med 1957, 27:163-
165.
16. Frazer S, Stewart C: Acidosis and alkalosis: a modern view.
J Clin Pathol 1959, 12:195-206.
17. Stewart PA: Modern quantitative acidbase chemistry. Can J
Physiol Pharmacol 1983, 61:1444-1461.
18. Siggaard-Andersen O, Fogh-Andersen N: Base excess or buffer
base (strong ion difference) as a measure of a non-respira-
tory acidbase disturbance. Acta Anesthesiol Scand 1995, 39:
123-128.
19. Sirker AA, Rhodes A, Grounds RM, Bennett ED: Acidbase
physiology: the traditional and the modern approaches.
Anaesthesia 2002, 57:348-356.
20. Severinghaus JW: Siggaard-Andersen and the Great Trans-
Atlantic AcidBase Debate. Scand J Clin Lab Invest Suppl
1993, 214:99-104.
21. Kofstad J: Base excess: a historical reviewhas the calcula-
tion of base excess been more standardised the last 20
years? Clin Chim Acta 2001, 307:193-195.
22. Siggaard-Andersen O: The van Slyke equation. Scand J Clin
Lab Invest Suppl 1977, 37:15-20.
23. Schwartz WB, Relman AS: A critique of the parameters used in
the evaluation of acidbase disorders. N Engl J Med 1963,
268:1383-1388.
24. Schlichtig R, Grogono AW, Severinghaus JW: Human PaCO
2
and standard base excess compensation for acidbase
imbalance. Crit Care Med 1998, 26:1173-1179.
25. Bunker JP: The Great Trans-Atlantic AcidBase Debate. Anes-
thesiology 1965, 26:591-593.
26. DuBose TD: Acidosis and alkalosis. In Harrisons Principles of
Internal Medicine, 15th edition, volume 1. Edited by Braunwald E,
Fauci AS, Kasper DL, Hauser SL, Longo DL, Jameson JL. Philadel-
phia, PA: McGraw-Hill; 2001:283-291.
27. Stewart PA: Independent and dependent variables of
acidbase control. Respir Physiol 1978, 33:9-26.
28. Stewart PA: How to Understand AcidBase. New York: Elsevier;
1981.
29. Gilfix BM, Bique M, Magder S: A physical chemical approach to
the analysis of acidbase balance in the clinical setting. J Crit
Care 1993, 8:187-197.
30. Kellum JA: Determinants of blood pH in health and disease.
Crit Care 2000, 4:6-14.
31. Fencl V, Jabor A, Kazda A, Figge J: Diagnosis of metabolic
acidbase disturbances in critically ill patients. Am J Respir
Crit Care Med 2000, 162:2246-2251.
32. Figge J, Mydosh T, Fencl V: Serum proteins and acidbase
equilibria: a follow-up. J Lab Clin Med 1992, 120:713-719.
33. Wilkes P: Hypoproteinemia, strong-ion difference, and acid
base status in critically ill patients. J Appl Physiol 1998, 84:
1740-1748.
34. Liskaser FJ, Bellomo R, Hayhoe M, Story D, Poustie S, Smith B,
Letis A, Bennett M: Role of pump prime in the etiology and
pathogenesis of cardiopulmonary bypass-associated acido-
sis. Anesthesiology 2000, 93:1170-1173.
35. Figge J, Jabor A, Kazda A, Fencl V: Anion gap and hypo-
albuminemia. Crit Care Med 1998, 26:1807-1810.
36. Story DA, Poustie S, Bellomo R: Estimating unmeasured anions
in critically ill patients: anion-gap, base-deficit, and strong-ion-
gap. Anaesthesia 2002, 57:1109-1114.
37. Kellum JA, Kramer DJ, Pinsky MR: Strong ion gap: a method-
ology for exploring unexplained anions. J Crit Care 1995,
10:51-55.
38. Story DA, Morimatsu H, Bellomo R: Strong ions, weak acids,
and base excess: a simplified FenclStewart approach to
clinical acidbase disorders. Br J Anaesth 2004, 92:1-7.
39. Boyle M, Lawrence J: An easy method of mentally estimating the
metabolic component of acid/base balance using the Fencl
Stewart approach. Anaesth Intensive Care 2003, 31:538-547.
40. Scheingraber S, Rehm M, Sehmisch C, Finsterer U: Rapid saline
infusion produces hyperchloremic acidosis in patients under-
going gynecologic surgery. Anesthesiology 1999, 90:1265-
1270.
41. Constable PD: Hyperchloremic acidosis: the classic example
of strong ion acidosis. Anesth Analg 2003, 96:919-922.
42. Rehm M, Finsterer U: Treating intraoperative hyperchloremic
acidosis with sodium bicarbonate or tris-hydroxymethyl
aminomethane: a randomized prospective study. Anesth
Analg 2003, 96:1201-1208.
43. Rocktaschel J, Morimatsu H, Uchino S, Ronco C, Bellomo R:
Impact of continuous veno-venous hemofiltration on acid
base balance. Int J Artif Organs 2003, 26:19-25.
44. Staempfli HR, Constable PD: Experimental determination of
net protein charge and A(tot) and K(a) of nonvolatile buffers
in human plasma. J Appl Physiol 2003, 95:620-630.
45. Wooten EW: Calculation of physiological acidbase parame-
ters in multicompartment systems with application to human
blood. J Appl Physiol 2003, 95:2333-2344.
258
46. Constable PD: Stewart approach is not always a practical clini-
cal tool response. Anesth Analg 2004, 98:271-272.
47. Stoelting RK: Pharmacology and Physiology in Anaesthetic Prac-
tice. Philadelphia, PA: Lippincott-Raven; 1999.
48. White SA, Goldhill DR: Is Hartmanns the solution? Anaesthesia
1997, 52:422-427.
49. Hayhoe M, Bellomo R, Liu G, McNicol L, Buxton B: The aetiology
and pathogenesis of cardiopulmonary bypass-associated
metabolic acidosis using polygeline pump prime. Intensive
Care Med 1999, 25:680-685.
184
AG = anion gap; [A
TOT
] = total concentration of weak acids; BE = base excess; PCO
2
= partial CO
2
difference; SCO
2
= CO
2
solubility; SID
+
=
strong ion difference; SIG = strong ion gap.
Critical Care April 2005 Vol 9 No 2 Corey
Abstract
Complex acidbase disorders arise frequently in critically ill
patients, especially in those with multiorgan failure. In order to
diagnose and treat these disorders better, some intensivists have
abandoned traditional theories in favor of revisionist models of
acidbase balance. With claimed superiority over the traditional
approach, the new methods have rekindled debate over the
fundmental principles of acidbase physiology. In order to shed
light on this controversy, we review the derivation and application
of new models of acidbase balance.
Introduction: Master equations
All modern theories of acidbase balance in plasma are
predicated upon thermodynamic equilibrium equations. In an
equilibrium theory, one enumerates some property of a
system (such as electrical charge, proton number, or proton
acceptor sites) and then distributes that property among the
various species of the system according to the energetics of
that particular system. For example, human plasma consists
of fully dissociated ions (strong ions such as Na
+
, K
+
, Cl
and lactate), partially dissociated weak acids (such as

albumin and phosphate), and volatile buffers (carbonate
species). C
B
, the total concentration of proton acceptor sites
in solution, is given by
C
B
= C +
i
C
i
e
i
D (1)
Where C is the total concentration of carbonate species
proton acceptor sites (in mmol/l), C
i
is the concentration of
noncarbonate buffer species i (in mmol/l), e
i
is the average
number of proton acceptor sites per molecule of species i,
and D is Riccis difference function (D = [H
+
] [OH
]).
Equation 1 may be regarded as a master equation from which
all other acidbase formulae may be derived [1].
Assuming that [CO
3
2
] is small, Eqn 1 may be re-expressed:
C
B
= [HCO
3
] +
i
C
i
e
i
(2)
Similarly, the distribution of electrical charge may be
expressed as follows:
SID
+
= C
i
C
i
Z
i
(3)
Where SID
+
is the strong ion difference and Z
i
is the
average charge per molecule of species i.
The solution(s) to these master equations require rigorous
mathematical modeling of complex protein structures.
Traditionally, the mathematical complexity of master Eqn 2
has been avoided by setting C
i
= 0, so that C
B
=
[HCO
3
]. The study of acidbase balance now becomes

appreciably easier, simplifying essentially to the study of
volatile buffer equilibria.
Stewart equations
Stewart, a Canadian physiologist, held that this
simplification is not only unnecessary but also potentially
misleading [2,3]. In 1981, he proposed a novel theory of
acidbase balance based principally on an explicit
restatement of master Eqn 3:
Bicarbonate ion formation equilibrium:
[H
+
] [HCO
3
] = K
1
S PCO
2
(4)
Where K
1
is the apparent equilibrium constant for the
HendersonHasselbalch equation and S is the solubility of
CO
2
in plasma.
Review
Bench-to-bedside review: Fundamental principles of acid-base
physiology
Howard E Corey
Director, The Childrens Kidney Center of New Jersey, Atlantic Health System, Morristown, New Jersey, USA
Corresponding author: Howard E Corey, howard.corey@ahsys.org
Published online: 29 November 2004 Critical Care 2005, 9:184-192 (DOI 10.1186/cc2985)
185
Carbonate ion formation equilibrium:
[H
+
] [CO
3
2
] = K
3
[HCO
3
] (5)
Where K
3
is the apparent equilibrium dissociation constant
for bicarbonate.
Water dissociation equilibrium:
[H
+
] [OH
] = K
w
(6)
Where K
w
is the autoionization constant for water.
Electrical charge equation:
[SID
+
] = [HCO
3
] + [A
] + [CO
3
2
] + [OH
] [H
+
] (7)
Where [SID
+
] is the difference in strong ions ([Na
+
] + [K
+
]
[Cl
] [lactate
]) and [A
] is the concentration of dissociated

weak acids, mostly albumin and phosphate.
Weak acid dissociation equilibrium:
[H
+
] [A
] = K
a
[HA] (8)
Where K
a
is the weak acid dissociation constant for HA.
In addition to these five equations based principally on the
conservation of electrical charge, Stewart included one
additional equation.
Conservation of mass for A:
[A
TOT
] = [HA] + [A
] (9)
Where [A
TOT
] is the total concentration of weak acids.
Accordingly, [H
+
] may be determined only if the constraints of
all six of the equations are satisfied simultaneously [2,3].
Combining equations, we obtain:
a[H
+
]
4
+ b[H
+
]
3
+ c[H
+
]
2
+ d[H
+
] + e = 0 (10)
Where a = 1; b = [SID
+
] + K
a
; c = {K
a
([SID
+
] [A
TOT
])
K
w
K
1
S PCO
2
}; d = {K
a
(K
w
+ K
1
S PCO
2
)
K
3
K
1
S PCO
2
}; and e = K
a
K
3
K
1
S PCO
2
.
If we ignore the contribution of the smaller terms in the
electrical charge equation (Eqn 7), then Eqn 10 simplifies to
become [4]:
pH = pK
1
+ log
[SID
+
] K
a
[A
TOT
]/K
a
+ 10
pH
(11)
S PCO
2
In traditional acidbase physiology, [A
TOT
] is set equal to 0
and Eqn 11 is reduced to the well-known Henderson
Hasselbalch equation [5,6]. If this simplification were valid,
then the plot of pH versus log PCO
2
(the buffer curve) would
be linear, with an intercept equal to log [HCO
3
]/K
1
SCO
2
[7,8]. In fact, experimental data cannot be fitted to a linear
buffer curve [4]. As indicated by Eqn 11, the plot of pH
versus log PCO
2
is displaced by changes in protein
concentration or the addition of Na
+
or Cl
, and becomes
nonlinear in markedly acid plasma (Fig. 1). These observa-
tions suggest that the HendersonHasselbalch equation may
be viewed as a limiting case of the more general Stewart
equation. When [A
TOT
] varies, the simplifications of the
traditional acidbase model may be unwarranted [9].
The Stewart variables
The Stewart equation (Eqn 10) is a fourth-order polynomial
equation that relates [H
+
] to three independent variables
([SID
+
], [A
TOT
] and PCO
2
) and five rate constants (K
a
, K
w
, K
1
,
K
3
and SCO
2
), which in turn depend on temperature and ion
activities (Fig. 2) [2,3].
Strong ion difference
The first of these three variables, [SID
+
], can best be
appreciated by referring to a Gamblegram (Fig. 3). The
apparent strong ion difference, [SID
+
]
a
, is given by the
following equation:
[SID
+
]
a
= [Na
+
] + [K
+
] [Cl
] [lactate]
[other strong anions] (12)
In normal plasma, [SID
+
]
a
is equal to [SID
+
]
e
, the effective
strong ion difference:
[SID
+
]
e
= [HCO
3
] + [A
] (13)
Where [A
] is the concentration of dissociated weak

noncarbonic acids, principally albumin and phosphate.
Strong ion gap
The strong ion gap (SIG), the difference between [SID
+
]
a
and
[SID
+
]
e
, may be taken as an estimate of unmeasured ions:
SIG = [SID
+
]
a
[SID
+
]
e
= AG [A
] (14)
Unlike the well-known anion gap (AG = [Na
+
] + [K
+
] [Cl
]
[HCO
3
]) [10], the SIG is normally equal to 0.

SIG may be a better indicator of unmeasured anions than the
AG. In plasma with low serum albumin, the SIG may be high
(reflecting unmeasured anions), even with a completely
normal AG. In this physiologic state, the alkalinizing effect of
hypoalbuminemia may mask the presence of unmeasured
anions [1118].
Weak acid buffers
Stewart defined the second variable, [A
TOT
], as the
composite concentration of the weak acid buffers having a
single dissociation constant (K
A
= 3.0 10
7
) and a net
maximal negative charge of 19 mEq/l [2,3]. Because Eqn 9
invokes the conservation of mass and not the conservation of
charge, Constable [19] computed [A
TOT
] in units of mass
186
(mmol/l) rather than in units of charge (mEq/l), and found that
[A
TOT
(mmol/l)] = 5.72 0.72 [albumin (g/dl)].
Although thermodynamic equilibrium equations are
independent of mechanism, Stewart asserted that his three
independent parameters ([SID
+
], [A
TOT
] and PCO
2
) determine
the only path by which changes in pH may arise (Fig. 4).
Furthermore, he claimed that [SID
+
], [A
TOT
] and PCO
2
are true
biologic variables that are regulated physiologically through
the processes of transepithelial transport, ventilation, and
metabolism (Fig. 5).
Base excess
In contrast to [SID
+
], the traditional parameter base excess
(BE; defined as the number of milliequivalents of acid or base
that are needed to titrate 1 l blood to pH 7.40 at 37C while
the PCO
2
is held constant at 40 mmHg) provides no further
insight into the underlying mechanism of acidbase
disturbances [20,21]. Although BE is equal to SID
+
when
nonvolatile buffers are held constant, BE is not equal to
SID
+
when nonvolatile acids vary. BE read from a standard
nomogram is then not only physiologically unrevealing but
also numerically inaccurate (Fig. 2) [1,9].
The Stewart theory: summary
The relative importance of each of the Stewart variables in the
overall regulation of pH can be appreciated by referring to a
spider plot (Fig. 6). pH varies markedly with small changes in
PCO
2
and [SID
+
]. However, pH is less affected by
perturbations in [A
TOT
] and the various rate constants [19].
Figure 1
The buffer curve. The line plots of linear in vitro (, , , ) and
curvilinear in vivo (dots) log PCO
2
versus pH relationship for plasma.
, plasma with a protein concentration of 13 g/dl (high [A
TOT
]);
, plasma with a high [SID
+
] of 50 mEq/l; , plasma with a normal
[A
TOT
] and [SID
+
]; , plasma with a low [SID
+
] of 25 mEq/l; dots,
curvilinear in vivo log PCO
2
versus pH relationship. [A
TOT
], total
concentration of weak acids; PCO
2
, partial CO
2
tension; SID
+
, strong
ion difference. Reproduced with permission from Constable [4].
Figure 2
Graph of independent variables (PCO
2
, [SID
+
] and [A
TOT
]) versus pH.
Published values were used for the rate constants K
a
, Kw, K
1
, K
3
, and
SCO
2
. Point A represents [SID
+
] = 45 mEq/l and [A
TOT
] = 20 mEq/l,
and point B represents [SID
+
] = 40 mEq/l and [A
TOT
] = 20 mEq/l. In
moving from point A to point B, SID
+
= AB = base excess. However,
if [A
TOT
] decreases from 20 to 10 mEq/l (point C), then AC SID
+
base excess. [A
TOT
], total concentration of weak acids; PCO
2
, partial
CO
2
tension; SCO
2
, CO
2
solubility; SID
+
, strong ion difference.
Reproduced with permission from Corey [9].
Figure 3
Gamblegram a graphical representation of the concentration of
plasma cations (mainly Na
+
and K
+
) and plasma anions (mainly Cl
,
HCO
3
and A
). SIG, strong ion gap (see text).

187
In summary, in exchange for mathematical complexity the
Stewart theory offers an explanation for anomalies in the
buffer curve, BE, and AG.
The FiggeFencl equations
Based on the conservation of mass rather than conservation
of charge, Stewarts [A
TOT
] is the composite concentration of
weak acid buffers, mainly albumin. However, albumin does
not exhibit the chemistry described by Eqn 9 within the range
of physiologic pH, and so a single, neutral [AH] does not
actually exist [22]. Rather, albumin is a complex poly-
ampholyte consisting of about 212 amino acids, each of
which has the potential to react with [H
+
].
From electrolyte solutions that contained albumin as the sole
protein moiety, Figge and coworkers [23,24] computed the
individual charges of each of albumins constituent amino
acid groups along with their individual pKa values. In the
FiggeFencl model, Stewarts [A
TOT
] term is replaced by
[Pi
x
] and [Pr
y
] (the contribution of phosphate and albumin to
charge balance, respectively), so that the four independent
variables of the model are [SID
+
], PCO
2
, [Pi
x
], and [Pr
y
].
Omitting the small terms
[SID
+
] [HCO
3
] [Pi
x
] [Pr
y
] = 0 (15)
Figure 5
The Stewart model. pH is regulated through manipulation of the three
Stewart variables: [SID
+
], [A
TOT
] and PCO
2
. These variables are in turn
upset, regulated, or modified by the gastrointestinal (GI) tract, the
liver, the kidneys, the tissue circulation, and the intracellular buffers.
[A
TOT
2
, partial CO
2
tension;
SID
+
Figure 6
Spider plot of the dependence of plasma pH on changes in the three
independent variables ([SID
+
], PCO
2
, and [A
TOT
]) and five rate
constants (solubility of CO
2
in plasma [S], apparent equilibrium
constant [K
1
], effective equilibrium dissociation constant [K
a
],
apparent equilibrium dissociation constant for HCO
3
[K
3
], and ion
product of water [K
w
]) of Stewarts strong ion model. The spider plot
is obtained by systematically varying one input variable while holding
the remaining input variables at their normal values for human plasma.
The influence of S and K
1
on plasma pH cannot be separated from
that of PCO
2
, inasmuch as the three factors always appear as one
expression. Large changes in two factors (K
3
and K
w
) do not change
plasma pH. [A
TOT
2
, partial
CO
2
tension; SID
+
, strong ion difference. Reproduced with
permission from Constable [19].
Figure 4
Stewarts independent variables ([SID
+
], [A
TOT
] and PCO
2
), along with
the water dissociation constant (K
w
), determine the dependent
variables [H
+
] and [HCO
3
]. When [A
TOT
] = 0, Stewarts model
simplifies to the well-known HendersonHasselbalch equation. [A
TOT
],
total concentration of weak acids; PCO
2
, partial CO
2
tension; SID
+
,
strong ion difference.
188
The FiggeFencl equation is as follows [25]:
SID
+
+ 1000 ([H
+
] Kw/[H
+
] Kc1 PCO
2
/
[H
+
] Kc1 Kc2 PCO
2
/[H
+
]
2
) [Pi
tot
] Z
+ {1/(1 + 10
[pH 8.5]
)
98/(1 + 10
[pH 4.0]
)
18/(1 + 10
[pH 10.9]
)
+ 24/(1 + 10
+[pH 12.5]
)
+ 6/(1 + 10
+[pH 7.8]
)
+ 53/(1 + 10
+[pH 10.0]
)
+ 1/(1 + 10
+[pH 7.12 + NB]
)
+ 1/(1 + 10
+[pH 7.22 + NB]
)
+ 1/(1 + 10
+[pH 7.10 + NB]
)
+ 1/(1 + 10
+[pH 7.49 + NB]
)
+ 1/(1 + 10
+[pH 7.01 + NB]
)
+ 1/(1 + 10
+[pH 7.31]
)
+ 1/(1 + 10
+[pH 6.75]
)
+ 1/(1 + 10
+[pH 6.36]
)
+ 1/(1 + 10
+[pH 4.85]
)
+ 1/(1 + 10
+[pH 5.76]
)
+ 1/(1 + 10
+[pH 6.17]
)
+ 1/(1 + 10
+[pH 6.73]
)
+ 1/(1 + 10
+[pH 5.82]
)
+ 1/(1 + 10
+[pH 6.70]
)
+ 1/(1 + 10
+[pH 4.85]
)
+ 1/(1 + 10
+[pH 6.00]
)
+ 1/(1 + 10
+[pH 8.0]
)
1/(1 + 10
[pH 3.1]
)} 1000 10 [Alb]/66500 = 0
(16)
Where [H
+
] = 10
pH
; Z = (K1 [H
+
]
2
+ 2 K1 K2 [H
+
] +
3 K1 K2 K3)/([H
+
]
3
+ K1 [H
+
]
2
+ K1 K2 [H
+
] +
K1 K2 K3); and NB = 0.4 (1 1/(1 + 10
[pH 6.9]
)).
The strong ion difference [SID
+
] is given in mEq/l, PCO
2
is
given in torr, the total concentration of inorganic phosphorus
containing species [Pi
tot
] is given in mmol/l and [Alb] is given
in g/dl. The various equilibrium constants are Kw = 4.4
10
14
(Eq/l)
2
; Kc1 = 2.46 10
11
(Eq/l)
2
/torr; Kc2 = 6.0
10
11
(Eq/l); K1 = 1.22 10
2
(mol/l); K2 = 2.19 10
7
(mol/l); and K3 = 1.66 10
12
(mol/l).
Watson [22] has provided a simple way to understand the
FiggeFencl equation. In the pH range 6.87.8, the pKa
values of about 178 of the amino acids are far from the
normal pH of 7.4. As a result, about 99 amino acids will have
a fixed negative charge (mainly aspartic acid and glutamic
acid) and about 79 amino acids will have a fixed positive
charge (mostly lysine and arginine), for a net fixed negative
charge of about 21 mEq/mol. In addition to the fixed charges,
albumin contains 16 histidine residues whose imidazole
groups may react with H
+
(variable charges).
The contribution of albumin to charge, [Pr
x
], can then be
determined as follows:
[Pr
x
] = 21 (16 [1
pH
]) 10,000/66,500
[albumin (g/dl)] (17)
Where 21 is the number of fixed negative charges/mol
albumin, 16 is the number of histidine residues/mol albumin,
and
pH
is the ratio of unprotonated to total histadine at a
given pH. Equation 17 yields identical results to the more
complex FiggeFencl analysis.
Linear approximations
In the linear approximation taken over the physiologic range of
pH, Eqn 16 becomes
[SID
+
]
e
=[HCO
3
] + [Pr
X
] + [Pi
Y
] (18)
Where [HCO
3
] = 1000 Kcl PCO

2
/(10
pH
); [Pr
X
] =
[albumin (g/dl)] (1.2 pH6.15) is the contribution of
albumin to charge balance; and [Pi
Y
] = [phosphate (mg/dl)]
(0.097 pH0.13) is contribution of phosphate to charge
balance [1,2325].
Combining equations yields the following:
SIG = AG [albumin (g/dl)] (1.2 pH6.15)
[phosphate (mg/dl)] (0.097 pH0.13) (19)
According to Eqn 18, when pH = 7.40 the AG increases by
roughly 2.5 mEq/l for every 1 g/dl decrease in [albumin].
Buffer value
The buffer value () of plasma, defined as = base/pH, is
equal to the slope of the line generated by plotting (from Eqn
18) [SID
+
]
e
versus pH [9]:
= 1.2 [albumin (g/dl)] + 0.097 [phosphate (mg/dl)]
(20)
When plasma is low, the pH is higher for any given BE
than when is normal.
The may be regarded as a central parameter that relates the
various components of the HendersonHasselbalch, Stewart
and FiggeFencl models together (Fig. 7). When non-
carbonate buffers are held constant:
BE = [SID
+
]
e
= [HCO
3
] + pH (21)
When non-carbonate buffers vary, BE = [SID
+
]
e
; that is,
[SID
+
]
a
referenced to the new weak buffer concentration.
The FiggeFencl equations: summary
In summary, the FiggeFencl model relates the traditional to
the Stewart parameters and provides equations that permit ,
[SID
+
]
e
, and SIG to be calculated from standard laboratory
measurements.
189
The Wooten equations
Acidbase disorders are usually analyzed in plasma.
However, it has long been recognized that the addition of
hemoglobin [Hgb], an intracellular buffer, to plasma causes a
shift in the buffer curve (Fig. 8) [26]. Therefore, BE is often
corrected for [Hgb] using a standard nomogram [20,21,27].
Wooten [28] developed a multicompartmental model that
corrects the FiggeFencl equations for [Hgb]:
= (1 Hct) 1.2 [albumin (g/dl)] + (1 Hct) 0.097
[phosphate (mg/dl)] + 1.58 [Hgb (g/dl)] + 4.2 (Hct) (22)
[SID
+
]
effective, blood
= (1 0.49 Hct)[HCO
3
] +
(1 Hct)(C
alb
[1.2 pH6.15] +C
phos
[0.097
pH0.13]) + C
Hgb
(1.58 pH11.4) + Hct (4.2 pH3.3)
(23)
With C
alb
and C
Hgb
expressed in g/dl and C
phos
in mg/dl.
In summary, the Wooten model brings Stewart theory to the
analysis of whole blood and quantitatively to the level of
titrated BE.
Application of new models of acidbase
balance
In order to facilitate the implementation of the Stewart
approach at the bedside, Watson [29] has developed a
computer program (AcidBasics II) with a graphical user
interface (Fig. 9). One may choose to use the original Stewart
or the FiggeFencl model, vary any of the rate constants, or
adjust the temperature. Following the input of the
independent variables, the program automatically displays all
of the independent variables, including pH, [HCO
3
] and [A
].
In addition, the program displays SIG, BE, and a
Gamblegram (for an example, see Fig. 3).
One may classify acidbased disorders according to
Stewarts three independent variables. Instead of four main
acidbase disorders (metabolic acidosis, metabolic alkalosis,
respiratory acidosis, and respiratory alkalosis), there are six
disorders based on consideration of PCO
2
, [SID
+
], and [A
TOT
]
(Table 1). Disease processes that may be diagnosed using
the Stewart approach are listed in Table 2.
Example
Normal plasma may be defined by the following values: pH =
7.40, PCO
2
= 40.0 torr, [HCO
3
] = 24.25 mmol/l, [albumin] =

4.4 g/dl, phosphate = 4.3 mg/dl, sodium = 140 mEq/l,
potassium = 4 mEq/l, and chloride = 105 mEq/l. The
corresponding values for traditional and Stewart acidbase
parameters are listed in Table 3.
Consider a hypothetical case 1 with pH = 7.30, PCO
2
=
30.0 torr, [HCO
3
] = 14.25 mmol/l, Na
2+
= 140 mEq/l, K
+
=
4 mEq/l, Cl
= 115 mEq/l, and BE = 10 mEq/l. The

traditional interpretation based on BE and AG is a normal
anion gap metabolic acidosis with respiratory compensation.
The Stewart interpretation based on [SID
+
]
e
and SIG is low
[SID
+
]
e
/normal SIG metabolic acidosis and respiratory
compensation. The Stewart approach corrects the BE read
from a nomogram for the 0.6 mEq/l acid load absorbed by
the noncarbonate buffers. In both models, the differential
diagnosis for the acidosis includes renal tubular acidosis,
diarrhea losses, pancreatic fluid losses, anion exchange
resins, and total parenteral nutrition (Tables 2 and 3).
Now consider a hypothetical case 2 with the same arterial
blood gas and chemistries but with [albumin] = 1.5 g/dl. The
Figure 8
The effect of hemoglobin (Hb) on the buffer curve: (left) in vitro and
(right) in vivo. PCO
2
, partial CO
2
tension. Reproduced with permission
from Davenport [26].
Figure 7
(a) The effective strong ion difference ([SID
+
]
e
; Eqn 18) can be
understood as a combination of [HCO
3
], the buffer value () and

constant terms. The [HCO
3
] parameter can be determined from the

(b) HendersonHasselbalch equation, whereas (d) the buffer value is
derived partly from the albumin data of Figge and Fencl (c). When
noncarbonate buffers are held constant, [SID
+
]
e
is equal to the base
excess (BE). (e) In physiologic states with a low , BE may be an
insensitive indicator of important acidbase processes. (f) The strong
ion gap (SIG), which quantifies unmeasured anions, can be
calculated from the anion gap (AG) and . In physiological states with
a low , unmeasured anions may be present (high SIG) even with a
normal AG.
190
traditional interpretation and differential diagnosis of the
disorder remains unchanged from case 1 because BE and
AG have not changed. However, the Stewart interpretation is
low [SID
+
]
e
/high SIG metabolic acidosis and respiratory
compensation. Because of the low , the pH is greater for
any given BE than in case 1. The Stewart approach corrects
BE read from a nomogram for the 0.2 mEq/l acid load
absorbed by the noncarbonate buffers. The differential
diagnosis for the acidosis includes ketoacidosis, lactic
acidosis, salicylate intoxication, formate intoxication, and
methanol ingestion (Tables 2 and 3).
Summary
All modern theories of acidbase balance are based on
physiochemical principles. As thermodynamic state equations
are independent of path, any convenient set of parameters
(not only the one[s] used by nature) may be used to describe
a physiochemical system. The traditional model of acidbase
balance in plasma is based on the distribution of proton
acceptor sites (Eqn 1), whereas the Stewart model is based
on the distribution of electrical charge (Eqn 2). Although
sophisticated and mathematically equivalent models may be
derived from either set of parameters, proponents of the
traditional or proton acceptor site approach have
advocated simple formulae whereas proponents of the
Stewart electrical charge method have emphasized
mathematical rigor.
The Stewart model examines the relationship between the
movement of ions across biologic membranes and the
consequent changes in pH. The Stewart equation relates
changes in pH to changes in three variables, [SID
+
], [A
TOT
]
and PCO
2
. These variables may define a biologic system and
so may be used to explain any acidbase derangement in
that system.
Figge and Fencl further refined the model by analyzing
explicitly each of the charged residues of albumin, the main
component of [A
TOT
]. Wooten extended these observations
to multiple compartments, permitting the consideration of
both extracellular and intracellular buffers.
In return for mathematical complexity, the Stewart model
corrects the traditional computations of buffer curve, BE,
and AG for nonvolative buffer concentration. This may be
important in critically ill, hypoproteinuric patients.
Conclusion
Critics note that nonvolatile buffers contribute relatively little
to BE and that a corrected AG (providing similar information
to the SIG) may be calculated without reference to Stewart
theory by adding about 2.5 (4.4 [albumin]) to the AG.
To counter these and other criticisms, future studies need to
demonstrate the following: the validity of Stewarts claim that
his unorthodox parameters are the sole determinants of pH in
plasma; the prognostic significance of the Stewart variables;
the superiority of the Stewart parameters for patient
management; and the concordance of the Stewart equations
Figure 9
AcidBasics II. With permission from Dr Watson.
Table 1
Classification of acidbase disorders
Stewart variables/constants Classification Acidosis Alkalosis
PCO
2
Respiratory
[SID
+
] Metabolic
Chloride excess/deficit
Strong ion gap
[A
TOT
]
a
Modulator
Extracellular
Albumin
Phosphate
Intracellular
b
Hgb
DPG
Rate constants Modulator
(K
a
, K
w
, K
1
, K
3
, and SCO
2
)
Temperature
c

a
Changes in [A
TOT
] modulate and do not necessarily cause acidbase
disorders.
b
Result in negligible changes in pH.
c
May be clinically
significant in hypothermia. [A
TOT
], total concentration of weak acids;
DPG, 2,3-diphosphoglycerate; Hgb, hemoglobin; PCO
2
, partial CO
2
tension; SCO
2
, CO
2
solubility; SID
+
191
with experimental data obtained from ion transporting
epithelia.
In the future, the Stewart model may be improved through a
better description of the electrostatic interaction of ions and
polyelectroles (PoissonBoltzman interactions). Such
interactions are likely to have an important effect on the
electrical charges of the nonvolatile buffers. For example, a
detailed analysis of the pH-dependent interaction of albumin
with lipids, hormones, drugs, and calcium may permit further
refinement of the FiggeFencl equation [25].
Perhaps most importantly, the Stewart theory has re-
awakened interest in quantitative acidbase chemistry and
has prompted a return to first principles of acidbase
physiology.
Competing interests
The author(s) declare that they have no competing interests.
Acknowledgments
I would like to acknowledge the helpful discussions I have had with
Dr E Wrenn Wooten and Dr P Watson during the preparation of the
manuscript.
References
1. Wooten EW: Analytic calculation of physiological acidbase
parameters. J Appl Physiol 1999, 86:326-334.
2. Stewart PA: How to understand acid base balance. In A Quan-
titative AcidBase Primer for Biology and Medicine. New York:
Elsevier; 1981.
3. Stewart PA: Modern quantitative acid-base chemistry. Can J
4. Constable PD: A simplified strong ion model for acid-base
equilibria: application to horse plasma. J Appl Physiol 1997,
83:297-311.
5. Hasselbalch KA, Gammeltoft A: The neutral regulation of the
gravid organism [in German]. Biochem Z 1915, 68:206.
6. Hasselbalch KA: The reduced and the regulated hydrogen
number of the blood [in German]. Biochem Z 1918, 174:56.
7. Van Slyke DD: Studies of acidosis: XVII. The normal and
abnormal variations in the acid base balance of the blood. J
Biol Chem 1921, 48:153.
8. Siggaard-Andersen 0: The pH-log PCO
2
blood acidbase
nomogram revised. Scand J Clin Lab Invest 1962, 14:598-604.
9. Corey HE: Stewart and beyond: new models of acid-base
balance. Kidney Int 2003, 64:777-787.
10. Oh MS, Carroll HJ: Current concepts: the anion gap. N Engl J
Med 1977, 297:814.
11. Constable PD: Clinical assessment of acidbase status.
Strong ion difference theory. Vet Clin North Am Food Anim
Pract 1999, 15:447-472.
12. 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.
13. Jurado RL, Del Rio C, Nassar G, Navarette J, Pimentel JL Jr: Low
anion gap. South Med J 1998, 91:624-629.
14. McAuliffe JJ, Lind LJ, Leith DE, Fencl V: Hypoproteinemic alkalo-
sis. Am J Med 1986, 81:86-90.
15. Rossing TH, Maffeo N, Fencl V: Acidbase effects of altering
plasma protein concentration in human blood in vitro. J Appl
Physiol 1986, 61:2260-2265.
16. Constable, PD, Hinchcliff KW, Muir WW: Comparison of anion
gap and strong ion gap as predictors of unmeasured strong
ion concentration in plasma and serum from horses. Am J Vet
Res 1998, 59:881-887.
17. Kellum JA, Kramer DJ, Pinsky MR: Strong ion gap: a methodology
for exploring unexplained anions. J Crit Care 1995, 10:51-55.
18. Figge J, Jabor A, Kazda A, Fencl V: Anion gap and hypoalbu-
minemia. Crit Care Med 1998, 26:1807-1810.
19. Constable PD: Total weak acid concentration and effective dis-
sociation constant of nonvolatile buffers in human plasma. J
Appl Physiol 2001, 91:1364-1371.
20. Siggaard-Andersen O, Engel K: A new acidbase nomogram,
an improved method for calculation of the relevant blood
acidbase data. Scand J Clin Lab Invest 1960, 12:177.
21. Siggaard-Andersen O: Blood acidbase alignment nomogram.
Scales for pH, PCO
2
, base excess of whole blood of different
hemoglobin concentrations, plasma bicarbonate and plasma
total CO
2
. Scand J Clin Lab Invest 1963, 15:211-217.
Table 3
An example of Stewart formulae (Eqns 1821) in practice
Parameter Control Case 1 Case 2
BE (mEq/l) 0 10 10
AG (mEq/l) 14.8 14.8 14.8
5.7 5.7 2.2
BE
corrected
0 10.6 10.2
[SID
+
]
e
(mEq/l) 39 29 20.7
[SID
+
]
a
(mEq/l) 39 29 29
SIG (mEq/l) 0 0 8.3
AF, anion gap; , buffer value; BE, base excess; SID
+
, strong ion
difference; SIG, strong ion gap.
Table 2
Disease states classified according to the Stewart approach
Acidbase disturbance Disease state Examples
Metabolic alkalosis Low serum albumin Nephrotic syndrome, hepatic cirrhosis
High SID
+
Chloride loss: vomiting, gastric drainage, diuretics, post-hypercapnea, Cl
wasting
diarrhea due to villous adenoma, mineralocorticoid excess, Cushings syndrome,
Liddles syndrome, Bartters syndrome, exogenous corticosteroids, licorice
Na
2+
load (such as acetate, citrate, lactate): Ringers solution, TPN, blood
transfusion
Metabolic acidosis Low SID
+
and high SIG Ketoacids, lactic acid, salicylate, formate, methanol
Low SID
+
and low SIG RTA, TPN, saline, anion exchange resins, diarrhea, pancreatic losses
RTA, renal tubular acidosis; SIG, strong ion gap; SID
+
, strong ion difference; TPN, total parenteral nutrition.
192
22. Watson PD: Modeling the effects of proteins on pH in plasma.
J Appl Physiol 1999, 86:1421-1427.
23. Figge J, Rossing TH, Fencl V: The role of serum proteins in
acidbase equilibria. J Lab Clin Med 1991, 117:453-467.
24. Figge J, Mydosh T, Fencl V: Serum proteins and acid-base
25. Figge J: An Educational Web Site about Modern Human Acid-
Base Physiology: Quantitative Physicochemical Model
[http://www.figgefencl.org]
26. Davenport HW: The A.B.C. of AcidBase Chemistry. Chicago:
University of Chicago Press; 1974.
27. Singer RB, Hastings AB: Improved clinical method for estima-
tion of disturbances of acid-base balance of human blood.
Medicine 1948, 27:223-242.
28. Wooten EW: Calculation of physiological acidbase parame-
29. Watson PD: USC physiology acidbase center: software and
data sets. [http://www.med.sc.edu:96/watson/Acidbase/Acidbase.
htm]
259
ALI = acute lung injury; ARDS = acute respiratory distress syndrome; [A
tot
] = total concentration of weak acids; [H
+
] = H
+
concentration; PCO
2
=
partial CO
2
tension; [SID] = strong ion difference; THAM = tris-hydroxymethyl aminomethane.
Introduction
Acidemia occurs commonly in critically ill patients. Certain
acidoses have specific remedies, for example insulin for the
patient with diabetic ketoacidosis, or fomepizole for the treat-
ment of methanol intoxication. However, the optimal manage-
ment of other forms of acidosis, such as lactic acidosis from
sepsis, is controversial. Specifically, it is unclear for many of
these disorders whether it is appropriate to attempt to correct
arterial pH through the administration of sodium bicarbonate or
other buffering agents, while efforts to treat the underlying
cause of the acidosis proceed apace. Similarly, whether pH
should be corrected in patients with hypercapnea as a result of
lung protective strategies of mechanical ventilation is unknown.
Herein we describe the properties of several buffering agents
and review the evidence for their clinical efficacy. We do not
discuss the administration of sodium bicarbonate to patients
with bicarbonate-losing metabolic acidoses such as occurs with
diarrhea or renal tubular acidosis a practice that enjoys
widespread acceptance. Similarly, the role of buffering agents in
treating intoxication is beyond the scope of the present review.
What is the harm associated with low pH?
Because we understand poorly both the effects of an
elevated arterial H
+
concentration ([H
+
]) as well as the effects
of attempting to correct it, deciding whether to administer a
buffering agent such as sodium bicarbonate to patients with
non-bicarbonate-losing forms of metabolic acidosis is
difficult. Proponents of such an approach typically argue
along the following lines [1].
An elevated arterial [H
+
], in and of itself, is harmful.
The administration of buffer X intravenously will lower the
arterial [H
+
].
Lowering the [H
+
] with buffer X confers clinical benefit.
Any adverse effects of buffer X will be outweighed by its
benefit.
We first consider the evidence supporting the first assertion.
The remaining ones are discussed below in the context of
each individual agent.
What are the effects of an elevated [H
+
]?
Because protein function is sensitive to the [H
+
] of its
environment, an increase in arterial [H
+
] might be expected to
have important detrimental effects on a host of bodily
functions. However, it is unclear to what extent the arterial
blood pH reflects the intracellular pH, which seems likely to
be more relevant. By way of example, consider the effect of
decreasing blood flow to a tissue by 50%. According to the
Review
Bench-to-bedside review: Treating acidbase abnormalities in
the intensive care unit the role of buffers
Brian K Gehlbach
1
and Gregory A Schmidt
2
1
Instructor of Medicine, Section of Pulmonary and Critical Care, University of Chicago, Chicago, Illinois, USA
2
Professor of Medicine, Section of Pulmonary and Critical Care, University of Chicago, Chicago, Illinois, USA
Corresponding author: Gregory A Schmidt, gschmidt@medicine.bsd.uchicago.edu
Published online: 5 May 2004 Critical Care 2004, 8:259-265 (DOI 10.1186/cc2865)
Abstract
The recognition and management of acidbase disorders is a commonplace activity for intensivists.
Despite the frequency with which non-bicarbonate-losing forms of metabolic acidosis such as lactic
acidosis occurs in critically ill patients, treatment is controversial. This article describes the properties
of several buffering agents and reviews the evidence for their clinical efficacy. The evidence
supporting and refuting attempts to correct arterial pH through the administration of currently
available buffers is presented.
Keywords acid-base, acidosis, bicarbonate, buffer, tromethamine
260
Critical Care August 2004 Vol 8 No 4 Gehlbach and Schmidt
Fick relationship the arterialvenous partial CO
2
tension
(PCO
2
) difference will double, assuming that local CO
2
production is constant. This will have the effect of raising the
tissue PCO
2
and lowering its pH; however, the arterial PCO
2
and pH are unchanged and hence do not reveal the
abnormality. The meaning of an individual arterial blood pH is
further limited when one considers the diversity of micro-
circulations and tissue metabolisms throughout the body. The
effects of the elevated [H
+
] may also be difficult to separate
from the effects of the accompanying anion; lactate buffered
to a pH of 7.4, for example, causes a decrease in cardiac
contractility in animal models [2]. Finally, discerning the effect
of an elevated [H
+
] from that of the underlying process
causing the acidosis hypoperfusion, sepsis, or diabetic
ketoacidosis for example is difficult.
Nevertheless, lowering the arterial pH has rather convincingly
been shown to cause a decrease in cardiac contractility. This
effect has been demonstrated in isolated [3,4] and whole
animal heart preparations [5,6], as well as in excised human
ventricular muscle [7]. The net influence of acidosis on the
cardiovascular system is complicated, however, by
concomitant stimulation of the sympatheticadrenal axis. As a
result, acidemia has been shown to increase cardiac output
and pulmonary artery pressure, whereas pulmonary vascular
resistance is not changed [8]. The responsiveness of adre-
nergic receptors to circulating catecholamines is decreased
[911], and the load tolerance of the right ventricle is reduced
[12]. It is unclear whether resuscitability from induced
ventricular fibrillation is impaired [1315]. Fewer patients with
an arterial pH below 7.1 have been studied, making it difficult
to draw any conclusions. Both respiratory and metabolic
acidoses appear to have similar effects, although the effects of
respiratory acidosis are more rapid, presumably because of
rapid diffusion of CO
2
across cell membranes.
Acute hypercapnea causes a decrease in diaphragmatic
contractility and endurance time [16], along with an increase in
cerebral blood flow. In fact, acute elevation in PCO
2
to more
than 70 mmHg may cause loss of consciousness and seizures
[17]. In contrast, more gradual elevations in PCO
2
are well
tolerated, as exhibited by patients with chronic obstructive
pulmonary disease. Broad clinical experience with the
application of lung protective strategies of mechanical
ventilation in patients with acute lung injury (ALI) and status
asthmaticus suggests that modest acidemia (typically pH
7.157.30, PCO
2
5070 mmHg) is remarkably well tolerated. In
general, patients with so-called permissive hypercapnea have a
decrease in systemic vascular resistance, an increase in heart
rate, cardiac output, oxygen delivery, mean pulmonary artery
pressure, and mixed venous oxygen saturation, and unchanged
mean arterial pressure and pulmonary vascular resistance.
The effects of acidosis may differ according to type and
magnitude. Disparate effects of three types of extracellular
acidosis inorganic, respiratory, and lactic on left ventricular
function in isolated rabbit hearts have been described [18].
Lactic acidosis caused a significant increase in the time to
peak left ventricular pressure while retarding ventricular
relaxation, reinforcing the concept that lactate ions have an
independent effect on myocardial function. Different types
and severity of acidosis may also induce different patterns of
inflammatory response. For example, murine macrophage-like
cells stimulated with lipopolysaccharide exhibited an
essentially proinflammatory response when the media
contained hydrochloric acid, but an anti-inflammatory
response when the media contained lactic acid [19].
Furthermore, hydrochloric acid infusion decreased the blood
pressure in septic rats in a dose dependent manner, but
whereas rats with moderately severe acidosis (standard base
excess of 510 mEq/l) had increased plasma nitrate/nitrite
levels, rats with severe acidosis did not [20].
Are there beneficial effects to an elevation in [H
+
] in
critical illness?
Interesting data are emerging regarding potential protective
effects of acidosis, particularly hypercapnic acidosis, in various
experimental models. Acidosis has been shown to protect
cells in a variety of organs (heart, lung, brain, and liver)
against injury from a number of insults, including hypoxia
[2125]. In contrast, hypocapnic alkalosis worsened
ischemiareperfusion ALI in isolated rabbit lungs [26],
whereas hypercapnic and metabolic acidosis afforded
protection [27]. Buffering the hypercapnic acidosis attenuated
the protection conferred. Similarly, rabbits ventilated with
injurious tidal volumes exhibited less ALI histologically when
hypercapnea was present [28]. A protective effect of
hypercapnea on the development of ALI has also been
demonstrated for an experimental model of extrapulmonary
ALI in which rats were subjected to splanchnic ischemia
reperfusion injury [29]. Hypercapnic acidosis was effective at
attenuating endotoxin-induced ALI in an in vivo rat model
[30]; in fact, both prophylactic and therapeutic hypercapnic
acidosis ameliorated lung injury. Conceivably, reducing cells
mechanical work (e.g. in cardiac cells) and metabolic demand
during hypoxia may protect them from ischemia.
Interestingly, the ARDS Network trial [31], which
demonstrated reduced mortality in ALI and acute respiratory
distress syndrome (ARDS) using a protocol employing low
tidal ventilation, allowed for sodium bicarbonate infusion for
acidemia. Whether this therapy had any effect, either negative
or positive, on patient outcome is unclear.
In summary, the negative impact of an elevated arterial [H
+
] is
frequently difficult to discern. We consider the evidence for
and against the administration of different buffering agents
within the context of each agent below.
Buffering agents
Buffers have conventionally been defined in acidbase
chemistry as substances that allow a solution to resist
261
changes in pH in response to administration of H
+
. Problems
exist with this definition, however. First, as discussed below,
conventionally defined buffers such as NaHCO
3
may cause
an increase in arterial [H
+
] in certain circumstances when
they are administered intravenously, while Stewart [32]
demonstrated that a solution containing weak acids (buffers)
such as blood containing albumin resists changes in
[H
+
] much less effectively than the same solution without any
weak acid. Also, the use of the term buffer obscures the
unique mechanisms of each agent. Neverthess, because of
its widespread use, we employ the term buffer to refer to any
agent whose intent is to raise the arterial pH when given
intravenously.
Sodium bicarbonate
Does sodium bicarbonate lower the arterial [H
+
]?
The effects of sodium bicarbonate infusion can be
understood within the following context. Although the
Henderson equation ([H
+
] = 24 PCO
2
/[HCO
3
]) accurately
describes the dissociation equilibrium for carbonic acid, it is
misleading to assume that [HCO
3
] is an independent
determinant of [H
+
]. In fact, the independent determinants of
[H
+
] in the blood are the strong ion difference [SID], the total
concentration of weak acids [A
tot
], and the PCO
2
[32]. Weak
acids [A
tot
] include substances such as albumin and PO
4
,
change relatively little acutely, and have little impact on [H
+
].
Strong ions are those that dissociate fully (or nearly so) in
aqueous solutions, such as Na
+
and Cl
. Because they are

fully dissociated, strong ions do not participate in chemical
reactions in blood like weak ions (such as H
+
or HCO
3
) do.
Because they do not react chemically, all that matters (for
acidbase purposes) is the net difference in their charges.
The [SID] is defined as the difference between the sum of the
major cations (Na
+
, K
+
, Ca
2+
, Mg
2+
) and the sum of the major
anions (Cl
, SO
4
, lactate) in the blood. [SID] is so important

because the difference in charges affects how much water
will dissociate into the charged species H
+
and OH
(i.e.
[SID] is the major determinant of pH).
The arterial [HCO
3
] and pH depend simply and quite

inextricably on the [SID], [A
tot
], and PCO
2
. The intravenous
infusion of sodium bicarbonate solution typically lowers
arterial [H
+
] (raising the pH) through an increase in [SID].
This occurs because Na
+
is a strong cation whereas HCO
3
is not, but rather reacts with [H

+
] to create CO
2
. When
ventilation is not limited, the excess CO
2
that is produced can
be eliminated, and arterial pH is increased so that most
[5,3336], but not all [37,38], whole animal studies have
shown an increase in arterial pH when sodium bicarbonate is
administered. Additionally, two prospective, randomized
controlled trials conducted in mechanically ventilated patients
with lactic acidosis [39,40] demonstrated that sodium
bicarbonate given intravenously causes a modest increase in
arterial pH. When ventilation is fixed, however, as commonly
occurs in mechanically ventilated patients, the effect of
sodium bicarbonate may be to lower arterial pH, as was seen
in patients ventilated with a lung protective strategy [41].
However, evidence supporting an increase in arterial pH with
bicarbonate infusion does not alone support its use for the
treatment of acidosis. First, bicarbonate infusion has been
shown to stimulate the production of lactate in animal models
of hypoxic lactic acidosis [34,38], phenformin-induced lactic
acidosis [37], hemorrhagic shock [35], and diabetic keto-
acidosis [36,42]. As mentioned above, lactate is itself a strong
anion, which may have independent negative effects on
cardiac contractility [2]. Furthermore, the effects of bicarbonate
administration on intracellular pH are far from clear. Because
CO
2
diffuses readily across cell membranes, sodium bicar-
bonate administration may cause a decrease in intracellular
pH. In fact, the findings of cellular and whole animal model
studies examining the effects of bicarbonate infusion on
intracellular pH are variable, with intracellular [H
+
] rising [36],
falling [37,38,4348], not changing [4,14,34,35], or either
rising or falling depending on the buffer used [49,50]. Two
studies of normal volunteers using very different experimental
designs have investigated the effect of bicarbonate on
intracellular pH using magnetic resonance spectroscopy. In
one study [51] bicarbonate attenuated the decrease in
intracellular muscle pH during exercise induced metabolic
acidosis while raising the arterial pH and PCO
2
. In the other
study [46] sodium bicarbonate caused a fall in brain pH.
The effect of bicarbonate on intracellular pH may depend on
the extracellular nonbicarbonate buffering capacity [52]. In
this model, bicarbonate reacts with H
+
to form H
2
O and CO
2
(reaction 1). The abrupt decrease in [H
+
] caused by reaction
1 causes the dissociation of [H
+
] from nonbicarbonate buffer
(back titration of the buffer), which in turn reacts with
bicarbonate to produce more CO
2
. Finally, the CO
2
diffuses
readily into cells, decreasing intracellular pH (an effect that
may be minimized by intracellular bicarbonate buffer).
Does sodium bicarbonate confer any beneficial effects?
In general, whole animal studies fail to demonstrate any
hemodynamic benefit of sodium bicarbonate therapy over
isotonic saline [5,33,34,37,38,53,54]. Additionally, two
randomized controlled trials of sodium bicarbonate therapy in
patients with lactic acidosis [39,40] found no benefit from this
therapy over sodium chloride in improving global hemodynamics
or the cardiovascular response to infused catecholamines.
The effects of sodium bicarbonate therapy in patients with
permissive hypercapnea have received little study,
notwithstanding the inclusion of sodium bicarbonate in the
aforementioned ARDS Network low tidal volume protocol
[31]. One small, uncontrolled study of patients receiving lung
protective ventilation for ALI showed a decrease in arterial pH
with bicarbonate therapy [41]. No benefit from sodium
bicarbonate has been found in the management of diabetic
ketoacidosis [55,56].
262
Summary
Intravenous sodium bicarbonate may decrease the arterial
[H
+
] when ventilation is not limited, but its effect on intra-
cellular pH is unclear. Perhaps more importantly, no clinical
benefit from sodium bicarbonate has been demonstrated in
the setting of lactic or ketoacidosis, but volume overload,
hyperosmolarity [57], and a decrease in ionized calcium [40]
are known to complicate its use.
Carbicarb
Carbicarb is an equimolar mixture of sodium bicarbonate and
sodium carbonate that is not currently available clinically.
Carbicarb raises the [SID] (lowering the arterial [H
+
]) far
more [33,34,43,58] and boosts the PCO
2
far less [33,34,45]
than does sodium bicarbonate when given intravenously to
animals with metabolic acidosis. If the inability of sodium
bicarbonate to demonstrate a benefit in patients with non-
bicarbonate-wasting forms of metabolic acidosis is due to
increased CO
2
generation, then carbicarb should be a
superior agent. In fact, although carbicarb more consistently
lowers intracellular [H
+
] [34,43,45], studies of its effects on
hemodynamics have yielded conflicting findings [4,33,34,43].
This agent deserves further study.
Tromethamine
Tris-hydroxymethyl aminomethane (THAM) is a weak alkali
(pK = 7.8) that reduces arterial [H
+
] without producing CO
2
.
Because it penetrates cells easily, it also reduces intracellular
[H
+
]. Protonated THAM is excreted by the kidneys.
Although THAM has been commercially available for some time
and has seen considerable use outside North America, there
are few studies of its efficacy. THAM incompletely buffered
metabolic acidosis but significantly improved contractility and
relaxation in an isolated blood perfused rabbit heart model [59].
The combination of THAM and sodium bicarbonate perfectly
buffered acidosis without modifying CO
2
, resulting in a
significant improvement in contractility. Weber and colleagues
[60] studied the effect of THAM on systemic hemodynamics in
12 patients with ARDS in whom permissive hypercapnea was
induced with a target CO
2
of 80 mmHg. Hypercapnea had the
following effects on hemodynamics in control patients, in whom
no attempt was made to correct the pH: reduced systemic
vascular resistance, mean arterial pressure and myocardial
contractility, and increased cardiac output and pulmonary
artery pressure. Patients who received THAM experienced
significantly less myocardial depression when compared with
control patients, whereas the effects of hypercapnea on mean
arterial pressure and mean pulmonary artery pressure were
ameliorated. Administration of THAM to 10 patients with
acidosis and ALI caused significant improvements in arterial pH
and base deficit, as well as a decrease in CO
2
that was not
adequately explained by the effects of ventilation [41].
Whether it is even desirable to buffer hypercapnea in ALI
and hypoperfusion states is unclear, as discussed above.
THAM also has potentially serious side effects, including
hypoglycemia, hyperkalemia, extravasation related necrosis,
and, in neonates, hepatic necrosis [61]. Nevertheless, THAM
is an interesting agent that deserves further study, including
as a potential therapy for patients with lactic acidosis.
Alternative agents for lactic acidosis
Dichloroacetate
Conceivably, the lactic acidosis of sepsis may be due in part
to impaired pyruvate oxidation. The pyruvate dehydrogenase
complex is a key regulator of carbohydrate metabolism. This
complex is inactivated by a pyruvate dehydrogenase kinase
that may be activated by sepsis [62], leading to pyruvate
accumulation and subsequently an increase in lactate.
Dichloroacetate stimulates pyruvate kinase, increasing the
oxidation of pyruvate to acetyl coenzyme A.
Initial studies of dichloroacetate in animals and humans were
indeed promising, demonstrating that dichloroacetate
effectively reduced arterial [H
+
] and lactate levels [6365].
There has been one large, randomized, placebo-controlled
trial of dichloroacetate in patients with lactic acidosis due to
sepsis, cardiogenic shock, or massive hemorrhage. Although
dichloroacetate reduced the arterial blood lactate
concentration and improved the arterial pH, it had no effect
on hemodynamics or survival [66]. Further studies of
dichloroacetate in other patient populations and using
different dosing schedules are warranted. Currently, this
therapy is investigational.
Thiamine
Patients with lactic acidosis due to thiamine deficiency (beri
beri) may respond promptly to its administration. Patients at
risk include those with chronic alcoholism, malignancy,
chronic illness, and short bowel syndrome. Lactic acidosis
may also develop in HIV infected patients receiving nucleo-
side analog reverse transcriptase inhibitors [67]. This
disorder is thought to represent drug induced mitochondrial
dysfunction, and there are anecdotal reports of improvement
with thiamine [68]. Although thiamine is an essential cofactor
for pyruvate dehydrogenase, its utility in sepsis with lactic
acidosis has not been studied.
Volume expanders and acidbase disorders
Considerable debate exists regarding the relative merits of
sodium chloride, lactated Ringers solution, or various colloid
solutions in the resuscitation of patients in shock. The
different chemical compositions of these fluids translate into
different acidbase consequences. For example, infusing
large volumes of normal saline intravenously lowers the [SID]
(because the [SID] of saline is zero), raising [H
+
] (and
lowering pH). Whether the dilutional acidosis that results is
harmful, inconsequential, or even protective to the patient is
unclear. Lactated Ringers solution also has an [SID] of zero
but, because lactate is metabolized in the liver (assuming
adequate hepatic perfusion and function), the effect is similar
263
to infusing a fluid with a positive [SID]. Whether this might be
advantageous is not known. New formulations of colloids have
been investigated; in an animal model of septic shock, volume
expansion with Hextend (Bio Time, Inc., Berkeley, CA, USA)
a synthetic colloid in a balanced electrolyte solution that does
not produce metabolic acidosis in humans conferred longer
survival when compared with 0.9% normal saline [69].
Conceivably, the differing effects of various volume
expanders on acidbase status may be important clinically,
but it is the authors view that considerably more work remains
to be done in this area before volume expanders other than
normal saline can be recommended. A detailed analysis of
this subject is beyond the scope of the present review.
When should I administer a buffering agent?
The lack of evidence supporting buffer therapy in human
acidosis makes it difficult to provide explicit recommendations.
Currently, it is unclear whether it is ever advantageous to
administer a buffering agent to a patient with lactic acidosis or
ketoacidosis. In fact, we do not recommend administration of
sodium bicarbonate to patients with lactic acidosis, regardless
of the pH. This includes lactic acidosis caused by
hypoperfusion, sepsis, mitochondrial dysfunction, or liver failure,
or in the setting of cardiopulmonary bypass. If the decision is
made to administer sodium bicarbonate, then slow infusion is
preferable and objective measures of benefit (or harm) should
be sought. Further study into the efficacy of alternative buffering
agents such as THAM and carbicarb is merited.
In patients with severe hyperchloremic metabolic acidosis from
diarrhea or renal tubular acidosis, the administration of sodium
bicarbonate is reasonable. Whether a patient will benefit from
this therapy is difficult to predict and probably depends on the
clinical circumstance. Patients with critical respiratory compro-
mise, who cannot easily compensate for acidemia, could also
benefit. Nevertheless, we find these patients to be quite rare. In
the much more common circumstance of modest hyper-
chloremic acidosis, attempting treatment with buffers is unlikely
to be helpful and may serve to distract the clinician from
addressing the underlying problem.
When buffer therapy is given its effect can be monitored by
serial determination of arterial blood pH, PCO
2
, and serum
anion gap corrected for albumin concentration. Failure to
correct for the nearly ubiquitous hypoalbuminemia present in
the critically ill introduces a systematic error in the detection
of unidentified anions such as lactate or ketoacids [70]. An
alternative approach is to calculate the strong ion gap, but
this requires measurement of albumin and phosphate
concentrations as well as a little more mathematics, and this
may be too cumbersome for regular clinical use.
Conclusion
Acidemia has both harmful and beneficial biological effects.
Sodium bicarbonate is generally ineffective in raising pH
when ventilation is limited, as in patients with ARDS. Even
when alkalinizing agents can correct the pH, evidence of
efficacy is lacking. Thus, these treatments should not be
considered standard therapy in patients with organic
acidoses, such as lactic acidosis. Rather, attention should be
directed toward correcting the underlying basis for the
acidosis. Alternative buffer agents, such as tromethamine,
offer potential advantages over sodium bicarbonate, but
clinical trials in humans are lacking.
Competing interests
None declared.
References
1. Forsythe SM, Schmidt GA: Sodium bicarbonate for the treat-
ment of lactic acidosis. Chest 2000, 117:260-267.
2. Yatani A, Fujino T, Kinoshita K, Goto M: Excess lactate modu-
lates ionic currents and tension components in frog atrial
muscle. J Mol Cell Cardiol 1981, 13:147-161.
3. Poole-Wilson PA, Langer GA: Effect of pH on ionic exchange
and function in rat and rabbit myocardium. Am J Physiol 1975,
229:570-581.
4. Shapiro JI: Functional and metabolic responses of isolated
hearts to acidosis: effects of sodium bicarbonate and Car-
bicarb. Am J Physiol 1990, 258:H1835-H1839.
5. Cooper DJ, Herbertson MJ, Werner HA, Walley KR: Bicarbonate
does not increase left ventricular contractility during L-lactic
academia in pigs. Am Rev Respir Dis 1993, 148:317-322.
6. Wildenthal K, Mierzwiak DS, Myers RW, Mitchell JH: Effects of
acute lactic acidosis on left ventricular performance. Am J
Physiol 1968, 214:1352-1359.
7. Cingolani HE, Faulkner SL, Mattiazzi AR, Bender HW, Graham TP
Jr: Depression of human myocardial contractility with respira-
tory and metabolic acidosis. Surgery 1975, 77:427-432.
8. Thorens J-B, Jolliet P, Ritz M, Chevrolet JC: Effects of rapid per-
missive hypercapnia on hemodynamics, gas exchange, and
oxygen transport and consumption during mechanical ventila-
tion for the acute respiratory distress syndrome. Intensive
Care Med 1996, 22:182-191.
9. Nakanishi T, Okuda H, Kamata K, Seguchi M, Nakazawa M, Takao
A: Influence of acidosis on inotropic effect of catecholamines
in newborn rabbit hearts. Am J Physiol 1987, 253:H1441-
H1448.
10. Marsh JD, Margolis TI, Kim D: Mechanism of diminished con-
tractile response to catecholamines duringacidosis. Am J
Physiol 1988, 254:H20-H27.
11. Campbell GS, Houle DB, Crisp NW Jr, Weil MH, Brown EB Jr:
Depressed response to intravenous sympathicomimetic
agents in humans during acidosis. Dis Chest 1958, 33:18-22.
12. Rose CE Jr, Benthuysen KV, Jackson JT, Tucker CE, Kaiser DL,
Grover RF, Weil JV: Right ventricular performance during
increased afterload impaired by hypercapnicacidosis in con-
scious dogs. Circ Res 1983, 52:76-84.
13. von Planta I, Weil MH, von Planta M, Gazmuri RJ, Duggal C:
Hypercarbic acidosis reduces cardiac resuscitability. Crit Care
Med 1991, 19:1177-1182.
14. Kette F, Weil MH, von Planta M, Gazmuri RJ, Rackow EC: Buffer
agents do not reverse intramyocardial acidosis during cardiac
resuscitation. Circulation 1990, 81:1660-1666.
15. Kerber RE, Pandian NG, Hoyt R, Jensen SR, Koyanagi S, Grayzel
J, Kieso R: Effect of ischemia, hypertrophy, hypoxia, acidosis,
and alkalosis on canine defibrillation. Am J Physiol 1983, 244:
H825-H831.
16. Juan G, Calverley P, Talamo C, Schnader J, Roussos C: Effect of
carbon dioxide on diaphragmatic function in human beings. N
Engl J Med 1984, 310:874-879.
17. Barnett VT, Schmidt GA: Acid-base disorders. In Principles of
Critical Care, 2nd ed. Edited by Hall JB, Schmidt GA, Wood LDH.
New York: McGraw-Hill; 1998:1169-1181.
18. Berger DS, Fellner SK, Robinson KA, Vlasica K, Godoy IE, Shroff
SG: Disparate effects of three types of extracellular acidosis
on left ventricular function. Am J Physiol 1999, 276:H582-H594.
264
19. Kellum JA, Song M, Li J: Lactic, and hydrochloric acids induce
different patterns of inflammatory response in LPS-stimulated
RAW 264.7 cells. Am J Physiol Regul Integr Comp Physiol 2004,
286:R686-R692.
20. Kellum JA, Song M, Venkataraman R: Effects of hyperchloremic
acidosis on arterial pressure and circulating inflammatory
molecules in experimental sepsis. Chest 2004, 125:243-248.
21. Bonventre JV, Cheung JY: Effects of metabolic acidosis on via-
bility of cells exposed to anoxia. Am J Physiol 1985, 249:
C149-C159.
22. Preckel B, Schlack W, Obal D, Barthel H, Ebel D, Grunert S,
Thamer V: Effect of acidotic blood reperfusion on reperfusion
injury after coronary artery occlusion in the dog heart. J Car-
diovascular Pharmacol 1998, 31:179-186.
23. Kitakaze M, Takashima S, Funaya H, Minamino T, Node K, Shi-
nozaki Y, Mori H, Hori M: Temporary acidosis during reperfu-
sion limits myocardial infarct size in dogs. Am J Physiol 1997,
272:H2071-H2078.
24. Vannucci RC, Towfighi J, Heitjan DF, Brucklacher RM: Carbon
dioxide protects the perinatal brain from hypoxic-ischemic
damage: an experimental study in the immature rat. Pediatrics
1995, 95:868-874.
25. Nomura F, Aoki M, Forbess JM, Mayer JE: Effects of hypercarbic
acidotic reperfusion on recovery of myocardial function after
cardioplegic ischemia in neonatal lambs. Circulation 1994, 90:
321-327.
26. Laffey JG, Engelberts D, Kavanagh BP: Injurious effects of
hypocapnic alkalosis in the isolated lung. Am J Respir Crit
Care Med 2000, 162:399-405.
27. Laffey JG, Engelberts D, Kavanagh BP: Buffering hypercapnic
acidosis worsens acute lung injury. Am J Respir Crit Care Med
2000, 161:141-146.
28. Sinclair SE, Kregenow DA, Lamm WJ, Starr IR, Chi EY, Hlastala
MP: Hypercapnic acidosis is protective in an in vivo model of
ventilator-induced lung injury. Am J Respir Crit Care Med
2002, 166:403-408.
29. Laffey JG, Jankov RP, Engelberts D, Tanswell AK, Post M, Lindsay
T, Mullen JB, Romaschin A, Stephens D, McKerlie C, Kavanagh
BP: Effects of therapeutic hypercapnia on mesenteric
ischemia-reperfusion injury. Am J Respir Crit Care Med 2003,
168:1383-1390.
30. Laffey JG, Honan D, Hopkins N, Hyvelin J-M, Boylan JF, McLough-
lin P: Hypercapnic acidosis attenuates endotoxin-induced
acute lung injury. Am J Respir Crit Care Med 2004, 169:46-56.
31. The Acute Respiratory Distress Syndrome Network: Ventilation
with lower tidal volumes as compared with traditional tidal
volumes for acute lung injury and the acute respiratory dis-
tress syndrome. N Engl J Med 2000, 342:1301-1308.
32. Stewart PA: How to Understand Acidbase: a Quantitative Acid-
base Primer for Biology and Medicine. New York: Elsevier; 1981.
33. Benjamin E, Oropello JM, Abalos AM, Hannon EM, Wang JK,
Fischer E, Iberti TJ: Effects of acid base correction on hemody-
namics, oxygen dynamics, and resuscitability in severe canine
hemorrhagic shock. Crit Care Med 1994, 22:1616-1623.
34. Rhee KH, Toro LO, McDonald GG, Nunnally RL, Levin DL: Car-
bicarb, sodium bicarbonate, and sodium chloride in hypoxic
lactic acidosis: effect on arterial blood gases, lactate concen-
trations, hemodynamic variables, and myocardial intracellular
pH. Chest 1993, 104:913-918.
35. Beech JS, Nolan KM, Iles RA, Cohen RD, Williams SC, Evans SJ:
The effects of sodium bicarbonate and a mixture of sodium
bicarbonate and carbonate (Carbicarb) on skeletal muscle pH
and hemodynamic status in rats with hypovolemic shock.
Metabolism 1994, 43:518-522.
36. Beech JS, Williams SCR, Iles RA, Cohen RD, Nolan KM, Evans
SJ, Going TC: Haemodynamic and metabolic effects in dia-
betic ketoacidosis in rats of treatment with sodium bicarbon-
ate or a mixture of sodium bicarbonate and sodium
carbonate. Diabetologia 1995, 38:889-898.
37. Arieff AI, Leach W, Park R, Lazarowitz VC: Systemic effects of
NaHCO
3
in experimental lactic acidosis in dogs. Am J Physiol
1982, 242:F586-F591.
38. Graf H, Leach W, Arieff AI: Metabolic effets of NaHCO
3
in
hypoxic lactic acidosis in dogs. Am J Physiol 1985, 249:F630-
F635.
39. Mathieu D, Neviere R, Billard V, Fleyfel M, Wattel F: Effects of
bicarbonate therapy on hemodynamics and tissue oxygena-
tion in patients with lactic acidosis: a prospective, controlled
clinical study. Crit Care Med 1991, 19:1352-1356.
40. Cooper DJ, Walley KR, Wiggs BR, Russell JA: Bicarbonate does
not improve hemodynamics in critically ill patients who have
lactic acidosis: a prospective, controlled clinical study. Ann
Intern Med 1990, 112:492-498.
41. Kallet RH, Jasmer RM, Luce JM, Lin LH, Marks JD: The treatment
of acidosis in acute lung injury with tris-hydroxymethyl
aminomethane (THAM). Am J Respir Crit Care Med 2000, 161:
1149-1153.
42. Bureau MA, Begin R, Berthiaume Y, Shapcott D, Khoury K,
Gagnon N: Cerebral hypoxia from bicarbonate infusion in dia-
betic acidosis. J Pediatr 1980, 96:968-973.
43. Bersin RM, Arieff AI: Improved hemodynamic function during
hypoxia with carbicarb, a new agent for the management of
acidosis. Circulation 1988, 77:227-233.
44. Shapiro JI, Whalen M, Kucera R, Kindig N, Filley G, Chan L: Brain
pH responses to sodium bicarbonate and carbicarb during
systemic acidosis. Am J Physiol 1989, 25:H1316-H1321.
45. Shapiro JI, Whalen M, Chan L: Hemodynamic and hepatic pH
responses to sodium bicarbonate and carbicarb during sys-
temic acidosis. Magnetic Res Med 1990, 16:403-410.
46. Nakashima K, Yamashita T, Kashiwagi S, Nakayama N, Kitahara T,
Ito H: The effect of sodium bicarbonate on CBF and intracellu-
lar pH in man: stable Xe-CT and
31
P-MRS. Acta Neurol Scand
Suppl 1996, 166:96-98.
47. Bjerneroth G, Sammeli O, Li YC, Wiklund L: Effects of alkaline
buffers on cytoplasmic pH in lymphocytes. Crit Care Med
1994, 22:1550-1556.
48. Ritter JM, Doktor HS, Benjamin N: Paradoxical effect of bicar-
bonate on cytoplasmic pH. Lancet 1990, 335:1243-1246.
49. Goldsmith DJA, Forni LG, Hilton PJ: Bicarbonate therapy and
intracellular acidosis. Clin Sci 1997, 93:593-598.
50. Levraut J, Labib Y, Chave S, Payan P, Raucoules-Aime M,
Grimaud D: Effect of sodium bicarbonate on intracellular pH
under different buffering conditions. Kidney Int 1996, 49:1262-
1267.
51. Nielsen HB, Hein L, Svendsen LB, Secher NH, Quistorff B: Bicar-
bonate attenuates intracellular acidosis. Acta Anaesthesiol
Scand 2002, 46:579-584.
52. Levraut J, Giunti C, Ciebiera J-P, de Sousa G, Ramhani R, Payan
P, Grimaud D: Initial effect of sodium bicarbonate on intracel-
lular pH depends on the extracellular nonbicarbonate buffer-
ing capacity. Crit Care Med 2001, 29:1033-1039.
53. Tanaka M, Nishikawa T, Mizutani T: Normovolemic hemodilution
attenuates cardiac depression induced by sodium bicarbon-
ate in canine metabolic acidosis. Br J Anaesth 1996, 77:408-
412.
54. Tanaka M, Nishikawa T: Acute haemodynamic effects of
sodium bicarbonate administration in respiratory and meta-
bolic acidosis in anaesthetized dogs. Anaesth Intensive Care
1997, 25:615-620.
55. Lever E, Jaspan JB: Sodium bicarbonate therapy in severe dia-
betic ketoacidosis. Am J Med 1983, 75:263-268.
56. Morris LR, Murphy MB, Kitabchi AE: Bicarbonate therapy in
severe diabetic ketoacidosis. Ann Intern Med 1986, 105:836-
840.
57. Mattar JA, Weil MH, Shubin H, Stein L: Cardiac arrest in the crit-
ically ill: II. Hyperosmolal states following cardiac arrest. Am J
Med 1974, 56:162-168.
58. Sun JH, Filley GF, Hord K, Kindig NB, Bartle EJ: Carbicarb: an
effective substitute for NaHCO3 for the treatment of acidosis.
Surgery 1987, 5:835-839.
59. Sirieix D, Delayance S, Paris M, Massonnet-Castel S, Carpentier
A, Baron J-F: Tris-hydroxymethyl aminomethane and sodium
bicarbonate to buffer metabolic acidosis in an isolated heart
model. Am J Respir Crit Care Med 1997, 155:957-963.
60. Weber T, Tschernich H, Sitzwohl C, Ullrich R, Germann P,
Zimpfer M, Sladen RN, Huemer G: Tromethamine buffer modi-
fies the depressant effect of permissive hypercapnia on
myocardial contractility in patients with acute respiratory dis-
tress syndrome. Am J Respir Crit Care Med 2000, 162:1361-
1365.
61. Adrogue HJ, Madias NE: Management of life-threatening
acidbase disorders. N Engl J Med 1998, 338:26-34.
62. Vary TC: Increased pyruvate dehydrogenase kinase activity in
response to sepsis. Am J Physiol 1991, 260:E669-E674.
265
63. Park R, Arieff AI: Treatment of lactic acidosis with dichloroac-
etate in dogs. J Clin Invest 1982, 70:853-862.
64. Stacpoole PW, Lorenz AC, Thomas RG, Harman EM:
Dichloroacetate in the treatment of lactic acidosis. Ann Intern
Med 1988, 108:58-63.
65. Stacpoole PW, Harman EM, Curry SH, Baumgartner TG, Misbin
RI: Treatment of lactic acidosis with dichloroacetate. N Engl J
Med 1983, 309:390-399.
66. Stacpoole PW, Wright EC, Baumgartner TG, Bersin RM, Buchal-
ter S, Curry SH, Duncan CA, Harman EM, Henderson GN, Jenkin-
son S, Lachin JM, Lorenz A, Schneider SH, Siegel JH, Summer
WR, Thompson D, Wolfe CL, Zorovich B, and the DCA Lactic
Acidosis Study Group: A controlled clinical trial of dichloroac-
etate for treatment of lactic acidosis in adults. N Engl J Med
1992, 327:1564-1569.
67. Falco V, Rodriguez D, Ribera E, Martinez E, Miro JM, Domingo P,
Diazaraque R, Arribas JR, Gonzalez-Garcia JJ, Montero F,
Sanchez L, Pahissa A: Severe nucleoside-associated lactic aci-
dosis in human immunodeficiency virus-infected patients:
report of 12 cases and review of the literature. Clin Infect Dis
2002, 34:838-846.
68. Schramm C, Wanitschke R, Galle PR: Thiamine for the treat-
ment of nucleoside analogue-induced severe lactic acidosis.
Eur J Anaesthesiol 1999, 16:733-735.
69. Kellum JA: Fluid resuscitation and hyperchloremic acidosis in
experimental sepsis: improved short-term survival and acid-
base balance with Hextend compared with saline. Crit Care
Med 2002, 30:300-305.
Care Med 2000, 162:2246-2251.
108
ARF = acute renal failure; Atot = total concentration of nonvolatile weak acid; CVVH = continuous venovenous hemofiltration; CVVHDF = continuous
venovenous hemodiafiltration; HVHF = high-volume hemofiltration; IHD = intermittent hemodialysis; PCO
2
= partial carbon dioxide tension; RRT =
renal replacement therapy; SID = strong ion difference; SIG = strong ion gap.
Critical Care April 2004 Vol 8 No 2 Naka and Bellomo
Introduction
Acute renal failure (ARF) in the critically ill is still associated
with a poor prognosis [1,2]. Metabolic acidbase disorders
are particularly common in these patients, especially acidosis.
The pathogenesis of such acidosis remains poorly under-
stood because its main cause in ARF patients is not fully
understood. However, the nature of this metabolic acidosis is
likely multifactorial and probably includes the effect of chlo-
ride-rich fluid resuscitation [3] and the accumulation of
lactate, phosphate, and unexcreted metabolic acids such as
sulfate [4]. This multifactorial metabolic acidosis associated
with ARF often leads to acidemia. Furthermore, persistent
Review
Bench-to-bedside review: Treating acidbase abnormalities in
the intensive care unit the role of renal replacement therapy
Toshio Naka
1
and Rinaldo Bellomo
2
1
Research Fellow, Department of Intensive Care and Department of Medicine, Austin Hospital, Melbourne, Australia
2
Professor, Director of Intensive Care Research, Department of Intensive Care, Austin Hospital, Heidelberg, Victoria, and University of Melbourne,
Melbourne, Australia
Correspondence: Rinaldo Bellomo, rinaldo.bellomo@armc.org.au
Published online: 17 February 2004 Critical Care 2004, 8:108-114 (DOI 10.1186/cc2821)
2004 BioMed Central Ltd (Print ISSN 1364-8535; Online ISSN 1466-609X)
Abstract
Acidbase disorders are common in critically ill patients. Metabolic acidbase disorders are
particularly common in patients who require acute renal replacement therapy. In these patients,
metabolic acidosis is common and multifactorial in origin. Analysis of acidbase status using the
StewartFigge methodology shows that these patients have greater acidemia despite the presence of
hypoalbuminemic alkalosis. This acidemia is mostly secondary to hyperphosphatemia, hyperlactatemia,
and the accumulation of unmeasured anions. Once continuous hemofiltration is started, profound
changes in acidbase status are rapidly achieved. They result in the progressive resolution of acidemia
and acidosis, with a lowering of concentrations of phosphate and unmeasured anions. However, if
lactate-based dialysate or replacement fluid are used, then in some patients hyperlactatemia results,
which decreases the strong ion difference and induces an iatrogenic metabolic acidosis. Such
hyperlactatemic acidosis is particularly marked in lactate-intolerant patients (shock with lactic acidosis
and/or liver disease) and is particularly strong if high-volume hemofiltration is performed with the
associated high lactate load, which overcomes the patients metabolic capacity for lactate. In such
patients, bicarbonate dialysis seems desirable. In all patients, once hemofiltration is established, it
becomes the dominant force in controlling metabolic acidbase status and, in stable patients, it
typically results in a degree of metabolic alkalosis. The nature and extent of these acidbase changes
is governed by the intensity of plasma water exchange/dialysis and by the buffer content of the
replacement fluid/dialysate, with different effects depending on whether lactate, acetate, citrate, or
bicarbonate is used. These effects can be achieved in any patient irrespective of whether they have
acute renal failure, because of the overwhelming effect of plasma water exchange on nonvolatile acid
balance. Critical care physicians must understand the nature, origin, and magnitude of alterations in
acidbase status seen with acute renal failure and during continuous hemofiltration if they wish to
provide their patients with safe and effective care.
Keywords acidosis, alkalosis, bicarbonate, hemofiltration, hemodialysis, renal replacement therapy
109
acidosis has been demonstrated to be an indicator of poor
prognosis [5]. The rationale behind the perceived need to
correct severe acidosis lies in the potential adverse cellular
effects of such metabolic disturbance on myocardial function,
likelihood of arrhythmias, and pulmonary vascular tone.
However, very few studies [6] have in fact established that
clinically significant benefits might arise from the correction of
such acidosis.
Nonetheless, renal replacement therapy (RRT) such as inter-
mittent hemodialysis (IHD), continuous venovenous hemofil-
tration (CVVH), continous venovenous hemodailysis, and
continuous venovenous hemodiafiltration (CVVHDF) has
been applied to the treatment of critically ill patients with ARF
to improve fluid overload, uremia, and acidbase disorders.
The use of RRT and adjustments in the replacement solutions
administered to acidotic critically ill patients with ARF can
have a substantial effect on acidbase homeostasis. Further-
more, high-volume hemofiltration (HVHF) may have an even
stronger effect on acidbase disorders. Therefore, improving
our understanding of the impact of RRT on acidbase disor-
ders and gaining insights into the nature of such disorders
and the mechanisms of action of RRT are important.
In the present review we explore the acidbase disorders
seen in ARF, the effect of RRT and its modalities on
acidbase disorders, the effect of replacement fluid on
acidbase balance, and the effect of HVHF on acidbase
balance. A strong focus is given to the clinical implications of
these interventions, with the aim of helping clinicians better
understand and manage the acidbase disorders in ARF and
critically ill patients in general.
Acidbase analysis using the StewartFigge
methodology
As described above, the pathogenesis of acidbase disor-
ders of ARF remains unknown and the cause of acidosis in
ARF patients is probably multifactorial. It is hard to quantita-
tively approciate such multifactorial metabolic disorders by
means of the classical HendersonHasselbach method.
Recently, however, quantitative acidbase analysis using the
StewartFigge approach [7,8] was introduced. This method
first involves calculating the apparent strong ion difference
(SID; all concentrations in mEq/l):
Apparent SID = [Na
+
] + [K
+
] + [Mg
2+
] + [Ca
2+
] [Cl
] [lactate]
The calculation then takes into account the role of weak acids
(carbon dioxide, albumin, and phosphate) in the balance of
electrical charges in plasma water, as expressed through cal-
culation of the effective SID (partial carbon dioxide tension
[PCO
2
] in mmHg, albumin in g/l, and phosphate in mmol/l):
Effective SID = 1000 2.46 10
11
PCO
2
/(10
pH
) +
[albumin] (0.12 [pH 0.631]) + [phosphate] (0.309
[pH 0.469])
Once weak acids are quantitatively taken into account, the
difference between apparent and effective SID should be
zero, unless there are unmeasured charges (anions). Such
charges are then described by the strong ion gap (SIG):
SIG= apparent SID effective SID.
The component of albumin and phosphate is defined as the
total concentration of nonvolatile weak acid (Atot). [Atot],
along with SID and PCO
2
, is an independent determinant of
[H
+
] or pH. According to the StewartFigge approach, meta-
bolic acidosis can then result from a reduction in the SID or
from an increase in Atot, and respiratory acidosis can result
from a gain in PCO
2
. The changes in each of these variables
can be quantified to express how much each one is responsi-
ble (in mEq/l) for the findings on blood analysis.
Acidbase balance in acute renal failure
Classically, metabolic acidosis in renal failure is described as
a high anion gap metabolic acidosis. However, in the clinical
setting, the anion gap is not always elevated. These findings
might lead clinicians to diagnostic and therapeutic confusion.
In these situations, quantitative analysis using the Stewart
Figge approach can be helpful. In this regard, Rocktaeschel
and coworkers [9] recently examined the acidbase status of
ARF patients using the StewartFigge methodology and
demonstrated several features. First, critically ill patients with
ARF were typically acidemic compared with control patients
(Fig. 1). Second, this acidemia appeared secondary to meta-
bolic acidosis with a mean base excess of approximately
7 mEq/l, which appeared secondary to the accumulation of
lactate, phosphate, and unmeasured anions (possible candi-
dates for these unmeasured anions include sulfate, urate,
hydroxypropionate, oxalate, and furanpropionate [10]; Fig. 2).
Third, in these patients there was also a marked failure to
alter the apparent SID to achieve a degree of metabolic com-
pensation (Fig. 3). Despite this finding, half of the ARF
patients had an anion gap within the normal range. Further-
more, these acidifying disorders were attenuated by a con-
comitant metabolic alkalosis, which was essentially
secondary to hypoalbuminemia. Hypoalbuminemia lowered
the anion gap and masked the presence of acidifying anions
to those clinicians using conventional acidbase analysis.
Effect of renal replacement therapy on
acidbase balance
There are two major modalities of RRT. One is intermittent
and the other continuous. Few studies have been done to
detect which modality is better in terms of acidbase control.
Uchino and coworkers [11] compared the effect on
acidbase balance of IHD and CVVHDF. Before treatment,
metabolic acidosis was common in both groups (63.2% for
IHD and 54.3% for CVVHDF). Both IHD and CVVHDF cor-
rected metabolic acidosis. However, the rate and degree of
correction differed significantly. CVVHDF normalized meta-
bolic acidosis more rapidly and more effectively during the
110
first 24 hours than did IHD (P < 0.01). IHD was also associ-
ated with a higher incidence of metabolic acidosis than was
CVVHDF during the subsequent 2 week treatment period
(P < 0.005; Fig. 4). Accordingly, CVVHDF can be considered
physiologically superior to IHD in the correction of metabolic
acidosis. The overwhelming superiority of continuous RRT in
terms of control of acidosis was also recently established in
comparison with peritoneal dialysis, with all patients random-
ized to CVVH achieving correction of acidosis by 50 hours of
treatment, compared with only 15% of those treated by peri-
toneal dialysis (P < 0.001) [12]. How does continuous RRT
correct acidosis?
To gain insights into the mechanisms by which continuous
RRT corrects metabolic acidosis in ARF, Rocktaschel and
coworkers [13] studied the effect of CVVH on acidbase
balance using the StewartFigge methodology. Before com-
mencing CVVH, patients had mild acidemia secondary to
metabolic acidosis. This acidosis was due to increased
unmeasured anions (SIG 12.3 mEq/l), hyperphosphatemia,
and hyperlactatemia. It was attenuated by the alkalizing effect
of hypoalbuminemia. Once CVVH was commenced, acidemia
was corrected within 24 hours. This change was associated
with a decreased SIG, and decreased phosphate and chlo-
ride concentrations. This correction was so powerful and
dominant that, after 3 days of CVVH, patients developed alka-
lemia secondary to metabolic alkalosis (bicarbonate
29.8 mmol/l, base excess 6.7 mmol/l; Fig. 1). This alkalemia
appeared due to a further decrease in SIG and a further
decrease in serum phosphate concentration in the setting of
persistent hypoalbuminemia. Hence, CVVH appears to
Figure 2
Differences in strong ion gap (SIG) between (ARF) patients and
controls in an intensive care unit.
Figure 3
Differences in apparent strong ion difference (SIDa) between acute renal
failure (ARF) patients and control individuals in an intensive care unit.
Figure 1
Difference in pH between patients with acute renal failure (ARF) in an
intensive care unit (ICU) and a control population of ICU patients.
111
correct metabolic acidosis in ARF through its effects on
unmeasured anions, phosphate, and chloride. Once hemofil-
tration is established, it becomes the dominant force in con-
trolling metabolic acidbase status, and in stable patients it
typically results in a degree of metabolic alkalosis.
Effect of replacement fluid composition
(lactate, acetate, bicarbonate, and citrate)
The exchange of approximately 30 l plasma water per day is
necessary to achieve adequate control of uremia and
acidbase disorders in ARF [14]. During continuous RRT,
according to conventional acidbase thinking, there is a sub-
stantial loss of endogenous bicarbonate, which must be sub-
stituted by the addition of buffer substances. (According to
the StewartFigge approach, the explanation for this is that
there is loss of a fluid with an SID of approximately 40 mEq/l,
which must be replaced by a fluid with a similar SID.)
Lactate, acetate, and bicarbonate have been used as buffers
(or SID generators according to Stewart [7]) during RRT.
Citrate has been used as a buffer and for anticoagulation.
These buffers affect acidbase balance, and therefore we
must understand their physiologic characteristics.
Bicarbonate has the major advantage of being the most phys-
iologic anion equivalent. However, the production of a com-
mercially available bicarbonate-based solution is not easy
because of the formation of calcium and magnesium salts
during long-term storage. Furthermore, the cost of this solu-
tion is approximately three times greater than that of other
buffer solutions. Accordingly, acetate and lactate have been
used widely for RRT. Under normal conditions, acetate is
rapidly converted on a 1:1 basis to carbon dioxide and then
bicarbonate by both liver and skeletal muscle. Lactate is also
rapidly converted in the liver on a 1:1 basis [15].
Studies of acetate-based solutions appear to exert a negative
influence on the mean arterial blood pressure and cardiac
function in the critically ill [1618]. Morgera and coworkers
[19] compared acidbase balance between acetate-buffered
and lactate-buffered replacement fluids, and reported that the
acetate-buffered solution was associated with a significant
lower pH and bicarbonate levels than was the lactate-
buffered solution. However, the acetate-buffered solution had
9.5 mmol/l less buffer than the lactate-buffered one. There-
fore, the difference is probably simply a matter of dose rather
than choice of buffer. From the StewartFigge perspective,
the acetate-buffered solution contained 8 mmol/l chloride
more than the lactate-buffered solution to achieve electrical
equilibrium. This reduces the SID of the replacement fluid and
acidifies blood more.
Thomas and coworkers [20] compared the effects of lactate-
buffered versus bicarbonate-buffered fluids. Hemofiltration
fluids contained either 44.5 mmol/l sodium lactate or
40.0 mmol/l sodium bicarbonate with 3 mmol/l lactate
(43 mmol/l). Lactate-buffered fluids contained 142 mmol/l
sodium and 103 mmol/l chloride (SID 39 mEq/l), and bicar-
bonate-buffered fluids contained 155 mmol/l sodium and
120 mmol/l chloride (SID 35 mEq/l). Lactate rose from
approximately 2 mmol/l to 4 mmol/l when lactate-based fluids
were given but not with bicarbonate. Both therapies resulted
in a similar improvement in metabolic acidosis. Potentially, the
lactate-buffered fluid could have had a more alkalinizing
effect. However, the accumulation of lactate in blood might
have offset this effect and attenuated the trend toward a
higher base excess with the lactate-buffered fluids.
Tan and coworkers [21] studied the acidbase effect of
CVVH with lactate-buffered and bicarbonate-buffered solu-
tions. The lactate-buffered solution had an SID of 46 mEq/l,
as compared with 35 mEq/l for the bicarbonate fluid. From
the StewartFigge point of view, the lactate-buffered solution
should have led to a greater amount of alkalosis. However,
that study found a significant increase in plasma lactate levels
and a decrease in base excess with the lactate-buffered solu-
tion (Figs 5 and 6). Lactate, if not metabolized and still
present in blood, acts as a strong anion, which would have
the same acidifying effect of chloride. Accordingly, iatrogenic
hyperlactatemia can cause a metabolic acidosis (Fig. 7). The
controversy can, of course, also be resolved by failure to
convert exogenous lactate into bicarbonate.
Most commercially available replacement fluids are buffered
with approximately 4046 mmol/l lactate. In the vast majority
of patients, the administration of such replacement fluid main-
tains a normal serum bicarbonate level without any significant
increase in blood lactate concentration. Because the ability of
the liver to metabolize lactate is in the region of
100 mmol/hour [22], even aggressive CVVH at 2 l/hour
exchange would still deliver less than the normal liver can
handle.
Figure 4
Box plot illustrating bicarbonate control with intermittent dialysis (IHD)
and continuous therapy (continuous venovenous hemodiafiltration
[CVVHDF]).
112
However, if lactate-based dialysate or replacement fluids are
used in some patients with liver dysfunction or shock, then
the administration of lactate-buffered fluids can induce signifi-
cant hyperlactatemia and acidosis because the metabolic
rate is insufficient to meet the additional lactate load.
Although lactate normally acts as a buffer by being removed
from the circulation and thereby lowering the SID, if lactate is
only partly metabolized and accumulates in plasma water
then it acts like a strong anion. Thus, hyperlactatemia
decreases the apparent SID, which results in increased dis-
sociation of plasma water and thereby lowers the pH.
Citrate has been used for regional anticoagulation. During
this procedure, citrate is administered to the circuit before the
filter and chelates calcium, thus impeding coagulation. Once
citrate enters the circulation, it is metabolized to carbon
dioxide and then bicarbonate on a 1: 3 basis; thus, 1 mmol
citrate yields 3 mmol carbon dioxide and then bicarbonate.
Under these circumstances, citrate acts as the buffer as well
as the anticoagulant. If the method described by Mehta and
coworkers [23] is applied, then approximately 48 mmol/hour
bicarbonate equivalent is given as citrate. This rate of alkali
administration may result in metabolic alkalosis (in up to 25%
of cases). Caution is warranted in patients with liver disease,
who may not be able to metabolize citrate. In these patients,
citrate may accumulate and result in severe ionized hypocal-
cemia and metabolic acidosis because the citrate anion
(C
6
H
5
O
7
3
) acts as an unmeasured anion and increases the
SIG, which has acidifying effects.
When oxidizable anions are used in the replacement fluids,
the anion (acetate, lactate, and citrate) must be completely
oxidized to carbon dioxide and water in order to generate
bicarbonate. If the metabolic conversion of nonbicarbonate
anions proceeds without accumulation, then their buffering
capacity is equal to that of bicarbonate. Thus, the effect on
acidbase status depends on the buffer concentration
rather than on the kind of buffer used [15]. When the meta-
bolic conversion is impaired, the increased blood concentra-
tion of the anions leads to an increased strong anion in
lactate or unmeasured anions for acetate and citrate. All
lower the apparent SID and acidify blood. The nature and
extent of these acidbase changes is governed by the inten-
sity of plasma water exchange/dialysis, by the buffer content
of the replacement fluid/dialysate, and by the metabolic rate
for these anions.
Effect of high volume hemofiltration on
acidbase balance
Recently, HVHF was applied to the treatment of septic shock
patients, with favorable hemodynamic results [24]. However,
Figure 6
Effect of bicarbonate-based replacement fluids (bicarbonate RF) and
lactate-based replacement fluids (lactate RF) on base excess.
Figure 7
lactate-based replacement fluids (lactate RF) on serum bicarbonate
levels.
Figure 5
lactate-based replacement fluids (lactate RF) on blood lactate levels.
113
if commercial lactate-buffered replacement fluid is used
during HVHF, then patients might receive more than
270 mmol/hour exogenous lactate. This lactate load could
overcome endogenous lactate metabolism, even in healthy
subjects [25], and result in progressive hyperlactatemia.
Hyperlactatemia has been reported with lactate-buffered fluids
in critically ill ARF patients treated with intermittent hemofiltra-
tion and a lactate load of 190210 mmol/hour [16]. Such
hyperlactatemia might induce a metabolic acidosis. Cole and
coworkers [26] studied the effect of HVHF on acidbase
balance. HVHF with lactate-buffered replacement fluids
(6 l/hour of lactate-buffered fluids) induced iatrogenic hyper-
lactatemia. Plasma lactate levels increased from a median of
2.51 mmol/l to a median of 7.3 mmol/l at 2 hours (Fig. 8). This
change was accompanied by a significant decrease in bicar-
bonate and base excess. However, such hyperlactatemia had
only a mild and transient acidifying effect. A decrease in chlo-
ride and effective SID and the removal of unmeasured anions
(decrease in SIG) all rapidly compensated for this effect
(Fig. 9). Thus, the final effect was that HVHF induced only a
minor change in pH from 7.42 to 7.39 at 2 hours. In the period
from 2 to 8 hours, the blood lactate concentration remained
stable at around 78 mmol/l, whereas compensatory effects
continued, which restored bicarbonate levels to 27.2 mmol/l
and pH to 7.44 by 8 hours of treatment.
Although the chloride concentration in the replacement fluid
was high compared with the serum chloride level, a progres-
sive decrease in chloride was observed. This might be due to
chloride losses in excess of gains. Uchino and coworkers
[27] examined the sieving coefficient for chloride during
HVHF and found a sieving coefficient for chloride in excess of
1. Another possible explanation for hypochloremia would be
the intracellular movement of chloride in response to meta-
bolic acidosis (chloride shift). A decrease in effective SID
was explained by the aggregate minor changes in arterial
PCO
2
, albumin, and phosphate. The changes in SIG appeared
most likely to be due to simple filtration of unmeasured anion.
Consequently, HVHF with lactate-buffered fluids induced a
marked hyperlactatemia but did not induce a progressive aci-
dosis. However, caution is warranted in particular patients
who have marked pretreatment hyperlactatemia (>5 mmol/l)
or liver dysfunction, or where the intensity of HVHF exceeds
6 l/hour plasma water exchange. Bicarbonate use is war-
ranted in such patients.
Conclusion
RRT can strongly affect acidbase disorders and can be
used to correct severe metabolic acidosis. If the dose of
treatment is titrated to achieve such a goal, essentially even
the most dramatic metabolic acidosis can be corrected.
Replacement fluid solutions containing buffers such as
lactate, acetate, bicarbonate, and citrate can have a variable
effect on acidbase balance, depending on the dose and rate
of metabolic disposition, as clearly seen in the setting of
HVHF. Critical care physicians must understand the nature,
origin, and magnitude of the alterations in acidbase status
seen with ARF and associated disorders, and the powerful
effects of continuous hemofiltration if they wish to provide
their patients with safe and effective care.
Competing interests
None declared.
References
1. Spiegel DM, Ullian ME, Zerbe GO, Berl T: Determinant of sur-
vival and recovery in acute renal failure patients dialysed in
the intensive care unit. Am J Nephrol 1991, 11:44-47.
2. Brivet F, Kleinknecht D, Loirat P: Acute renal failure in intensive
care units causes, outcome, and prognostic factors: a
prospective, multicenter study. Crit Care Med 1996, 24:192-
198.
3. Kellum JA, Bellomo R, Kramer DJ, Pinsky MR: Etiology of meta-
bolic acidosis during saline resuscitation in endotoxemia.
Shock 1998, 9:1-5.
Figure 9
Effect of high-volume hemofiltration (HVHF) on chloride, effective
strong ion difference (SIDe), and strong ion gap (SIG).
Figure 8
Effect of high-volume hemofiltration (HVHF) on lactate, bicarbonate,
and base excess.
114
4. Prough DS, Bidani A: Hyperchloremic metabolic acidosis is a
predictable consequence of intraoperative infusion of 0.9%
saline. Anesthesiology 1999, 90:1247-1249.
5. Wendon J, Smithies M, Sheppard M, Bullen K, Tinker J, Bihari DJ:
Continuous high volume venous-venous hemofiltration in
acute renal failure. Intensive Care Med 1989, 15:358-363.
6. Kirshbaum B: Effect of hemodialysis on the hypersalphatemia
of chronic renal failure. ASAIO J 1998, 44:314-318.
equilibria: a follow up. J Lab Clin Med 1992, 120:713-719.
9. Rocktaeschel J, Morimatsu H, Uchino S, Goldsmith D, Pousie S,
Story DA, Gutteridge G, Bellomo R: Acid-base status of criti-
cally ill patients with acute renal failure: analysis based on
Stewart-Figge methodology. Crit Care 2003, 7:60-66.
10. Niwa T: Organic acids and the uremic syndrome: protein
metabolite hypothesis in the progression of chronic renal
failure. Semin Nephrol 1996, 16:167-182.
11. Uchino S, Bellomo R, Ronco C: Intermittent versus continous
renal replacement therapy in the ICU: impact on electrolyte
and acid-base balance. Intensive Care Med 2001, 27:1037-
1043.
12. Phu NH, Hien TT, Mai NT, Chau TT, Chuong LV, Loc PP, Winearls
C, Farrar J, White N, Day N: Hemofitlration and peritoneal dialy-
sis in infection-associated acute renal failure in Vietnam. N
Engl J Med 2002, 347:895-902.
13. Rocktaschel J, Morimatsu H, Uchino S, Ronco C, Bellomo R:
Impact of continuous veno-venous hemofiltration on acid-
base balance. Int J Artif Organs 2003, 26:19-25.
14. Kierdorf H, Sieberth HG: Continuous treatment modalities in
acute renal failure. Nephrol Dial Transplant 1995, 10:2001-
2008.
15. Heering P, Ivens K, Thumer O, Brause M, Grabensee B: Acid-
base balance and substitution fluid during continuous
hemofiltration. Kidney Int 1999, 56:s37-s40.
16. Davenport A, Will E, Davison AM: The effect of lactate-buffered
solutions on the acid-base status of patients with renal
failure. Nephrol Dial Transplant 1989, 4:800-804.
17. Mansell MA, Morgan SH, Moore L, Kong CH, Laker MF, Wing AJ:
Cardiovascular and acid-base effects of acetate and bicar-
bonate haemodialysis. Nephrol Dial Transplant 1987, 1:229-
232.
18. Saman S, Opie LH: Mechanism of reduction of action potential
duration of ventricular myocardium by exogenous lactate. J
Mol Cell Cardiol 1984, 10:659-662.
19. Morgera S, Heering P, Szentandrasi T, Manassa E, Heintzen M,
Willers R, Passlick-Deetjen J, Grabensee B: Comparison of a
lactate- versus acetate-based hemofiltration replacement
fluid in patients with acute renal failure. Renal Fail 1997, 19:
155-164.
20. Thomas AN, Guy JM, Kishen R, Geraghty IF, Bowles BJM,
Vadgama P: Comparison of lactate and bicarbonate buffered
haemofiltration fluids: use in critically ill patients. Nephrol Dial
Transplant 1997, 12:1212-1217.
21. Tan HK, Uchino S, Bellomo R: The acid-base effects of continu-
ous hemofiltration with lactate or bicarbonate buffered
replacement fluids. Int J Artif Organs 2003, 26:477-483.
22. Cohen RD, Iles RA: Lactic acidosis. Clin Endocrinol Metab
1980, 9:513-527.
23. Mehta RL, McDonald B, Aguilar M, Ward DM: Regional citrate
anticoagulation for continuous arteriovenous hemodialysis in
critically ill patients. Kidney Int 1990, 38:976-981.
24. Cole L, Bellomo R, Journois D, Davenport P, Baldwin I, Tipping P:
High volume hemofiltration in human septic shock. Intensive
Care Med 2001, 27:978-986.
25. Levraut J, Ciebera JP, Jambou P, Ichiai C, Labib Y, Grimaud D:
Effect of continuous veno-venous hemofiltration with dialysis
on lactate clearance in critically ill patients. Crit Care Med
1997, 25:58-62.
26. Cole L, Bellomo R, Baldwin I, Hayhoe M, Ronco C: The impact of
lactate-buffered high volume hemofiltration on acid-base
balance. Intensive Care Med 2003, 29:1113-1120.
27. Uchino S, Cole L, Morimatsu H, Goldsmith D, Ronco C, Bellomo
R: Solute mass balance during isovolaemic high volume
haemofiltration. Intensive Care Med 2003, 29:1541-1546.
508
A
TOT
= total amount of weak acids and proteins in plasma; ICU = intensive care unit; ISE = ion selective electrode; PCO
2
= partial carbon dioxide
tension; SBE = standard base excess; SID = strong ion difference; SIDa = apparent strong ion difference; SIDe = effective strong ion difference;
SIG = strong ion gap; Vd = volume of distribution.
Critical Care October 2005 Vol 9 No 5 Gunnerson
Abstract
Acidbase abnormalities are common in critically ill patients. Our
ability to describe acidbase disorders must be precise. Small
differences in corrections for anion gap, different types of analytical
processes, and the basic approach used to diagnose acidbase
aberrations can lead to markedly different interpretations and
treatment strategies for the same disorder. By applying a quantitive
acidbase approach, clinicians are able to account for small
changes in ion distribution that may have gone unrecognized with
traditional techniques of acidbase analysis. Outcome prediction
based on the quantitative approach remains controversial. This is in
part due to use of various technologies to measure acidbase
variables, administration of fluid or medication that can alter
acidbase results, and lack of standardized nomenclature. Without
controlling for these factors it is difficult to appreciate the full effect
that acidbase disorders have on patient outcomes, ultimately
making results of outcome studies hard to compare.
Introduction
Critically ill and injured patients commonly have disorders of
acidbase equilibrium. Acidosis may occur as a result of
increases in arterial partial carbon dioxide tension (PCO
2
;
respiratory acidosis) or from a variety organic or inorganic,
fixed acids (metabolic acidosis). There appears to be a
difference in physiologic variables and outcomes between
patients with respiratory acidosis and those with metabolic
acidosis [1,2], leading some investigators to hypothesize that
it is the cause of acidosis rather than the acidosis per se that
drives the association with clinical outcomes. Even though
metabolic acidosis is a common occurrence in the intensive
care unit (ICU), the precise incidence and prevalence of
metabolic acidosis has not been established for critically ill
patients. Often these disorders are markers for underlying
pathology. Although the true causeeffect relationship
between acidosis and adverse clinical outcomes remains
uncertain, metabolic acidosis remains a powerful marker of
poor prognosis in critically ill patients [3-5].
Common etiologies of metabolic acidosis include lactic
acidosis, hyperchloremic acidosis, renal failure, and ketones.
All types of metabolic acidosis have a contributing anion
responsible for the acidosis. Some causes may be obvious
with a single contributing anion, such as a pure lactate
acidosis, whereas other complex disorders may not have a
single and identifiable, causative anion and only the strong
ion gap (SIG) is elevated. There is recent evidence
suggesting that outcomes may be associated with the
predominant anion contributing to the metabolic acidosis.
In this review we use modern physical chemical analysis and
interpretation to describe why these acidbase disorders
occur, what is considered normal, and how variations in
analytical technology affect results. We also attempt to
describe the incidence between various etiologies of
acidbase disorders in ICU patients and examine whether
they might affect clinical outcomes. Finally, we discuss
limitations of the current nomenclature system, or the lack
thereof, with regard to acidbase definitions, and propose a
standard approach to describing physical chemical
influences on acidbase disorders.
The physical chemical approach
Critically ill patients commonly have acidbase disorders.
When applying evolving technology in analytical techniques
to measure acidbase variables, the quantitative acidbase
(or physical chemical) approach is slowly emerging as a
valuable tool in identifying the causative forces that drive
acidbase disorders [6]. This review is built on the physical
chemical approach (also referred to as the Stewart
Review
Clinical review: The meaning of acidbase abnormalities in the
intensive care unit epidemiology
Kyle J Gunnerson
Assistant Professor, The Virginia Commonwealth University Reanimation Engineering and Shock Center (VCURES) Laboratory, Departments of
Anesthesiology/Critical Care and Emergency Medicine, Virginia Commonwealth University Medical Center, Richmond, Virginia, USA
Corresponding author: Kyle Gunnerson, kgunnerson@vcu.edu
Published online: 10 August 2005 Critical Care 2005, 9:508-516 (DOI 10.1186/cc3796)
509
approach or the quantitative approach) to analyzing acid
base disorders, and there are many well written reviews that
detail the intricacies of these approaches [7-10].
Traditional approaches to the analysis of acidbase disorders
adapted from Henderson and Hasselbalch or those proposed
by Siggaard-Andersen and colleagues are inadequate for
appreciating causative mechanisms. These traditional approa-
ches may identify the presence of a metabolic acidosis, but
the categorization ends with a broad differential based on the
presence or absence of an anion gap. Controversy has existed
for many years over which approach to the analysis of
acidbase balance is more accurate, but in general the results
of these differing approaches are nearly identical [8,9,11].
The physical chemical approach allows the clinician to
quantify the causative ion. The basic principle of the physical
chemical approach revolves around three independent
variables: PCO
2
, strong ion difference (SID), and the total
amount of weak acids (A
TOT
). SID is the resulting net charge
of all of the strong ions. This includes both the cations (Na
+
,
K
+
, Ca
2+
, and Mg
2+
) and anions (Cl
and lactate). This

measurable difference is referred to as the apparent SID
(SIDa), with the understanding that not all ions may be
accounted for. In healthy humans this number is close to
+40 mEq/l [12]. The law of electroneutrality states that there
must be an equal and opposing charge to balance the
positive charge, and so the +40 mEq/l is balanced by an
equal negative force comprised mostly of weak acids (A
TOT
).
These weak acids include plasma proteins (predominately
albumin) and phosphates. The total charge of these must
equal the SIDa. The product of all of the measurable anions
contributing to the balancing negative charge is referred to as
the effective SID (SIDe). Theoretically, the SIDa and SIDe
should equal each other, but a small amount of unmeasurable
anions may be present, even in good health, and so the
resulting difference in healthy humans appears to be less
than 2 mEq/l [12].
The role played by plasma proteins, specifically albumin, in
acidbase balance is curiously neglected in the traditional
approaches. This has led to numerous controversies
regarding the usefulness of the anion gap [13] and the
classification of metabolic acidbase disorders [14]. Several
studies have supported the observation that a significant
number of abnormal anion gaps go unrecognized without
correction for the albumin level (which, in the critically ill, is
usually low) [14-16]. The importance of correcting the anion
gap for albumin is not limited to the adult population. Quite
the contrary, there is a high incidence of hypoalbuminemia in
pediatric patients who are critically ill, and the effect on anion
gap measurements are similar to those in the adult population
[17,18]. Hatherill and colleagues [18] demonstrated that,
when the anion gap is not corrected in critically ill pediatric
patients, approximately 10 mEq acid and up to 50% of
abnormally elevated anion gaps are missed.
What is normal?
Strong ion gap metabolic acidosis
The SIG can simply be described as the sum of unmeasured
ions. More specifically, it is the difference between the SIDa
and the SIDe. The SIG and traditional anion gap differ in the
sense that the traditional anion gap exists in a broad range
of normal values, whereas the SIG takes into account the
effect of a wider range of ions, including weak acids, and thus
should approach zero. Any residual charge represents
unmeasured ions and has been termed SIG [19]. Even
though this theoretical value of zero should exist for patients
who have no known acidbase abnormalities, a wide range
(013 mEq/l) has been reported in the literature [14,19-22].
In the USA ranges for SIG in survivors tend to be low and are
predictive of survival in critical illness [15,23]. However, in
England and Australia countries that routinely use gelatins
for resuscitation values of SIG have been reported as high
as 11 mEq/l in ICU survivors [20] and do not appear to be
predictive of outcome [20,24]. Gelatins are a class of colloid
plasma expanders that are comprised of negatively charged
polypeptides (mean molecular weight between 20 and
30 kDa) dissolved in a crystalloid solution commonly
comprised of 154 mEq sodium and 120 mEq chloride. These
negatively charged polypeptides have been shown to
contribute to both an increased anion gap [25] and SIG [26],
most likely due to their negative charge and relatively long
circulating half-life. Moreover, these high levels of SIG may be
seen in the absence of acidbase abnormalities using
traditional acidbase measurements (e.g. PCO
2
, standard
base excess [SBE], pH).
We recently compared quantitative acidbase variables
between healthy volunteers (control) and stable ICU
patients. There were significant differences between these
two groups. The control group had a SIDe (mean standard
deviation) of 40 3.8 mEq/l and SIG of 1.4 1.8 mEq/l. The
ICU patients had a SIDe of 33 5.6 mEq/l and a SIG of
5.1 2.9 mEq/l. The control group also had a higher albumin
level (4.5 g/dl versus 2.6 g/dl in the ICU group). Interestingly,
traditional acidbase variables (pH, PCO
2
, and SBE) were
similar between the groups [12]. Controversy remains, but it
appears that a normal range of SIG in healthy patients is
02 2 mEq/l, and in stable ICU patients without renal
failure SIG appears to be slightly higher, at 5 3 mEq/l.
The SIG calculation is somewhat cumbersome to use at the
bedside [19], and attempts have been made to simplify this
technique based on normalizing the anion gap for the serum
albumin, phosphate, and lactate concentrations [8,16,21,27].
By substituting the corrected anion gap in place of the SIG,
we found a strong correlation between the two (r
2
= 0.96)
[28]. The corrected anion gap was calculated as follows:
([Na
+
+ K
+
] [Cl
+ HCO
3
]) 2.0(albumin [g/dl])
0.5(phosphate [mg/dl]) lactate (mEq/l) [8]. An even simpler
formula (Na
+
+ K
+
) (Cl
+ HCO
3
) 2.5(albumin [g/dl])
lactate (mmol/l) for the corrected anion gap without the use
510
of phosphate can be used and retain a strong correlation with
SIG (r
2
= 0.93) [8,28]. For international units, the following
conversion can be substituted for albumin and phosphate:
0.2(albumin [g/l]) 1.5(phosphate [mmol/l]).
Hyperchloremic metabolic acidosis
One of the obstacles in identifying the incidence of
hyperchloremic metabolic acidosis is the actual definition
itself. There are many references to hyperchloremic metabolic
acidosis or dilutional acidosis in the literature, and there are
just as many definitions of hyperchloremic metabolic acidosis.
In fact, classifying hyperchloremia as a metabolic acidosis is
misleading because chloride is not a byproduct of meta-
bolism. This multitude of definitions is akin to the difficulty in
defining acute renal failure, for which more than 30 different
definitions have been reported in the literature [29]. It is more
common to base the diagnosis of hyperchloremic metabolic
acidosis on an absolute chloride value rather than to take into
account the physicochemical principles of either the
decreased ratio of sodium to chloride or the decreased
difference between them. With regard to plasma, the addition
of normal saline increases the value from baseline of chloride
more so than does sodium. This difference in the ratio of
sodium to chloride change is what is important. The increase
in chloride relative to that of sodium reduces the SID,
resulting in a reduction in the alkalinity of blood. The Na
+
/Cl
ratio has been proposed as a simple way to delineate the

contribution of chloride to the degree metabolic acidosis
[30]. In other words, euchloremia or normal chloride is
completely dependent on the concentration of sodium. In this
sense, chloride must always be interpreted with the sodium
value because they both change with respect to the patients
volume status and the composition of intravenous fluids.
For example, a 70 kg person has 60% total body water and
a serum Na
+
of 140 mEq/l and Cl
of 100 mEq/l, resulting

in a SIDa of approximately 40 mEq/l. This patient is now
given 10 l saline (154 mEq of both Na
+
and Cl
) over the
course of his resuscitation. Accounting for his volume of
distribution (Vd), the serum Na
+
would increase only to
143 mEq/l but the Cl
would increase to 111 mEq/l.

Although the true Vd of Cl
is extracellular fluid, the

movement of salt and water together creates an effective
Vd equal to that of total body water [31]. The SBE would
decrease at a similar rate but the Cl
would be regarded as
normal range on most analyzers. In spite of the normal
absolute reading of Cl
, the patient has had a reduction in

SIDa from 40 mEq/l to 32 mEq/l. This patient now has a
hyperchloremic metabolic acidosis with a normal absolute
value of chloride, and thus would likely be overlooked by
applying traditional principles and nomenclature.
Regardless of how it is diagnosed, hyperchloremic
metabolic acidosis is common in critically ill patients, is
most likely iatrogenic, and surprisingly remains controversial
regarding the cause of the acidosis (strong ion addition
[chloride] versus bicarbonate dilution) [32,33].
Lactic acidosis
Lactic acidosis is a concerning pathophysiologic state for
critically ill patients, and there is a wealth of literature
reporting on the significance of various etiologies of elevated
lactate as it pertains to the critically ill patient [34-36]. During
basal metabolic conditions, arterial lactate levels exist in a
range between 0.5 and 1 mEq/l. Levels may be higher in
hypoperfused or hypoxic states. However, critically ill patients
may have conditions other than hypoperfusion that may lead
to lactate elevations, such as increased catecholamine
production in sepsis or trauma [37] or from production by
lung in acute lung injury [38,39].
Even though elevated lactate levels can be a sign of
underlying pathology, most patients in the ICU do not have
elevated lactate levels. Five recent outcome trials comparing
various approaches in diagnosing acidbase disorders had
relatively low mean lactate levels: 2.7 mEq/l in survivors [40];
1.88 mEq/l [24]; 1.0 mEq/l [30]; 2.3 mEq/l in survivors [20];
and 3.1 mEq/l [15]. In a cohort of 851 ICU patients with a
suspected lactic acidosis, and using the highest lactate value
if there were multiple values, the mean lactate level was still
only 5.7 mEq/l [28]. Therefore, when an elevated lactate is
present, it should not be dismissed without further
investigation into the underlying etiology.
Outcome data: does the type of acidosis
matter?
Metabolic acidosis may represent an overall poor prognosis,
but does this relationship exist among the various types of
metabolic acidosis? Lactic acidosis has garnered
considerable attention in critically ill patients, but metabolic
acidosis may result from a variety of conditions other than
those that generate lactate [8]. The existing literature does
not suggest a strong relationship between the type of
acidosis and outcome. However, traditional methods of
classifying and analyzing acidbase abnormalities have
significant limitations, especially in critically ill patients [13].
Studies have usually failed to identify the effects that
causative anions (lactate, chloride, and others) have on the
resulting pH and SBE. Findings are typically reported as
either nonlactate metabolic acidosis or anion gap metabolic
acidosis, without identifying a predominant source. These are
major limitations of the traditional approach.
A large, retrospective analysis of critically ill patients in which
clinicians suspected the presence of lactic acidosis [28]
revealed that differing etiologies of metabolic acidosis were in
fact associated with different mortality rates. It also appeared
that a varying distribution of mortality, within these subgroups
of metabolic acidoses existed between different ICU patient
populations (Fig. 1). The study suggests that the effects of
metabolic acidosis may vary depending on the causative ion.
Conflicting relationships have been reported between
acidbase abnormalities, their treatment, and outcomes in
511
critically ill patients [15,20,23,24,40,41]. Some studies have
suggested an independent association between low pH or
SBE and mortality [42-44], whereas others have not [4,15].
We address further the impact that three major classifications
of metabolic acidosis have on patient outcome.
Hyperchloremic metabolic acidosis
Even though many causes of metabolic acidosis may be
unavoidable, often the source of metabolic acidosis is
iatrogenic. In critically ill patients a common cause is related
to the volume of saline infused during resuscitation from
shock. Large volume saline infusion produces metabolic
acidosis by increasing the plasma Cl
concentration relative
to the plasma Na
+
concentration [45-48]. This results in a
decreased SID (the difference between positive and negative
charged electrolytes), which in turn produces an increase in
free H
+
ions in order to preserve electrical neutrality [8]. The
clinical effects of these changes have been documented over
the past several years.
The consequences of hyperchloremic metabolic acidosis are
traditionally downplayed and accepted as a necessary evil of
saline resuscitation. However, recent studies may change this
benign view of iatrogenic hyperchloremic metabolic acidosis,
especially as it pertains to choice of fluid composition for
resuscitation. Deusch and Kozek-Langenecker [49] recently
demonstrated better platelet function in vitro when samples
of whole blood were diluted with a hetastarch prepared in a
balanced electrolyte solution instead of using saline as the
solvent. In the same study, similar results were observed
when the starch molecule was removed and the samples
were diluted with either a balanced electrolyte solution or
0.9% saline. This supports the hypothesis that the electrolyte
composition of the solution may play a role in the
coagulopathy associated with starch solutions greater than
that of the starch molecule itself. Wilkes and colleagues [50]
also demonstrated an increase in adverse events and worse
acidbase balance when comparing similar hetastarch based
solutions prepared in either a saline solution or balanced
electrolyte solution. Gan and coworkers [51] reported similar
findings in large volume resuscitation in major surgery
comparing hetastarch prepared in a balanced electrolyte
solution or in saline, and similar findings were reported by
Williams and colleagues [52] when they compared lactated
Ringers with 0.9% saline. In all of these studies, saline fared
worse than did balanced electrolyte solutions.
Saline induced acidosis has a side effect profile similar to that
of ammonium chloride. This includes abdominal pain, nausea,
vomiting, headache, thirst, hyperventilation, and delayed
urination [53,54]. This striking similarity may be related to the
chloride concentration. Aside from avoiding these adverse
reactions, the treatment of metabolic acidosis per se has not
yet been shown to improve clinical outcome [41] and, based
on a large retrospective database [28], mortality does not
appear to be significantly increased. However, there is
mounting evidence that iatrogenic metabolic acidosis may be
harmful and should be avoided when possible.
Lactic acidosis
Much interest has been directed at lactate metabolism and its
role in metabolic acidosis in critically ill patients since the first
description of lactate associated with circulatory shock [55].
It has also been the focus of several recent reviews
[34,35,56,57]. An early approach to the broad classification
of elevated lactate levels based on the presence (type A) or
absence (type B) of hypoperfusion was described by Cohen
and Woods [58] in their classic monogram. Contemporary
understanding of the complexity of lactate production and
metabolism in critical illness has practically relegated this
classification system to that of a historical one [56].
Our improved understanding of the complexities of lactate
metabolism has fueled the controversy regarding lactates role
in the care of critically ill patients. Aside from hypoperfusion
leading to cellular dysoxia, elevated lactate has been
associated with a number of common cellular processes that
are present in critical illness. These include increased activity
of Na
+
/K
+
-ATPase in normoxia [59], increased pyruvate and
lactate due to increased aerobic glycolysis [60], and
decreased lactate clearance [61], to name but a few.
Regardless of the etiology, lactic acidosis has been
associated with worse outcomes in critically ill patients.
Elevated lactate has been associated with oxygen debt since
Figure 1
Distribution of patients and contributing ion responsible for majority of
metabolic acidosis present. Shown is the distribution of patients within
different types of intensive care unit (ICU) locations and their
respective hospital mortality associated with the major ion contributing
to the metabolic acidosis. These results were obtained from a large
teaching institution comprised of two hospitals and seven ICUs over a
1 year period and included patients with a suspected lactic acidosis.
No metabolic acidosis is defined as a standard base excess of
2 mEq/l or higher. CCU, cardiac (nonsurgical) ICU; CTICU,
cardiothoracic ICU; LTICU, liver transplant ICU; Med, medical ICU;
Neuro, neurosurgical and neurological ICU; Surg, general surgical ICU;
Trauma, trauma ICU.
512
the 1930s [62] and has been associated with poor outcome
since the 1960s [3,63-65]. Elevated lactate on presentation
[65] and serial measurements [36,66] are both associated
with worse outcome. More importantly, the ability to clear
lactate rapidly has been associated with improved mortality
[67-69]. Although our understanding of the metabolism of
lactate has greatly improved since these early studies [56],
critically ill patients with elevated lactate levels continue to
have worse outcomes than those who do not [35,36,69].
Recent goal-directed strategies incorporating lactate either
as an acute marker for acuity [70] or as an end-point of
resuscitation [71] have been shown to improve mortality.
Strong ion gap metabolic acidosis
Lactate serves not only as a marker for severity or an end-point
of resuscitation but also as an important variable in the
quantification and determination of the primary etiology of a
metabolic acidosis. In the presence of a metabolic acidosis and
a normal lactate and SIDa, the resulting charge balance must
be composed of unmeasured anions (SIG). There is still much
debate as to how well SIG acidosis predicts mortality
[15,20,23,24]. The ability of SIG to predict mortality in the
critically ill is not as clear as that of lactate. There have been
varying findings regarding absolute values and the significance
of all quantitative acidbase variables, especially SIG. It
appears that a pattern is emerging in which studies conducted
in different countries have shown different baseline levels of
SIG and have noted differences in their clinical significance
[15,20,23,24,40]. This may be related to the technology used
to measure acidbase variables [72-74] or administration of
medications or fluid (e.g. gelatins) [25,26] that alter the SIG.
Two recent prospective studies [23,40] controlled for the
limitations noted above when evaluating the ability of the SIG
to predict mortality. The findings of these two studies are
unique in the sense that they are the first reports of SIG
predicting mortality in patients with trauma [23] and severe
malaria [40]. Acidbase variables were measured, in both
studies, before any significant amount of volume resuscitation.
Kaplan and Kellum [23] evaluated the relationship between
SIG, before significant fluid resuscitation, and mortality. In
patients with major vascular injury requiring surgery, a SIG in
excess of 5 mEq/l was predictive of mortality. Interestingly,
SIG outperformed lactate as a predictor of mortality based on
receiver operator curve characteristics. SIG was also a
stronger predictor of mortality than was the Injury Severity
Score, based on multivariate logistic regression analysis.
Nonsurvivors had a mean SIG above 10 mEq/l. These levels
of unmeasured anions were generated in the absence of
resuscitative fluids known to contribute to unmeasured
anions such as gelatin based solutions, which are not used
for resuscitation in the USA. This important study supports
the hypothesis that SIG may be a rapidly accumulating
biomarker that reflects severity of injury or illness, similar to
other acute phase proteins.
Dondorp and colleagues [40] evaluated the relationship
between SIG and mortality in critically ill patients diagnosed
with severe malaria. Severe falciparum malaria is frequently
associated with metabolic acidosis and hyperlactatemia. The
etiology of both of these conditions has been thought to be
based on both hepatic dysfunction and hypoperfusion. The
authors found that even in fatal cases of this disease state,
the predominant form of metabolic acidosis was not lactate
but rather unaccounted anion, or SIG, acidosis. Mean lactate
levels were surprisingly low in both survivors (2.7 mEq/l) and
nonsurvivors (4.0 mEq/l), whereas SIG levels were elevated
in both (9.7 mEq/l and 15.9 mEq/l, respectively). SIG was
also a strong predictor of mortality in this study.
The overall value of SIG as a predictor of mortality is yet to be
determined. Future studies that control for technology and
the composition of resuscitative fluids are required. Regard-
less of the etiology of these anions, our understanding of the
importance of SIG is rapidly evolving.
Technology limitations
Technologic advances in the measurement of electrolytes
have an influence on how quantitive acidbase parameters
are calculated. Currently, there are three techniques
commonly used to measure quantitive acidbase variables:
flame photometry and potentiometry using direct ion selective
electrodes (ISEs) or indirect ISEs. Flame photometry is used
infrequently in developed countries. It is the measurement of
the wavelength of light rays emitted by excited metallic
electrons exposed to the heat energy of a flame. The intensity
of the emitted light is proportional to the concentration of
atoms in the fluid, such that a quantitative analysis can be
made on this basis. Examples are the measurements of
sodium, potassium, and calcium. The sample is dispersed
into a flame from which the metal ions draw sufficient energy
to become excited. On returning to the ground state, energy
is emitted as electromagnetic radiation in the visible part of
the spectrum, usually as a very narrow wavelength band (e.g.
sodium emits orange light, potassium purple, and calcium
red). The radiation is filtered to remove unwanted wave-
lengths and the resultant intensity measured. Thus, the total
concentration of the ion is measured.
Flame photometry has several limitations, one of the more
common being the influence of blood solids (lipids). These
lipids have been shown to interfere with the optical sensing
(due to increased turbidity) and by causing short sampling
errors (underestimating true sample volume) [75]. Flame
photometry also measures the concentration of ions, both
bound and unbound, whereas newer techniques (ISEs)
measure the disassociated form (or active form) of the ion.
An ISE measures the potential of a specific ion in solution,
even in the presence of other ions. This potential is measured
against a stable reference electrode of constant potential. By
measuring the electric potential generated across a
513
membrane by selected ions and comparing it with a reference
electrode, a net charge is determined. The strength of this
charge is directly proportional to the concentration of the
selected ion. The major advantage that ISEs have over flame
photometry is that ISEs do not measure the concentration of an
ion; rather, they measure its activity. Ionic activity has a specific
thermodynamic definition, but for most purposes it can be
regarded as the concentration of free ion in solution.
Because potentiometry measures the activity of the ion at the
electrode surface, the measurement is independent of the
volume of the sample, unlike flame photometry. In indirect
potentiometry, the concentration of ion is diluted to an activity
near unity. Because the concentration will take into account
the original volume and dilution factor, any excluded volume
(lipids, proteins) introduces an error (usually insignificant).
When a specimen contains very large amounts of lipid or
protein, the dilutional error in indirect potentiometric methods
can become significant. A classic example of this is seen with
hyperlipidemia and hyperproteinemia resulting in a pseudo-
hyponatremia by indirect potentiometry. However, direct
potentiometry will reveal the true sodium concentration
(activity). This technology (direct potentiometry) is commonly
used in blood gas analyzers and point-of-care electrolyte
analyzers. Indirect ISE is commonly used in the large, so-
called chemistry analyzers located in the central laboratory.
However, there are some centralized analyzers utilizing direct
ISE. The methodologies can produce significantly different
results [72-74,76].
Recent evidence reinforces how technology used to measure
acidbase variables affects results and may affect inter-
pretation of clinical studies. Morimatsu and colleagues [77]
have demonstrated a significant difference between a point-
of-care analysis and the central laboratory in detecting
sodium and chloride values. These differences ultimately
affect the quantitative acidbase measurements. The study
emphasizes that differences in results may be based on
technology rather than pathophysiology. One reason may be
related to the improving technology of chloride and sodium
specific probes. On a similar note, it also appears that there
is variation in the way in which the blood gas analyzers
calculate base excess [78].
Unfortunately, many studies evaluating acidbase balance
have failed to report details of the technology used to
measure these variables. This limitation was discussed by
Rocktaeschel and colleagues [24] in 2003. Since then,
detailed methods sections that include specific electrode
technology have become more common when acidbase
disorders are evaluated [23,40,79,80].
Incidence of metabolic acidosis in the
intensive care unit
The incidence of metabolic acidosis in the ICU is difficult to
extrapolate from the current literature. It is even harder to find
solid epidemiology data on the various types of metabolic
acidosis. A major hurdle is the various definitions used to
describe the types of acidbase disorder. The development
and implementation of the physical chemical approach has
made identifying the etiology of acidbase abnormalities
possible. Even though we can quantify these abnormalities, a
classification system has yet to be developed. The literature is
full of pre-Stewart acidbase descriptions, but the major
Table 1
Summary of quantitative acidbase studies in critically ill patients and the distribution of type of metabolic acidosis
Patient Sample Metabolic Unmeasured
Ref. population size acidosis acids Lactate Chloride Mixed
[30] Pediatric 540 samples 230 (45.5%) 120 (52%) M 22 (9.6%) M 88 (38.2%) M 57 (25%) M
ICU patients (282 patients)
a
44 base deficit
[80] Pediatric ICU 150 samples
a
24 anion gap 44 6 19 10
post-cardiac (44 patients)
a
57 anion gap
surgery corrected
[15] Pediatric ICU, 255 patients 69 (27%) 55 (79.7%) M N/A N/A N/A
patients only with
acidbase measurements
[79] Pediatric ICU 46 patients 42 (91%) 33 (72%) M 39 (85%) M 29 (63%) M N/A
in shock
[21] Adult ICU with 50 patients 50 (100%) 49 (98%) M,T 31 (62%) M,T 40 (80%) M,T N/A
met acidosis
[28] Adult ICU with 851 patients 548 (64%) T 204 (37%) M 239 (44%) M 105 (19%) M N/A
suspicion of lactic
acidosis (highest lactate used)
a
Authors defined metabolic acidosis using three different techniques; measurement of other variables by quantitive approach. M, the percentage of
the samples with a metabolic acidosis; T, the percentage of the total number (n) of patients.
514
taxonomy of metabolic acidoses was limited either to the
presence or to the absence of an anion gap, which also has
major limitations. Even when reviewing the quantitative
acidbase literature specifically, there is no agreement on
how to classify patients with metabolic acidosis.
In a retrospective review of 851 ICU patients, we classified
patients into categories representing the predominant
causative anion associated with the metabolic acidosis [28].
However, others simply reported absolute values of SID, SIG,
chloride, anion gap, and SBE in association with mortality
prediction rather than attempting to classify various subtypes
of metabolic acidosis [15,20,24]. Still others used a
combination of quantitative acidbase variables and the
sodium/chloride ratio [30] or absolute chloride levels [21,80]
to further classify disorders. Table 1 summarizes several
recent studies using the same physical chemical approach to
address acidbase disorders. Even though the authors all
applied the same methodology to identify acidbase
disorders, each one used different classification schemes to
describe the acidbase state. The absence of a uniform
classification system and different study designs limit our
ability to appreciate fully the incidence of the various
acidbase categories. For example, the incidence of
unmeasured anions contributing to metabolic acidosis ranged
from 37% to 98%. Lactate as the major contributing ion had
an even wider distribution, from almost 10% to 85%. Until the
nomenclature can become standardized, the true incidence
of acidbase disorders may never be fully appreciated.
We recommend the use of a classification system that is
based on physicochemical principles and the predominant
anion responsible for the acidosis (Fig. 2). In this system,
metabolic acidosis is defined as a SBE below 2 mEq/l;
lactic acidosis is an acidosis in which lactate accounts for
more than 50% of the SBE; in SIG acidosis the SIG
(unmeasured ions) accounts for more than 50% of SBE (in
the absence of lactic acidosis); and hyperchloremic
acidosis is defined a SBE below 2 mEq/l that is not
accounted for by lactate or SIG. As one can see, an
absolute level of chloride was not used for the definition of
hyperchloremic acidosis because it is the relative
relationship between the sodium and chloride
concentrations that contribute to the SIDa, which is one of
the independent variables that comprise acidbase
equilibria. Therefore, if a metabolic acidosis is present and
the SIG or lactate does not make up the majority of the acid
load, then the only strong ion left is chloride. For example,
let us consider a scenario in which the SBE is 8 mEq/l,
lactate is 2 mEq/l, and SIG is 2 mEq/l. In this scenario,
lactate and SIG together account for only 50% of all of the
() charges, as represented by the SBE of 8 mEq/l. There
remain 4 mEq/l of unaccounted anions that would be
explained by a proportional excess of Cl
in relation to Na
+
.
Thus, the final classification would be hyperchloremic
metabolic acidosis, regardless of the absolute Cl
level.
This classification system will serve two major purposes. First,
we will have a way to describe consistently the predominant
anion that drives the acidbase status. This may potentially
contribute to a clearer understanding of the underlying
pathology. Second, by using the quantitative approach, the
clinician can still recognize a sizeable contribution of other
anions, regardless of the predominate anion. An example
would be that of a patient with a predominant hyperchloremic
metabolic acidosis but with a substantial amount of
unaccounted anions (SIG), even though SIG may not
account for more than 50% of the SBE. In this case, the
clinician may consider whether to pursue a possible
diagnosis of concomitant ethylene glycol toxicity (or other
unmeasured anions) along with the hyperchloremia.
Our classification scheme leaves open the possibility that a
combined lactic and SIG acidosis could be misclassified as
Figure 2
Proposed metabolic acidosis classification flow diagram based on the
contributing anion group. This flow diagram is one proposed way to
classify metabolic acidosis based on the major contributing anion
group. The definition of metabolic acidosis component is a standard
base excess (SBE) below 2 mEq/l. It is not based on pH because of
the possibility of respiratory compensation. SIDa, apparent strong ion
difference; SIDe, effective strong ion difference; SIG, strong ion gap.
515
hyperchloremic. Conversely, some cases of hyperchloremic
acidosis could also be misclassified as either SIG or lactic
acidosis if pre-existing or concomitant metabolic alkalosis
was also present, reducing the apparent impact of chloride.
However, these limitations exist with any acidbase
classification scheme, and given that hyperchloremic acidosis
is defined on the basis of acidosis without an anion gap,
rather than on the basis of chloride levels, some imprecision
is always going to be present.
Conclusion
Acidbase disorders in critically ill patients are common.
Traditional approaches used to measure acidbase disorders
may actually underestimate their presence. Currently, the
relationship between metabolic acidosis and clinical outcome
remains uncertain, but it appears that a difference in mortality
may depend on the varying contribution of causative anions.
Major limitations in the interpretation of current literature
evaluating outcomes can be condensed into three areas:
varying results based on technologic differences between
flame photometry, indirect ISEs, and direct ISEs; lack of
consistent nomenclature classifying subgroups of metabolic
acidosis; and confounding of results by administration of
medications or fluids used for resuscitation that will
exogenously elevate the SIG (e.g. gelatins). These limitations
can and should be addressed in future study designs.
Without consistency in reporting acidbase methodology,
conflicting reports will continue.
Competing interests
References
1. Kellum JA, Song M, Subramanian S: Acidemia: good, bad or
inconsequential? In Yearbook of Intensive Care and Emergency
Medicine. Edited by Vincent JL. Berlin: Springer; 2002:510-516.
2. Li J, Hoskote A, Hickey C, Stephens D, Bohn D, Holtby H, Van
Arsdell G, Redington AN, Adata I: Effect of carbon dioxide on
systemic oxygenation, oxygen consumption, and blood lactate
levels after bidirectional superior cavopulmonary anastomo-
sis. Crit Care Med 2005, 33:984-989.
3. Broder G, Weil MH: Excess lactate: an index of reversibility of
shock in human patients. Science 1964, 143:1457.
4. Hickling KG, Walsh J, Henderson S, Jackson R: Low mortality
rate in adult respiratory distress syndrome using low-volume,
pressure-limited ventilation with permissive hypercapnia: a
prospective study. Crit Care Med 1994, 22:1568-1578.
5. Stacpoole PW, Lorenz AC, Thomas RG, Harman EM:
Dichloroacetate in the treatment of lactic acidosis. Ann Intern
Med 1988, 108:58-63.
6. Gunnerson KJ, Kellum JA: Acid-base and electrolyte analysis in
critically ill patients: are we ready for the new millennium?
Curr Opin Crit Care 2003, 9:468-473.
7. Corey HE: Stewart and beyond: new models of acid-base
Crit Care 2000, 4:6-14.
9. Stewart P: Modern quantitative acid-base chemistry. Can J
10. Stewart PA: How to Understand Acid-base. A Quantitative Acid-
base Primer for Biology and Medicine. New York: Elsevier; 1981.
11. Sirker AA, Rhodes A, Grounds RM, Bennett ED: Acid-base phys-
iology: the traditional and the modern approaches. Anaes-
thesia 2002, 57:348-356.
12. Gunnerson KJ, Roberts G, Kellum JA: What is a normal strong
ion gap (SIG) in healthy subjects and critically ill patients
without acid-base abnormalities? [abstract]. Crit Care Med
2003, Suppl 12:A111.
13. Salem MM, Mujais SK: Gaps in the anion gap. Arch Intern Med
1992, 152:1625-1629.
Care Med 2000, 162:2246-2251.
15. Balasubramanyan N, Havens PL, Hoffman GM: Unmeasured
anions identified by the Fencl-Stewart method predict mortal-
ity better than base excess, anion gap, and lactate in patients
in the pediatric intensive care unit. Crit Care Med 1999, 27:
1577-1581.
16. Story DA, Poustie S, Bellomo R: Estimating unmeasured anions
in critically ill patients: anion-gap, base-deficit, and strong-ion-
gap. Anaesthesia 2002, 57:1109-1114.
17. Durward A, Mayer A, Skellett S, Taylor D, Hanna S, Tibby SM,
Murdoch IA: Hypoalbuminaemia in critically ill children: inci-
dence, prognosis, and influence on the anion gap. Arch Dis
Child 2003, 88:419-422.
18. Hatherill M, Waggie Z, Purves L, Reynolds L, Argent A: Correc-
tion of the anion gap for albumin in order to detect occult
tissue anions in shock. Arch Dis Child 2002, 87:526-529.
19. Kellum JA, Kramer DJ, Pinsky MR: Strong ion gap: a methodol-
ogy for exploring unexplained anions. J Crit Care 1995, 10:51-
55.
20. Cusack RJ, Rhodes A, Lochhead P, Jordan B, Perry S, Ball JA,
Grounds RM, Bennett ED: The strong ion gap does not have
prognostic value in critically ill patients in a mixed
medical/surgical adult ICU. Intensive Care Med 2002, 28:864-
869.
21. Moviat M, van Haren F, van der HH: Conventional or physico-
chemical approach in intensive care unit patients with meta-
bolic acidosis. Crit Care 2003, 7:R41-R45.
22. Wilkes P: Hypoproteinemia, strong-ion difference, and acid-
1740-1748.
23. Kaplan LJ, Kellum JA: Initial pH, base deficit, lactate, anion gap,
strong ion difference, and strong ion gap predict outcome
from major vascular injury. Crit Care Med 2004, 32:1120-
1124.
24. Rocktaeschel J, Morimatsu H, Uchino S, Bellomo R: Unmea-
sured anions in critically ill patients: can they predict mortal-
ity? Crit Care Med 2003, 31:2131-2136.
25. Sumpelmann R, Schurholz T, Marx G, Thorns E, Zander R: Alter-
ation of anion gap during almost total plasma replacement
with synthetic colloids in piglets. Intensive Care Med 1999, 25:
1287-1290.
Care Med 1999, 25:680-685.
28. Gunnerson KJ, Saul M, Kellum JA: Lactic versus non-lactic
metabolic acidosis: outcomes in critically ill patients.
[abstract]. Crit Care 2003, Suppl 2:S8.
29. Bellomo R, Ronco C, Kellum JA, Mehta RL, Palevsky P: Acute
renal failure definition, outcome measures, animal models,
fluid therapy and information technology needs: the Second
International Consensus Conference of the Acute Dialysis
Quality Initiative (ADQI) Group. Crit Care 2004, 8:R204-R212.
30. Durward A, Skellett S, Mayer A, Taylor D, Tibby SM, Murdoch IA:
The value of the chloride: sodium ratio in differentiating the
aetiology of metabolic acidosis. Intensive Care Med 2001, 27:
828-835.
Shock 1998, 9:364-368.
32. Kellum JA: Saline-induced hyperchloremic metabolic acidosis.
Crit Care Med 2002, 30:259-261.
33. Prough DS: Acidosis associated with perioperative saline
administration: dilution or delusion? Anesthesiol 2000, 93:
1167-1169.
34. De Backer D: Lactic acidosis. Minerva Anestesiol 2003, 69:281-
284.
516
35. Luft FC: Lactic acidosis update for critical care clinicians. J Am
Soc Nephrol 2001, Suppl 17:S15-S19.
36. Vincent JL, Dufaye P, Berre J, Leeman M, Degaute JP, Kahn RJ:
Serial lactate determinations during circulatory shock. Crit
Care Med 1983, 11:449-451.
37. James JH, Luchette FA, McCarter FD, Fischer JE: Lactate is an
unreliable indicator of tissue hypoxia in injury or sepsis.
Lancet 1999, 354:505-508.
38. Bellomo R, Kellum JA, Pinsky MR: Transvisceral lactate fluxes
during early endotoxemia. Chest 1996, 110:198-204.
39. De Backer D, Creteur J, Zhang H, Norrenberg M, Vincent JL:
Lactate production by the lungs in acute lung injury. Am J
Respir Crit Care Med 1997, 156:1099-1104.
40. Dondorp AM, Chau TT, Phu NH, Mai NT, Loc PP, Chuong LV,
Sinh DX, Taylor A, Hien TT, White NJ, et al.: Unidentified acids
of strong prognostic significance in severe malaria. Crit Care
Med 2004, 32:1683-1688.
42. Davis JW, Parks SN, Kaups KL, Gladen HE, ODonnell-Nicol S:
Admission base deficit predicts transfusion requirements and
risk of complications. J Trauma 1996, 41:769-774.
43. Dunham CM, Siegel JH, Weireter L, Fabian M, Goodarzi S,
Guadalupi P, Gettings L, Linberg SE, Very TC: Oxygen debt and
metabolic acidemia as quantitative predictors of mortality and
the severity of the ischemic insult in hemorrhagic shock. Crit
Care Med 1991, 19:231-243.
44. Smith I, Kumar P, Molloy S, Rhodes A, Newman PJ, Grounds RM,
Bennett ED: Base excess and lactate as prognostic indicators
for patients admitted to intensive care. Intensive Care Med
2001, 27:74-83.
Shock 1998, 9:364-368.
46. Rehm M, Orth V, Scheingraber S, Kreimeier U, Brechtelsbauer H,
Finsterer U: Acid-base changes caused by 5% albumin versus
6% hydroxyethyl starch solution in patients undergoing acute
normovolemic hemodilution: a randomized prospective study.
Anesthesiol 2000, 93:1174-1183.
going gynecologic surgery. Anesthesiol 1999, 90:1265-1270.
48. Waters JH, Miller LR, Clack S, Kim JV: Cause of metabolic aci-
dosis in prolonged surgery. Crit Care Med 1999, 27:2142-
2146.
49. Deusch E, Kozek-Langenecker S: Effects of hydroxyethyl starch
and calcium on platelet activation. Anesth Analg 2005, 100:
1538-1539.
50. Wilkes NJ, Woolf R, Mutch M, Mallett SV, Peachey T, Stephens R,
Mythen MG: The effects of balanced versus saline-based het-
astarch and crystalloid solutions on acid-base and electrolyte
status and gastric mucosal perfusion in elderly surgical
patients. Anesth Analg 2001, 93:811-816.
51. Gan TJ, Bennett-Guerrero E, Phillips-Bute B, Wakeling H,
Moskowitz DM, Olufolabi Y, Konstadt SN, Bradford C, Glass PS,
Machin SJ, et al.: Hextend, a physiologically balanced plasma
expander for large volume use in major surgery: a random-
ized phase III clinical trial. Hextend Study Group. Anesth Analg
1999, 88:992-998.
52. Williams EL, Hildebrand KL, McCormick SA, Bedel MJ: The
effect of intravenous lactated Ringers solution versus 0.9%
sodium chloride solution on serum osmolality in human vol-
unteers. Anesth Analg 1999, 88:999-1003.
53. Bushinsky DA, Coe FL: Hyperkalemia during acute ammonium
chloride acidosis in man. Nephron 1985, 40:38.
54. Wilcox CS: Regulation of renal blood flow by plasma chloride.
J Clin Invest 1983, 71:726-735.
55. Meakins J, Long C: Oxygen consumption, oxygen debt and
lactic acid in circulatory failure. J Clin Invest 1927, 4:273.
56. Gladden LB: Lactate metabolism: a new paradigm for the third
millennium. J Physiol 2004, 558:5-30.
57. Pittard AJ: Does blood lactate measurement have a role in the
management of the critically ill patient? Ann Clin Biochem
1999, 36:401-407.
58. Cohen R, Woods H: The clinical presentations and classifica-
tions of lactic acidosis. In Clinical and Biochemical Aspects of
Lactic Acidosis. Edited by Cohen R, Woods H. Boston: Blackwell
Scientific Publications; 1976:40-76.
59. James JH, Fang CH, Schrantz SJ, Hasselgren PO, Paul RJ,
Fischer JE: Linkage of aerobic glycolysis to sodium-potassium
transport in rat skeletal muscle. Implications for increased
muscle lactate production in sepsis. J Clin Invest 1996, 98:
2388-2397.
60. Gore DC, Jahoor F, Hibbert JM, DeMaria EJ: Lactic acidosis
during sepsis is related to increased pyruvate production, not
deficits in tissue oxygen availability. Ann Surg 1996, 224:97-
102.
61. Levraut J, Ciebiera JP, Chave S, Rabary O, Jambou P, Carles M,
Grimaud D: Mild hyperlactatemia in stable septic patients is
due to impaired lactate clearance rather than overproduction.
Am J Respir Crit Care Med 1998, 157:1021-1026.
62. Margaria R, Edwards R, Dill D: The possible mechanisms of
contracting and paying the oxygen debt and the role of lactic
acid in muscular contraction. Am J Physiol 1933, 106:689-715.
63. Cowley RA, Attar S, LaBrosse E, McLaughlin J, Scanlan E,
Wheeler S, Hanashiro P, Grumberg I, Vitek V, Mansberger A, et
al.: Some significant biochemical parameters found in 300
shock patients. J Trauma 1969, 9:926-938.
64. Schweizer O, Howland WS: Prognostic significance of high
lactate levels. Anesth Analg 1968, 47:383-388.
65. Weil MH, Afifi AA: Experimental and clinical studies on lactate
and pyruvate as indicators of the severity of acute circulatory
failure (shock). Circulation 1970, 41:989-1001.
66. Bakker J, Gris P, Coffernils M, Kahn RJ, Vincent JL: Serial blood
lactate levels can predict the development of multiple organ
failure following septic shock. Am J Surg 1996, 171:221-226.
67. Abramson D, Scalea TM, Hitchcock R, Trooskin SZ, Henry SM,
Greenspan J: Lactate clearance and survival following injury. J
Trauma 1993, 35:584-588.
68. Bakker J, Coffernils M, Leon M, Gris P, Vincent JL: Blood lactate
levels are superior to oxygen-derived variables in predicting
outcome in human septic shock. Chest 1991, 99:956-962.
69. Nguyen HB, Rivers EP, Knoblich BP, Jacobsen G, Muzzin A,
Ressler JA, Tomlanovich MC: Early lactate clearance is associ-
ated with improved outcome in severe sepsis and septic
shock. Crit Care Med 2004, 32:1637-1642.
70. Rivers E, Nguyen B, Havstad S, Ressler J, Muzzin A, Knoblich B,
Peterson E, Tomlanovich M; for the Early Goal-Directed Therapy
Collaborative Group: Early goal-directed therapy in the treat-
ment of severe sepsis and septic shock. N Engl J Med 2001,
345:1368-1377.
71. Rossi AF, Khan DM, Hannan R, Bolivar J, Zaidenweber M, Burke
R: Goal-directed medical therapy and point-of-care testing
improve outcomes after congenital heart surgery. Intensive
Care Med 2005, 31:98-104.
72. Burns RF, Russell LJ: Ion-selective electrode technology: an
overview. Contemp Issues Clin Biochem 1985, 2:121-130.
73. Fogh-Andersen N, Wimberley PD, Thode J, Siggaard-Andersen
O: Determination of sodium and potassium with ion-selective
electrodes. Clin Chem 1984, 30:433-436.
74. Worth HG: A comparison of the measurement of sodium and
potassium by flame photometry and ion-selective electrode.
Ann Clin Biochem 1985, 22:343-350.
75. Artiss JD, Zak B: Problems with measurements caused by
high concentrations of serum solids. Crit Rev Clin Lab Sci
1987, 25:19-41.
76. Stone JA, Moriguchi JR, Notto DR, Murphy PE, Dass CJ, Wessels
LM, Freier EF: Discrepancies between sodium concentrations
measured by the Kodak Ektachem 700 and by dilutional and
direct ion-selective electrode analyzers. Clin Chem 1992, 38:
2419-2422.
77. Morimatsu H, Rocktaschel J, Bellomo R, Uchino S, Goldsmith D,
Gutteridge G: Comparison of point-of-care versus central lab-
oratory measurement of electrolyte concentrations on calcu-
lations of the anion gap and the strong ion difference.
Anesthesiol 2003, 98:1077-1084.
78. Lang W, Zander R: The accuracy of calculated base excess in
blood. Clin Chem Lab Med 2002, 40:404-410.
79. Hatherill M, Waggie Z, Purves L, Reynolds L, Argent A: Mortality
and the nature of metabolic acidosis in children with shock.
Intensive Care Med 2003, 29:286-291.
80. Murray DM, Olhsson V, Fraser JI: Defining acidosis in postoper-
ative cardiac patients using Stewarts method of strong ion
difference. Pediatr Crit Care Med 2004, 5:240-245.
198
A
TOT
= sum of weak acids and proteins in human plasma; ICU = intensive care unit; IL = interleukin; LR = lactated Ringers; pCO
2
= Partial pres-
sure of carbon dioxide in arterial blood; SBE = standard base excess; SID = strong ion difference; SIDe = effective strong ion difference; SIG =
strong ion gap.
Critical Care April 2005 Vol 9 No 2 Kaplan and Frangos
Abstract
Acidbase abnormalities are common in the critically ill. The
traditional classification of acidbase abnormalities and a modern
physico-chemical method of categorizing them will be explored.
Specific disorders relating to mortality prediction in the intensive
care unit are examined in detail. Lactic acidosis, base excess, and
a strong ion gap are highlighted as markers for increased risk of
death.
Introduction
Deranged acidbase physiology drives admission to a critical
care arena for vast numbers of patients. Management of
diverse disorders ranging from diabetic ketoacidosis to hypo-
perfusion with lactic acidosis from hemorrhagic or septic
shock shares a variety of common therapies for disordered
acidbase balance. It is encumbent upon the intensivist to
decode the deranged physiology and to categorize the
disorder in a meaningful fashion to direct effective repair
strategies [1].
Besides the traditional classification of respiratory versus
metabolic, acidosis versus alkalosis, and gap versus nongap
(normal gap), the intensivist benefits from classifying acid
base disorders into three discrete groups: iatrogenically
induced (i.e. hyperchloremic metabolic acidosis), a fixed
feature of a pre-existing disease process (i.e. chronic renal
failure, hyperlactatemia), or a labile feature of an evolving
disease process (i.e. lactic acidosis from hemorrhage, shock
of any cause). The therapy for, and the outcome from, each of
these three categories may be distinctly different. A review of
the genesis of acidbase abnormalities is appropriate but will
be limited to metabolic derangements, as respiratory acid
base abnormalities are usually reparable with adjustments in
sedative or ventilator prescription.
Acidbase abnormality genesis
Traditional paradigms of acidbase abnormalities hinge on
generation of protons from the liberation of metabolic acids
such as lactate or carbonic acid from increased CO
2
. Most
traditional views rely on the HendersonHasselbach equation
to determine the pH and proton concentration. Other
attempts at classification rely upon nomograms with
imprecise grey zones to account for the imprecision in the
HendersonHasselbach equation solutions. The key fault
with these determinations is reliance upon bicarbonate as a
determinant of the pH. In 1983, Peter Stewart clarified the
physical chemistry principles that describe the independent
determinants of proton concentration and pH, allowing the
clinician to precisely and accurately determine the pH and to
understand the genesis of each acidbase disturbance
encountered [2].
The Stewartian methodology relies upon the relationships
between ions that completely dissociate at physiologic pH
so-called strong ions. There exist strong cations (Na
+
, K
+
,
Ca
2+
and Mg
2+
) as well as strong anions (Cl
, lactate, and
sulphates [most notable in renal failure]). These strong ions
establish a readily apparent strong ion difference (SID) that is
net strong ion-positive (normal approximately +40). Since
human acidbase physiology derives its homeostasis from
charge balance, according to the physical chemistry
principles articulated by Stewart the SID must be
counterbalanced by an equal and opposing charge termed
the effective strong ion difference (SIDe) (normal
approximately 40). The SIDe negative charge principally
stems from the dissociated moieties of plasma proteins
(~78% albumin) and phosphate (~20%). The sum of these
weak acids is known as A
TOT
since they exist in a dissociated
form (A
) as well as an associated form (AH). When the SID

Review
Clinical review: Acidbase abnormalities in the intensive care
unit
Lewis J Kaplan and Spiros Frangos
Yale University School of Medicine, Department of Surgery, Section of Trauma, Surgical Critical Care and Surgical Emergencies, New Haven,
Connecticut, USA
Corresponding author: Lewis J Kaplan, Lewis.Kaplan@yale.edu
Published online: 20 October 2004 Critical Care 2005, 9:198-203 (DOI 10.1186/cc2912)
199
and SIDe are equal, the plasma pH is exactly 7.4 at a pCO
2
of 40 torr. These relationships are demonstrated in Fig. 1.
Note that when the SID and SIDe are unequal, the difference
between the two is termed the strong ion gap (SIG) (SID SIDe,
normal = 0). This value is not discoverable by interrogation of
any other acidbase variables or scheme, and is buried within
the anion gap along with A
and lactate. It is important to note

that the generation or consumption of protons is driven by the
law of mass action upon the relationships identified in Fig. 2.
Saline is comprised of equal parts of sodium and chloride,
and as such appears electrically neutral. When equal
amounts of sodium and chloride are added to plasma,
however, the effects are different from those expected. The
plasma chloride level is less than that of sodium. The net
impact of adding equal amounts of sodium and chloride will
therefore raise the chloride to a greater degree than the
sodium. This results in a narrowed SID and a reduced plasma
positive net strong ion charge. When plasma positive charge
is reduced, as commonly occurs with significant chloride
loading (reduced SID), an immediate and compensatory
response is proton generation to aid in restoring charge
equilibrium. The clinician identifies this physiologic process
as a decreased pH. The genesis of hyperchloremic metabolic
acidosis is thus readily understandable based on the Stewart
principles [3]. It is important to recognize that the changes in
plasma electrolyte concentration are millimolar in scale while
the corresponding changes in proton concentration are
nanomolar. There is therefore an unfavorable electrochemical
gradient for simple plasma electrolyte and proton exchange;
the mechanism that underpins these changes is well explained
by Stewart [2].
Relatedly, an individual with chloride loss (vomiting, large
volume nasogastric losses without proton pump blockade)
would have a net increase in plasma positive charge. Exactly
the opposite process occurs to consume protons, leading to
an increased pH. Importantly, this clinical condition highlights
the mechanism underlying hypochloremic metabolic alkalosis
as well as the rationale behind chloride loading for repair
the Cl
therapeutically reduces the plasma excess positive

charge and the proton concentration in tandem. This process
is unassociated with mortality, reflects the common use of
loop diuretics for volume management, and will not be further
explored
A central tenet of the Stewart methodology identifies the
three independent control mechanisms for pH: SID, pCO
2
and A
TOT
. Bicarbonate is a dependent variable, and as such
does not determine the pH. This key concept aids in
constructing acidbase repair strategies in the critical care
environment. By way of example, patients with hyper-
chloremic metabolic acidosis may be corrected by altering
their intravenous fluid prescription. An ideal strategy reduces
plasma Cl
while preserving plasma Na

+
. This may be
achieved by prescribing D
5
W plus a variable amount of
NaHCO
3
as the maintenance fluid, with the amount of
NaHCO
3
dependent on the desired amount of Cl
and pH
change. This prescription provides a strong cation (Na
+
)
without a strong anion, resulting in an expected increase in
SID as Na
+
is maintained but Cl
falls; the increased SID

drives proton consumption and produces an increased pH.
Importantly, only changes in strong ions drive changes in
proton concentration. There are, however, readily identifiable
and compensatory changes in dependent ions such as
bicarbonate. Understanding the physiologic mechanisms
underpinning acidbase abnormalities thus provides a
rationale for therapeutic intervention. Indeed, a recent
comparison of traditional methods of acidbase interpretation
Figure 1
Charge balance in human plasma. SIDa, apparent strong ion
difference; SIDe, effective strong ion difference; SIG, strong ion gap.
Reproduced with permission from [1].
Figure 2
Charge interaction in human plasma. The equations demonstrate the
charge interactions in human plasma that serve as independent control
mechanisms for pH determination (pCO
2
, sum of weak acids and
proteins in human plasma [A
TOT
], and strong ion difference).
pCO
2
H
2
O + CO
2
H
2
CO
3
H
+
+ HCO
3
Weak acids/proteins
A
TOT
A
+ AH
(Na
+
+ K
+
+ Mg
2+
+ Ca
2+
) (Cl
+ lactate)
200
to guide therapy with Stewarts physicalchemical method
has championed the latter as an ideal means of determining
the mechanism, and of uncovering acidbase abnormalities
that were unappreciated using traditional classification and
interpretation schemes [4].
Lactic acidosis and hyperlactatemia
The most common acidbase abnormality in trauma patients
is lactic acidosis from hypovolemic shock and hypoperfusion.
Lactic acidosis is a gap metabolic acidosis that is a labile
feature of an evolving disease process. As such, lactic
acidosis is a final common feature of a variety of processes
that engender hypoperfusion, including diabetic ketoacidosis,
septic shock, cardiogenic shock, and a variety of intoxica-
tions. These entities will therefore not be discussed separately;
the discussion will instead focus on the consequences and
implications of lactic acidosis regardless of etiology.
Lactate generated from hypoperfusion generates acidosis as
the vast amount of lactate produced contributes a strong
anion, decreases the SID, and generates protons. In contrast,
lactate from lactated Ringers (LR) solution is in small
quantities (28 mmol/l) and is readily consumed, leaving
behind Na
+
as a strong cation; alkalinization results from the
more positive SID leading to proton consumption.
Resolution of lactic acidosis correlates well with survival in a
time-dependent fashion [5]. Moreover, resolving occult
hypoperfusion (normal vital signs, but a persistent lactic
acidosis) directly relates to infection risk as well as to
mortality [6,7]. Reduced infectious events (principally
respiratory complications) were realized using a protocol to
clear lactate, whether overt or occult, as an arbiter of
underlying hypoperfusion and systemic infection risk.
In order to avoid inappropriate therapy, it is important to
differentiate lactic acidemia from hyperlactatemia (normal pH,
elevated lactate level, constant lactate/pyruvate ratio). The
former indicates a condition that merits therapy (volume
expansion, inotropic support, septic source control), while
hyperlactatemia frequently stems from exogenous medications,
or as an endogenous accompaniment to persistently elevated
endogenous catecholamines after shock or trauma [8].
Lactic acidosis has long been utilized as an outcome
predictor with regard to survival after trauma, both blunt and
penetrating, as well as intra-abdominal catastrophe [57,9,
10]. However, lactate also performs quite well in the intensive
care unit (ICU) as a mortality gauge [11]. The presence of
this potent predictor of outcome is readily identifiable in the
ICU setting with physical examination using extremity
temperature as an arbiter (exclusive of patients with
peripheral occlusive vascular disease) [12].
Lactic acidosis, but not hyperlactatemia [13], closely
correlates with mortality risk and serves as a window into cell-
level oxygen-dependent processes. Moreover, clearance of
lactic acidemia portends an excellent likelihood of survival. In
one convenience sampling of surgical ICU patients (general
surgery and trauma) comparing lactate and base excess,
lactate appears superior in predicting mortality and morbidity
[14]. Relatedly, a separate study (prospective, consecutive,
mixed medicalsurgical patients) found that the combination
of the two variables appeared superior to either lactate or
base excess alone in predicting survival [15].
Standard base excess (base deficit)
A companion acidbase variable, base excess (commonly
presented as base deficit) has also been touted as a
prognostic variable in assessing outcome in the critically ill.
Base excess indicates metabolic acidosis or alkalosis, but
does not help place the acidosis into one or another category
with regard to genesis. It is, however, commonly and readily
assessed and is therefore the focus of a host of studies. A
plethora of studies present a mixed picture in the analysis of
base excess since the data derive from two distinct time
frames: Emergency Department arrival versus some time after
resuscitation. It is in the interpretation of base excess that the
Stewart principles are vital to guide interpretation. Indeed, it
has been demonstrated that the base excess may be
manipulated by fluid resuscitation. Generating a hyper-
chloremic metabolic acidosis will create a spuriously more
negative base deficit (or increased base excess) as the Cl
decreases the pH unaccompanied by hypoperfusion and

lactic acidemia [16]. Prognostication dependent on post-
resuscitation standard base excess (SBE) values must
therefore be interpreted with caution.
Nonetheless, presentation or pre-resuscitation base excess
values reliably indicate the degree of acid production
following injury [17]. Interestingly, in this large cohort analysis
of presentation SBE, the 50% lethal dose for the acid load
indicated by base deficit shifted to a substantially lower level
for a given age when combined with a traumatic brain injury; it
is unknown whether this is true for other injuries in isolation or
combination. The interpretation of SBE must therefore
incorporate the injury complex into decision-making, perhaps
limiting its utility. A recent study of salvageable trauma
patients who underwent arterial blood gas analysis identified
that SBE utility was greatest in predicting the outcome of
patients sustaining gunshot wounds and blunt injury versus
those with stab wounds or lacerations [18]. Mortality was
lower for stab/laceration patients at any given base deficit,
rendering interpretation in this subgroup problematic. Similar
to lactate, the rate of clearance of base deficit to normal,
rather than the absolute value, correlates better with survival
than do changes in pH [19].
It is important to note that, using an ex vivo model, base
excess values are CO
2
invariate (unlike pH), potentially aiding
in their initial utility and interpretation [20]. However, the
clinical milieu includes multiple elements that may impact
201
base excess, rendering the CO
2
base excess relationship
difficult to appreciate. Nonetheless, base excess correlates
with transfusion requirements and with length of stay [21].
In patients with major hepatic trauma, base deficit (50% lethal
dose, 11.8 mmol/l) and 24-hour transfusion requirement
(50% lethal dose, 5.4 l packed red blood cells) surfaced as the
strongest predictors of the risk of death, outperforming arterial
lactate [22]. Importantly, these observations and the model
were then tested on a different cohort with only pelvic
fractures, with excellent performance. Smaller studies in
pediatric trauma patients found that a base deficit less negative
than 5 predicted uniform survival since all study group deaths
occurred in patients with more negative base deficit values
[23]. It thus appears that pre-resuscitation base excess or
deficit correlates with survival and serves as another indicator
of an underlying disease (hypoperfusion), but interpretation
must be tempered by age and the mechanism of injury.
Hyperchloremic acidosis
While we touched upon hyperchloremic acidosis earlier, this
common iatrogenically induced entity deserves further
exploration. As already noted, the genesis of hyperchloremic
metabolic acidosis stems from excess chloride administration
relative to sodium, commonly as 0.9% normal saline solution,
0.45% normal saline solution, and even LR solution in large
quantities [2426]. This entity is thus an iatrogenic metabolic
acidosis of the nongap variety. Hyperchloremia has been
identified in up to 80% of patients admitted to a mixed
medicalsurgical ICU [26]. While not a predictor of outcome,
hyperchloremic metabolic acidosis may contribute to
morbidity and resource utilization. ICU admission for an
unexpected acidosis, increased and perhaps mechanically
supported minute ventilation to compensate for acidosis, and
more complex intravenous fluid prescriptions (especially
when utilizing hyperalimentation for nutritional support) are
but a few ICU care elements impacted by hyperchloremic
metabolic acidosis. While these events are probably
insignificant for the young and otherwise physiologically
sound patients, they may be significantly physiologically
challenging for the elderly or for those with physiologic
decompensation following significant trauma and hemor-
rhagic or septic shock.
The relationship between hyperchloremia and renal
dysfunction is well known [27,28]. Moreover, ICU survival has
been linked to Acute Pathophysiology and Chronic Health
Evaluation II/III scores and multiple organ dysfunction
syndrome, of which acute renal failure is a major element
[29]. Controversy has long surrounded whether patients die
from their renal failure or whether they die from the disease
process. Recent data strongly suggest that acute renal failure
is an independent risk factor for death despite renal
replacement therapy [30]. In this study of acute renal failure,
patients requiring renal replacement therapy suffered an
accelerated mortality (62.8%) compared with those without
renal failure (15.6%). The mortality differences remained
unexplained by differences in the severity of illness, thus
helping establish acute renal failure as an independent risk
factor for mortality. Moreover, complicated acidosis/alkalosis
was independently associated with death.
The deleterious impact of acute renal failure is thus potentially
minimized by avoiding iatrogenic hyperchloremia and its
attendant compromise of renal function. Further studies are
needed to ascertain the impact of this entity upon current
arbiters of morbidity including the ICU length of stay,
ventilator days, acute lung injury/acute respiratory distress
syndrome, and ventilator-associated pneumonia. Moreover,
virtually no research addresses hyperchloremia avoidance
strategies and their impact on morbidity such as acute renal
failure in at-risk populations, nor addresses mortality.
Both animal and human data identify a linearly decreased pH
and an increased SID with progressive chloride loading
[3133]. Interestingly, metabolic acidosis induced by
chloride from normal saline solution loading is associated
with impaired coagulation and the need for bicarbonate
buffering of the induced acidosis, while resuscitation with
comparable amounts of LR solution required no such therapy
[31,33]. Hyperchloremic acidosis, while not a predictor of
outcome, may therefore serve as a sentinel for hemorrhage
risk, for component transfusion therapy, and for accelerated
resource utilization. Importantly, one ex vivo study noted the
induction of a SIG with crystalloid-induced hyperchloremic
acidosis; no SIG was induced by adding comparable
amounts of large molecular weight hydroxyethyl starch [31].
In a related provocative study, sepsis survival was enhanced
by resuscitation with a large molecular weight hydroxyethyl
starch molecule suspended in a balanced salt solution
compared with LR solution or saline, and was unassociated
with hyperchloremic metabolic acidosis [34].
Immune effects of acidosis
The effects of metabolic acidosis span more than one system.
Immune activation has been intimately linked to the presence
of acidosis, and SIG generation may be but one feature.
Crystalloid resuscitation serves as a potent trigger for human
white blood cell count activation, manifested as an oxidative
burst and the expression of cell surface adhesion molecules
[35]. Activation of T-cell protein kinases has been demon-
strated with hypertonic saline, an effect whose downstream
cell-specific responses carry an uncertain significance [36].
More certainly, intravascular acid infusion reliably creates
acute lung injury and increases exhaled nitric oxide
concentration in a rat model [37]. This effect has been
demonstrated to stem from acidosis-stimulated expression of
inducible nitric oxide synthase, and was associated with
elaboration of the proinflammatory cytokine IL-6, also in a rat
preparation [38]. Importantly, this work suggests that
correction of acidosis may ameliorate inducible nitric oxide
synthase expression and reduce lung injury.
202
Relatedly, acidosis included by lactate, pyruvate or HCl has
been recently demonstrated to increase whole blood
viscosity at both high and low shear rates of flow. During
acidosis induction, hematocrit increases reflecting red blood
cell swelling were also observed. Most importantly, these
rheologic changes were reversible with the correction of
acidosis. These data lend support to the notion that
correcting acidosis represents more than treating numbers
and instead addresses important cellular and subcellular
events. It is possible that the increased viscosity and
hematocrit is responsible, in part, for regional hypoperfusion
despite normal or supranormal systemic flow. Clearly further
study is warranted, but one must consider that the time-
honored endpoint of mortality is not well suited to assess the
interventions targeting acidbase balance. Measures of
morbidity or resource utilization may be more appropriate
instead.
Strong ion gap
There are several studies that either support [39,40] or decry
the utility of the Stewart methodology in evaluating ICU
patients [26,41,42]. The SIG, as determined by Stewarts
physico-chemical method, is strongly associated with
metabolic acidosis, but is an independent entity that is
probably a labile feature of an evolving disease process. One
element that has surfaced from these studies is that the
Stewart methodology is a precise and readily utilizable means
of identifying the nature of the metabolic aberration; a
calculator to determine the individual components is
downloadable from the Internet [43]. How may one resolve
the seeming disparity of SIG utility identified in some studies
that is conspicuously lacking in others? The answer may be
found in the timing. Much like base excess, the value of the
SIG may be related to the time of assay. Since the natural
history of the SIG and its clearance value remains unknown
(similar to the early lactate observations), we must look to
pre-resuscitation SIG analysis as a more controlled evaluation
scheme.
In patients with major vascular injury requiring operative
repair, but prior to resuscitation, an increased SIG (>5) is
predictive of mortality [44]. Performance characteristics
based on receiveroperator characteristic curve analysis
indicated a SIG area of 0.991 for mortality (95% confidence
interval, 0.9720.998) and that for anion gap of 0.994 (95%
confidence interval, 0.9760.999), outperforming lactate
(receiveroperator characteristic curve area, 0.981; 95%
confidence interval, 0.9570.993). Multivariate logistic
regression analysis indicated that an increased SIG (odds
ratio, 3.6; 95% confidence interval, 1.996.78), more
strongly than injury severity score (odds ratio, 1.17; 95%
confidence interval, 1.061.31), was predictive of mortality.
In a related study in unselected trauma patients, the SIG
discriminated quite well between survivors and those who
died within 72 hours of Emergency Department arrival, again
outperforming lactate and base deficit [45]]. While the
absolute SIG levels were not identical, the import behind the
elevated level remains unaltered. It may be that the degree of
SIG elevation is disease specific. An increased SIG occurs in
patients with hepatic dysfunction [46] and renal dysfunction
[26], as well as during endotoxin-induced sepsis [47]. In a
large retrospective database analysis of patients requiring
ICU care, SIG >2 was independently linked with mortality in
patients evidencing metabolic acidosis [48].
Based on these studies, longitudinal assessments of changes
in the SIG as a predictor of outcome are underway. Nonethe-
less, it seems prudent to incorporate the pre-resuscitation
SIG into the mlange of information that guides outcome
prognostication. These data may be incorporated into daily
practice using a handheld calculator, or a computer-based
macro utilizing the relevant data points from the clinical
laboratory; automated abstraction is ideal but awaits the
development of appropriate interfaces with existing laboratory
devices. It is essential to note that no evaluation method
besides the physico-chemical one of Stewart allows the
clinician to ascertain the presence and magnitude of the SIG.
Conclusion
Traditional classification schemes of acidbase derange-
ments are too broad to aid in prognostication. Individual
acidbase element evaluation allows one to draw valid
conclusions regarding the likelihood of survival. The Stewart
physico-chemical approach to acidbase analysis readily
lends itself to these determinations by precisely evaluating
the independent determinants of pH as well as the important
SIG. At present, lactate, pre-resuscitation base deficit and
the SIG appear most predictive of outcome in the critically ill,
and they should be incorporated into a prognostication
method. Future studies of acidbase prediction of outcome
should strongly consider including each of these variables in their
methodology. Further evaluation of these and potentially other
markers of morbidity and resource utilization is appropriate.
Competing interests
References
1. Gunnerson K, Kellum JA: Acidbase and electrolyte analysis in
critically ill patients: are we ready for the new millennium?
Physiol Pharm 1983, 61:1444-1461.
3. Miller L, Waters JH: Mechanism of hyperchloremic nonanion
gap acidosis. Anesthesiology 1997; 87:1009-1010.
4. Fencl V, Jabor A, Kazda A, Figge J: Diagnosis of metabolic
acidbase disturbances in critically ill patients. Am J Respir
Crit Care Med 2000, 162:2246-2251.
5. Abramson D, Scalea TM, Hitchcock R, Trooskin SZ, Henry SM,
Greenspan J: Lactate clearance and survival following injury. J
Trauma 1993, 35:584-589.
6. Claridge JA, Crabtree TD, Pelletier SJ, Butler K, Sawyer RG,
Young JS: Persistent occult hypoperfusion is associated with
a significant increase in infection rate and mortality in major
trauma patients. J Trauma 2000, 48:8-14.
203
7. Blow O, Magliore L, Claridge JA, Butler K, Young JS: The golden
hour and the silver day: detection and correction of occult
hypoperfusion within 24 hours improves outcomes from
major trauma. J Trauma 1999, 47:964-969.
8. James JH, Luchette FA, McCarter FD, Fischer JE: Lactate is an
unreliable indicator of tissue hypoxia in injury or sepsis.
Lancet 1999, 354:505-508.
9. Jeng JC, Jablonski K, Bridgeman A, Jordan MH: Serum lactate,
not base deficit, rapidly predicts survival after major burns.
Burns 2002, 28:161-166.
10. Mikulaschek A, Henry SM, Donovan R, Scalea TM: Serum lactate
in not predicted by anion gap or base excess after trauma
resuscitation. J Trauma 1996, 40:218-224.
11. Mizock BM, Falk JL: Lactic acidosis in critical illness. Crit Care
Med 1992, 20:80-93.
12. Kaplan L, McPartland K, Santora TA, Trooskin SZ: Start with the
physical examination to identify hypoperfusion in ICU
patients. J Trauma 2001, 50:620-628.
13. Mizock BM: Significance of hyperlactatemia without acidosis
during hypermetabloic stress. Crit Care Med 1997, 25:1780-
1781.
14. Husain FA, Martin MJ, Mullenix PS, Steele SR, Elliott DC: Serum
lactate and base deficit as predictor of mortality and morbid-
ity. Am J Surg 2003, 185:485-491.
15. Smith I, Kumar P, Molloy S, Rhodes A, Newman PJ, Grounds RM,
Bennett ED: Base excess and lactate as prognostic indicators
for patients admitted to intensive care. Int Care Med 2001,
27:74-83.
16. Brill SA, Stewart TR, Brundage SI, Schreiber MA: Base deficit
does not predict mortality when secondary to hyperchloremic
acidosis. Shock 2002, 17:459-462.
17. Rutherford E, Morris JA, Reed GW, Hall KS: Base deficit strati-
fies mortality and determines therapy. J Trauma 1992, 33:417-
423.
18. Tremblay L, Feliciano DV, Rozycki G: Assessment of initial base
deficit as a predictor of outcome: mechanism of injury does
make a difference. Am Surg 2002, 68:689-693.
19. Davis J, Kaups KL, Parks SN: Base deficit is superior to pH in
evaluating clearance of acidosis after traumatic shock.
J Trauma 1998, 44:114-118.
20. Morgan TJ, Clark C, Endre Z: Accuracy of base excess an in
vitro evaluation of the Van Slyke equation. Crit Care Med
2000, 28:2932-2936.
21. Davis JW, Parks SN, Kaups KL, Gladen HE, O'Donnell-Nicol S:
Admission base deficit predicts transfusion requirements and
risk of complications. J Trauma 1996, 41:769-774.
22. Siegel JH, Rivkind AI, Dalal S, Godzari S: Early physiologic pre-
dictors of injury severity and death in blunt multiple trauma.
Arch Surg 1990, 125:498-508.
23. Randolph LC, Takacs M, Davis KA: Resuscitation in the pedi-
atric trauma population: admission base deficit remains an
important prognostic indicator. J Trauma 2002, 53:838-842.
going gynecologic surgery. Anesthesiology 1999, 90:1265-1270.
metabolic acidosis using polygeline pump prime. Int Care
Med 1999, 25:680-685.
26. Moviat M, van Haren F, van der Hoeven H: Conventional or
physicochemical approach in intensive care unit patients with
metabolic acidosis. Crit Care 2003, 7:R41-R45.
27. Wilcox CS: Regulation of renal blood flow by plasma chloride.
J Clin Invest 1983, 71:726-735.
effect of intravenous lactated ringers solution versus 0.9%
sodium chloride solution on serum osmolality in human vol-
unteers. Anesth Analg 1999, 88:999-1003.
29. Barie P, Hydo LJ, Fischer E: Utility of severity scoring for pre-
diction of prolonged critical care. J Trauma 1996, 40:513-519.
30. Metnitz PG, Krenn CG, Steltzer H, Lang T, Ploder J, Lenz K, Le
Gall JR, Druml W: Effect of renal failure requiring renal
replacement therapy on outcome in critically ill patients. Crit
Care Med 2002, 30:2051-2058.
31. Patterson T, Bailey H, Kaplan LJ: Hyperchloremia induces aci-
dosis, increases the strong ion gap, and impairs coagulation
[abstract]. Crit Care Med 2000, Suppl 28:A118.
32. Healey, MA, Davis RE, Liu FC, Loomis WH, Hoyt DB: Lactated
ringers in superior to normal saline in a model of massive
hemorrhage and resuscitation. J Trauma 1998, 45:894-899.
33. Waters JH, Gottlieb A, Schoenwald P, Popovich MJ, Sprung J,
Nelson DR: Normal saline versus lactated ringers solution for
intraoperative fluid management in patients undergoing
abdominal aortic aneurysm repair: an outcome study. Anesth
Analg 2001, 93:817-822.
experimental sepsis: improved short-term survival and
acidbase balance with Hextend compared with saline. Crit
Care Med 2002, 30:300-305.
35. Rhee P, Wang D, Ruff P, Austin B, DeBraux S, Wolcott K, Burris
D, Ling G, Sun L: Human neutrophil activation and increased
adhesion by various resuscitation fluids. Crit Care Med 2000,
28:74-78.
36. Junger WG, Hoyt DB, Hamreus M, Liu FC, Herdon-Remelius C,
Junger W, Altman A: Hypertonic saline activates protein
kinases and mitogen-activated protein kinase p38 in T-cells. J
Trauma 1997, 42:437-443.
37. Pedoto A, Caruso JE, Nandi J, Oler A, Hoffmann SP, Tassiopoulos
AK, McGraw DJ, Camporesi EM, Hakim TS: Acidosis stimulates
nitric oxide production and lung damage in rats. Am J Respir
Crit Care Med 1999, 159:397-402.
38. Haque IU, Huang CJ, Scumpia PO, Nasiroglu O, Skimming JW:
Intravascular infusion of acid promotes intrapulmonary
inducible nitric oxide synthase activity and impairs blood oxy-
genation in rats. Crit Care Med 2003, 31:1454-1460.
39. Kellum JA: Metabolic acidosis in the critically ill: lessons
learned from physical chemistry. Kidney Int 1998, 53 (Suppl
66):S81-S86.
anions identified by the FenclStewart method predict mortal-
admitted in the pediatric intensive care unit. Crit Care Med
1999, 27:1577-1581.
41. Hatherill M, Waggie Z, Purves L, Reynolds L, Argent A: Mortality
and the nature of metabolic acidosis in children with shock.
Int Care Med 2003, 29:286-291.
42. Cusack RJ, Rhodes A, Lochhead P, Jordan B, Perry S, Ball JA,
prognostic value in critically ill patients in a mixed
medical/surgical adult ICU. Int Care Med 2002, 28:864-869.
43. Kellum JA (Ed): The Acid Base pHorum [http://www.ccm.upmc.
edu/education/resources/phorum.html], accessed 10 December
2003.
strong ion difference and strong ion gap predicts outcome
1124.
45. Kaplan LJ, Bailey H, Klein A, et al.: Strong ion gap: a predictor of
early mortality following blunt or penetrating trauma
[abstract]. Crit Care Med 1999, Suppl 27:A42.
55.
47. Kellum JA, Bellomo R, Kramer DJ, Pinsky MR: Hepatic anion flux
during acute endotoxemia. J Appl Physiol 1995, 78:2212-
2217.
48. Gunnerson KJ, Saul M, Kellum JA: Lactic versus nonlactic meta-
bolic acidosis: outcomes in critically ill patients [abstract]. Crit
Care 2003; 7 (Suppl 2):S8.
204
A
TOT
= total concentration of weak acid; CO
2TOT
= total concentration of CO
2
; PaCO
2
= arterial CO
2
tension; PCO
2
= partial CO
2
tension; SBE =
standard base excess; SID = strong ion difference.
Critical Care April 2005 Vol 9 No 2 Morgan
Abstract
Stewarts quantitative physical chemical approach enables us to
understand the acidbase properties of intravenous fluids. In
Stewarts analysis, the three independent acidbase variables are
partial CO
2
tension, the total concentration of nonvolatile weak
acid (A
TOT
), and the strong ion difference (SID). Raising and
lowering A
TOT
while holding SID constant cause metabolic
acidosis and alkalosis, respectively. Lowering and raising plasma
SID while clamping A
TOT
cause metabolic acidosis and alkalosis,
respectively. Fluid infusion causes acidbase effects by forcing
extracellular SID and A
TOT
toward the SID and A
TOT
of the
administered fluid. Thus, fluids with vastly differing pH can have the
same acidbase effects. The stimulus is strongest when large
volumes are administered, as in correction of hypovolaemia, acute
normovolaemic haemodilution, and cardiopulmonary bypass. Zero
SID crystalloids such as saline cause a dilutional acidosis by
lowering extracellular SID enough to overwhelm the metabolic
alkalosis of A
TOT
dilution. A balanced crystalloid must reduce
extracellular SID at a rate that precisely counteracts the A
TOT
dilutional alkalosis. Experimentally, the crystalloid SID required is
24 mEq/l. When organic anions such as L-lactate are added to
fluids they can be regarded as weak ions that do not contribute to
fluid SID, provided they are metabolized on infusion. With colloids
the presence of A
TOT
is an additional consideration. Albumin and
gelatin preparations contain A
TOT
, whereas starch preparations do
not. Hextend is a hetastarch preparation balanced with L-lactate. It
reduces or eliminates infusion related metabolic acidosis, may
improve gastric mucosal blood flow, and increases survival in
experimental endotoxaemia. Stored whole blood has a very high
effective SID because of the added preservative. Large volume
transfusion thus causes metabolic alkalosis after metabolism of
contained citrate, a tendency that is reduced but not eliminated
with packed red cells. Thus, Stewarts approach not only explains
fluid induced acidbase phenomena but also provides a framework
for the design of fluids for specific acidbase effects.
Introduction
There is a persistent misconception among critical care
personnel that the systemic acidbase properties of a fluid
are dictated by its pH. Some even advocate pH-balanced
fluids, particularly when priming cardiopulmonary bypass
pumps [1]. This is not to deny the merit of avoiding very high
or very low pH in fluids intended for rapid administration.
Extremes of pH can cause thrombophlebitis, and on
extravasation tissue necrosis, and rapid administration is a
hemolysis risk (specific data on this topic are sparse).
However, these effects occur before equilibration. What must
be understood is that fluids with widely disparate pH values
can have exactly the same systemic acidbase effects. To
illustrate, the acidbase properties of pure 0.9% saline
(pH 7.0 at 25C) are identical to those of 0.9% saline
equilibrated with atmospheric CO
2
(pH 5.6 at 25C).
Until recently, the challenge was to find a logical basis for
predicting the acidbase properties of intravenous fluids. In this
review important concepts of quantitative physical chemistry
are presented, concepts originally set out by the late Peter
Stewart [25]. They provide the key to understanding fluid
induced acidbase phenomena and allow a more informed
approach to fluid design. On this background we consider the
effects of intravenous fluids on acidbase balance.
The Stewart approach in brief
There are just three independent variables that, when
imposed on the physical chemical milieu of body fluids,
dictate their acidbase status. They are strong ion difference
(SID), the total weak acid concentration (A
TOT
), and partial
CO
2
tension (PCO
2
). The interplay between SID, A
TOT
, and
PCO
2
is the sole determinant of pH, as well as of other
dependent variables such as [HCO
3
]. All acidbase
interventions, including fluid administration, act through SID,
A
TOT
and PCO
2
, alone or in combination. The single
exception is the addition of weak base (e.g. tris-hydroxymethyl
aminomethane) [6], which is normally absent from body fluids.
Review
Clinical review: The meaning of acidbase abnormalities in the
intensive care unit effects of fluid administration
Thomas J Morgan
Senior Specialist, Adult Intensive Care, Mater Misericordiae Hospitals, Brisbane, Australia
Corresponding author: Thomas J Morgan, thomas_morgan@mater.org.au
Published online: 3 September 2004 Critical Care 2005, 9:204-211 (DOI 10.1186/cc2946)
205
Elements such as Na
+
, K
+
, Ca
2+
, Mg
2+
, and Cl
exist in body
fluids as completely ionized entities. At physiologic pH this can
also be said of anions with pKa values of 4 or less, for example
sulphate, lactate, and -hydroxybutyrate. Stewart described all
such compounds as strong ions. In body fluids there is a
surfeit of strong cations, quantified by SID. In other words, SID
= [strong cations] [strong anions]. Being a charge space,
SID is expressed in mEq/l. SID calculated from measured
strong ion concentrations in normal plasma is 42 mEq/l.
Partial CO
2
tension
Arterial PCO
2
(PaCO
2
) is an equilibrium value determined by
the balance between CO
2
production (15,000 mmol/day) and
CO
2
elimination via the lungs. In areas where PCO
2
is less
directly controlled by alveolar ventilation (e.g. venous blood
and interstitial fluid during low flow states), the total CO
2
concentration (CO
2TOT
) becomes the independent variable.
Total concentration of weak acid (A
TOT
)
Body fluid compartments have varying concentrations of
nonvolatile (i.e. non-CO
2
) weak acids. In plasma these
consist of albumin and inorganic phosphate. The same
applies to interstitial fluid, although total concentrations here
are very small. In red cells the predominant source is
haemoglobin.
Nonvolatile weak acids dissociate in body fluids as follows:
HA H
+
+ A
The group of ions summarized as A
are weak anions (pKa

approximately 6.8). Unlike strong ions, weak ions in body fluids
vary their concentrations with pH by dissociation/association
of their respective parent molecules. The total concentration of
nonvolatile weak acid in any compartment is termed A
TOT
,
where A
TOT
= [HA] + [A
]. Although [A
] varies with pH, A

TOT
does not, and as such it is an independent variable.
Weak ions
The SID space is filled by weak ions, one of which is A
. The
only other quantitatively important weak ion is HCO
3
, but
there are also minute concentrations of CO
3
2
, OH
, and H
+
.
To preserve electrical neutrality, their net charge must always
equal the SID.
Stewarts equations
Stewart set out six simultaneous equations primarily
describing the behaviour of weak ions occupying the SID
space (Table 1). They are applications of the Law of Mass
Action to the dissociation of water, H
2
CO
3
, HCO
3
, and
nonvolatile weak acids, coupled with the expression for A
TOT
and a statement of electrical neutrality. If PCO
2
, SID and A
TOT
are known, then the equations in Table 1 can be solved for
the remaining six unknowns [A
], [HCO
3
], [OH
], [CO
3
2
],
[HA] and, most importantly, [H
+
].
Isolated abnormalities in strong ion difference and total
concentration of weak acid (A
TOT
)
From Stewarts equations, four simple rules can be derived
concerning isolated abnormalities in SID and A
TOT
(Table 2).
These can be verified by in vitro experimentation [7].
Standard base excess
The rules in Table 2 illustrate an important Stewart principle.
Metabolic acidbase disturbances arise from abnormalities in
SID and A
TOT
, either or both. However, to quantify metabolic
acidbase status at the bedside, neither SID nor A
TOT
needs
individual measurement. For this the standard base excess
(SBE) is sufficient. The SBE concept was developed by
Siggaard-Andersen and the Copenhagen group [8,9]. It is
calculated from buffer base offsets by assuming a mean
extracellular haemoglobin concentration of 50 g/l. A useful
formula is as follows (with SBE and [HCO
3
] values
expressed in mEq/l):
SBE = 0.93 {[HCO
3
] + 14.84 (pH 7.4) 24.4}

SBE supplements the Stewart approach as a practical tool
[1012]. A typical reference range is 3.0 to +3.0 mEq/l. The
SBE deviation from zero is the change in extracellular SID
needed to normalize metabolic acidbase status without
changing A
TOT
. If the SBE is below 3.0 mEq/l then there is
metabolic acidosis, either primary or compensatory. The
deviation below zero is the increase in extracellular SID
needed to correct the acidosis. Although this value should
also equate to the dose (in mmol) of NaHCO
3
required per
litre of extracellular fluid, in practice more is usually needed
a dose corresponding to an extracellular space of 30% body
weight rather than 20%. Similarly, if the SBE is greater than
3.0 mEq/l then there is metabolic alkalosis. The positive offset
from zero represents a theoretical dose calculation for HCl
rather than for NaHCO
3
.
Thinking about fluids in Stewarts terms
Fluids are administered into the physiological milieu. Their in
vivo properties can therefore be described using Stewarts
physical chemical language, in other words in terms of their
SID, A
TOT
and CO
2TOT
[13]. Acidbase effects come about
Table 1
Stewarts six simultaneous equations
[H
+
] [OH
] = Kw
[H
+
] [A
] = Ka HA
[HA] + [A
] = A
TOT
[H
+
] [HCO
3
] = Kc PCO
2
[H
+
] [CO
3
2
] = Kd [HCO
3
]
SID + [H
+
] [HCO
3
] [CO
3
2
] [A
] [OH
] = 0
All K values are known dissociation constants. PCO
2
, partial CO
2
tension; SID, strong ion difference.
206
as a fluid with a particular set of physical chemical properties
mixes and equilibrates with extracellular fluid (which itself
continually equilibrates across cell membranes with
intracellular fluid). This alters extracellular SID and A
TOT
, the
final determinants of metabolic acidbase status, toward the
SID and A
TOT
of the infused fluid.
The CO
2TOT
of infused fluid is worth mentioning separately.
First, it has no effect on extracellular SID and A
TOT
, and
therefore it does not influence the final metabolic acidbase
status. In other words, it is not the presence of HCO
3
in
bicarbonate preparations that reverses a metabolic acidosis;
rather, it is the high SID (1000 mEq/l for 1 mol/l NaHCO
3
)
and the absence of A
TOT
. The same metabolic effect would
be achieved if the weak anion were OH
rather than HCO

3
,
although the resultant high pH (14.0 rather than 7.7)
introduces a risk for haemolysis and tissue damage, and
mandates extremely slow administration via a central vein.
However, the CO
2TOT
of administered fluid can be important
for other reasons. Rapid infusion of fluids with high CO
2TOT
can transiently alter CO
2
homeostasis, mainly in areas under
less direct control of respiratory servo loops, such as venous
blood, the tissues and the intracellular environment [1418].
The crystalloid and colloid fluids discussed in this review are
not in this category.
Crystalloid effects from the Stewart perspective
No crystalloid contains A
TOT
. Crystalloid loading therefore
dilutes plasma A
TOT
, causing a metabolic alkalosis (Table 2).
Simultaneously, plasma and extracellular SID are forced
toward the SID of the infused crystalloid, primarily by
differential alteration in [Na
+
] and [Cl
]. If these changes
increase SID then the effects of A
TOT
dilution are enhanced,
and if they decrease SID then they oppose them (Table 2).
Dilutional acidosis
It has been reported on many occasions that large-scale
saline infusions can cause a metabolic acidosis [1921].
Although best documented during repletion of extracellular
fluid deficits, acute normovolaemic haemodilution [22,23] and
cardiopulmonary bypass [2326] have similar potential. The
mechanism is not bicarbonate dilution, as is commonly
supposed [27]. Bicarbonate is a dependent variable. The key
fact is that the SID of saline is zero, simply because the
strong cation concentration ([Na
+
]) is exactly the same as the
strong anion concentration ([Cl
]). Large volumes of saline

therefore reduce plasma and extracellular SID. This easily
overwhelms the concurrent A
TOT
dilutional alkalosis. A normal
(in fact reduced) anion gap metabolic acidosis is the end
result [28,29], albeit less severe than if A
TOT
had remained
constant.
The critical care practitioner should be alert to this possibility
when confronted with a patient who has a metabolic acidosis
and a normal anion gap. It is wise to check that the corrected
anion gap [30,31] and perhaps the strong ion gap [32,33]
are also normal. These are thought to be more reliable
screening tools for unmeasured anions [34,35]. (For a more
detailed discussion of the anion gap, corrected anion gap
and strong ion gap, see other reviews in this issue.) A history
of recent large volume saline infusion (e.g. >2 l in <24 hours)
in such a patient is highly suggestive of infusion related
metabolic acidosis. Even if there is an alternative explanation,
such as renal tubular acidosis or enteric fluid loss, saline
infusions will perpetuate and exacerbate the problem.
The phenomenon is not confined to 0.9% saline, and the
resultant metabolic acidosis may or may not be hyper-
chloraemic. Hypotonic NaCl solutions also have a zero SID.
Even fluids with no strong ions at all, such as dextrose
solutions, mannitol and water, have a zero SID. Infusion of any
of these fluids reduces plasma and extracellular SID by the
same equilibration mechanism, irrespective of whether
plasma [Cl
] rises or falls, forcing acidbase in the direction

of metabolic acidosis [36]. For a theoretical illustration of
dilutional SID effects, imagine adding 1 l of either saline or
water to a sealed 3 l mock extracellular compartment with a
SID of 40 mEq/l, as illustrated in Table 3. In either case the
SID is reduced to 30 mEq/l, but with a fall in [Cl
] after water
dilution.
Interestingly, hypertonicity makes solutions more acidifying
[36]. In this case the reduction in extracellular SID is
magnified by an added dilution effect, because water is
drawn by osmosis from the intracellular space. An unproven
corollary is that hypotonic solutions are less acidifying. The
important message here is that the intracellular space is a
participant in the final equilibrium, and can contribute
significantly to fluid induced acidbase effects.
Saline responsive metabolic alkalosis
Patients categorized as suffering from contraction alkalosis
or diminished functional extracellular fluid volume are said to
be saline responsive, and complex hormonal and renal
tubular mechanisms are often invoked [3739]. In fact, from
the perspective of physical chemistry, any metabolic alkalosis
is saline responsive, provided sufficient saline (or any zero
SID fluid) can be administered. Unfortunately, in the absence
Table 2
Rules for isolated abnormalities in strong ion difference (SID)
and total concentration of weak acid (A
TOT
)
SID/A
TOT
Isolated abnormality Result
SID Increased Metabolic alkalosis
SID Decreased Metabolic acidosis
A
TOT
Increased Metabolic acidosis
A
TOT
Decreased Metabolic alkalosis
207
of hypovolaemia the amount of saline required introduces a
risk for overload.
Hence, a diagnosis of volume depletion should be established
before treating metabolic alkalosis in this way. Signs of
extracellular volume depletion include reduced skin turgor,
postural hypotension, and systolic pressure variability [40].
There may also be a prerenal plasma biochemical pattern
(high urea:creatinine ratio), and if tubular function is preserved
then urinary [Na
] is normally under 20 mmol/l [41].

KCl and metabolic alkalosis
Some types of metabolic alkalosis are associated with
hypokalaemia and total body potassium deficits [37,42].
When dealing with these categories, correcting the deficit
with KCl is a particularly effective way to reverse the alkalosis.
From the Stewart perspective, this practice has similarities to
infusing HCl, minus the pH disadvantages of a negative SID.
This is because potassium and potassium deficits are
predominantly intracellular, and so all but a small fraction of
retained potassium ends up within the cells during correction.
The net effect of KCl administration is that the retained strong
anion (Cl
) stays extracellullar, whereas most of the retained

strong cation disappears into the intracellular space. This is a
potent stimulus for reducing plasma and extracellular SID.
To give another rough illustration, imagine the repletion of a
200 mmol total body potassium deficit using KCl. If the
extracellular [K
+
] is increased by 3 mmol/l during the process,
then approximately 50 mmol of K
+
has been retained in the
17 l extracellular space and about 150 mmol has crossed into
the cells. This means that 150 mmol Cl
is left behind in the

extracellular space, now unaccompanied by a strong cation.
This lowers extracellular SID and thus SBE by about 9 mEq/l.
Balanced crystalloids
To avoid crystalloid induced acidbase disturbances, plasma
SID must fall just enough during rapid infusion to counteract
the progressive A
TOT
dilutional alkalosis. Balanced
crystalloids thus must have a SID lower than plasma SID but
higher than zero. Experimentally, this value is 24 mEq/l
[23,43]. In other words, saline can be balanced by replacing
24 mEq/l of Cl
with OH
, HCO
3
or CO
3
2
. From this
perspective, and for now ignoring pH, solutions 1 and 3 in
Table 4 are balanced. However, it is noteworthy that, unless
stored in glass, solutions 1 and 3 both become solution 2 by
gradual equilibration with atmospheric CO
2
(Table 4).
Solution 2 is also balanced.
To eliminate the issue of atmospheric equilibration,
commercial suppliers have substituted various organic anions
such as L-lactate, acetate, gluconate and citrate as weak ion
surrogates. Solution 4 (Table 4) is a generic example of this
approach (for actual examples, see Table 5). L-lactate is a
strong anion, and the in vitro SID of solution 4 is zero.
However, solution 4 can also be regarded as balanced,
provided L-lactate is metabolized rapidly after infusion. In fact,
in the absence of severe liver dysfunction, L-lactate can be
metabolized at rates of 100 mmol/hour or more [44,45],
which is equivalent to nearly 4 l/hour of solution 4. The in vivo
or effective SID of solution 4 can be calculated from the
L-lactate component subject to metabolic disappearance. If
the plasma [lactate] stays at 2 mmol/l during infusion, then
solution 4 has an effective SID of 24 mEq/l.
Hence, despite wide variation in pH, solutions 14 in Table 4
have identical effective SID values. They are all balanced,
with identical systemic acidbase effects. However, other
attributes must be considered. Solution 1 (pH 12.38) is too
alkaline for peripheral or rapid central administration. The
situation for solution 2 is less clear. Atmospheric equilibration
has brought the pH to 9.35, which is less than that of sodium
thiopentone (pH 10.4) [46] a drug that is normally free of
venous irritation. Similarly Carbicarb, a low CO
2TOT
alternative
to NaHCO
3
preparations [47], has a pH of 9.6 [48]. Thus, the
pH of solution 2 may not preclude peripheral or more rapid
central administration. On the downside, and like Carbicarb,
solution 2 contains significant concentrations of carbonate,
which precipitates if traces of Ca
2+
or Mg
2+
are present. A
chelating agent such as sodium edetate may be required.
Choosing a balanced resuscitation crystalloid
Hartmanns solution (Table 5) is the best known commercial
balanced preparation. It contains 29 mmol/l of L-lactate. In
the absence of severe liver dysfunction, the effective SID is
therefore approximately 27 mEq/l. Although this should make
it slightly alkalinizing, much as Hartmann originally intended
[49], it is close to the ideal from an acidbase perspective.
Slight alkalinization is difficult to demonstrate in laboratory
and especially in clinical studies, but the available evidence
shows that Hartmanns solution reduces or eliminates
infusion related metabolic acidosis [5054].
The acidbase status of a patient before resuscitation is a
consideration. If it is normal to start with, then higher SID
fluids such as Plasma-Lyte 148 (effective SID 50 mEq/l;
Table 3
Equivalent strong ion difference reductions by adding 1 l water
or 1 l of 0.15 mol/l NaCl to a 3 l sample of mock extracellular
fluid
After saline After water
ECF dilution dilution
[Na
+
] 140 142.5 105
[Cl
] 100 112.5 75
[A
] + [HCO
3
] 40 30 30
SID 40 30 30
Electrolyte concentrations are given in mEq/l. ECF, extracellular fluid;
SID, strong ion difference.
208
Table 5) are likely to cause a progressive metabolic alkalosis
from the outset. Again, evidence is limited, but in support of
this statement Plasma-Lyte 148 priming cardiopulmonary
bypass pumps has been shown to increase arterial base
excess by the end of bypass [25]. On the other hand, if there
is a pre-existing metabolic acidosis, caused by diabetic
ketoacidosis or hypovolaemic shock for example, then fluids
with higher effective SID such as Isolyte E or Plasma-Lyte
148 will correct the acidosis more rapidly (provided their
organic anions are metabolized with efficiency) while
counteracting ongoing generation of acidosis. The problem
with high SID fluids is the potential for over-correction and
break through metabolic alkalosis, particularly when the
cause of the acidosis is accumulation of organic strong
anions such as ketoacids and lactate, which disappear as the
illness resolves.
Unfortunately, available commercial balanced preparations
have unresolved problems. Many contain either calcium or
magnesium (or sometimes both; Table 5). Calcium neutralizes
the anticoagulant effect of citrate, and both can precipitate in
the presence of HCO
3
and CO
2
2
. This restricts their range
of ex vivo compatibilities (e.g. there are incompatibilities with
stored blood and sodium bicarbonate preparations) and
makes them poor drug delivery vehicles. Another
disadvantage is that they all require an intermediary metabolic
step, often at times of severe metabolic stress, to achieve
their effective SID.
Hartmanns solution is also hypotonic relative to extracellular
fluid. Although a potential disadvantage in traumatic brain
injury [55], this was not borne out in a comparison with
hypertonic saline given prehospital to hypotensive brain-
injured patients [56]. Diabetic ketoacidosis is another
scenario that predisposes to brain swelling during fluid
loading [57], but here Hartmanns solution and other mildly
hypotonic preparations seem safe for a least part of the
repletion process [5861]. If used from the beginning, the
slightly alkalinizing Hartmanns SID of 27 mEq/l is probably
sufficient to ameliorate or even prevent the late-appearing
normal anion gap metabolic acidosis to which these patients
are prone [57], although this remains to be demonstrated.
Overcoming current shortcomings
Given the limitations of commercially available solutions and
assuming that infusion-related acidosis causes harm, as
seems likely [62], then an argument could be put for new
balanced resuscitation solutions. Ideally, these should be
normotonic and free of organic anion surrogates and divalent
cations. The design could be along the lines of solution 3 in
Table 4. However, because solution 3 requires CO
2
-
impermeable storage, solution 2 might be preferable,
provided its higher pH does not preclude rapid peripheral
administration. Such a fluid could become the first line
crystalloid in all large volume infusion scenarios, including
intraoperative fluid replacement, acute normovolaemic
haemodilution and cardiopulmonary bypass, as well as
resuscitation of hypovolaemic and distributive shock, diabetic
ketoacidosis and hyperosmolar nonketotic coma. Refine-
ments would include a selection of [Na
+
] and corresponding
[Cl
] values to cater for varying osmolality requirements. The

standard SID for neutral acidbase effects would be
24 mEq/l, perhaps with variations above or below to correct
pre-existing acidbase disturbances.
Colloids
The SAFE (Saline versus Albumin Fluid Evaluation) study has
lifted the cloud hanging over albumin solutions [63], and
clinicians should now feel more comfortable using colloid
preparations in general. Just as with crystalloids, the
Table 4
Four balanced crystalloids (see text)
Solution 1 Solution 2 Solution 3 Solution 4
[Na
+
] 140 140 140 140
[Cl
] 116 116 116 114

[HCO
3
] 19.2 24
[CO
3
2
] 4.8
[OH
] 24
[L-lactate] 26
PCO
2
(mmHg) 0 0.3
a
760 0.3
a
pH 12.38 9.35 6.04 6.49
Effective SID 24 24 24 24
a
Atmospheric sea level partial CO
2
tension (PCO
2
). Electrolyte
concentrations are given in mEq/l. SID, strong ion difference.
Table 5
Four commercial crystalloids
Plasma-Lyte Isolyte S
Hartmanns 148 (pH 7.4) Isolyte E
[Na
+
] 129 140 141 140
[Cl
] 109 98 98 103
[K
+
] 5 5 5 10
[Ca
2+
] 4 5
[Mg
2+
] 3 3 3
[L-lactate] 29
[Acetate] 27 27 49
[Gluconate] 23 23
[Citrate] 8
[Phosphate] 1
Effective SID 27
a
50 50 57
a
Assumes stable plasma lactate concentrations of 2 mmol/l (see text).
All concentrations are given in mEq/l.
209
effective SID of a colloid is a fundamental acidbase
property. This is tempered by two other factors. First, lower
infusion volumes are normally required for the same
haemodynamic effect [63], reducing the forcing function of
SID equilibration. Second, the colloid molecule itself may be
a weak acid. In other words some colloids contain A
TOT
, as is
the case with albumin and gelatin preparations (Table 6)
[64]. A
TOT
dilutional alkalosis is thus reduced or eliminated
when these fluids are infused, at least until the colloid
disappears from the extracellular space.
However, the SID values of commercially available weak acid
colloids are all significantly greater than zero (Table 6). On
infusion, the raised SID will tend to offset the acidbase
effects of A
TOT
infusion. As a result the overall tendency of
standard albumin and gelatin based colloids to cause
metabolic acidosis is probably similar to that of saline. By
contrast, hetastarch and pentastarch are not weak acids, and
the SID of standard starch preparations is zero (Table 6).
Their acidbase effects are therefore likely to be similar to
those of saline and the weak acid colloids [17].
Balanced colloids are still at the investigational stage.
Hextend (Table 6) is a balanced hetastarch preparation [65].
It contains L-lactate, which, by raising the effective SID to
26 mEq/l, reduces or eliminates infusion related metabolic
acidosis, and perhaps improves gastric mucosal blood flow
[66]. Experimentally, this appears to offer a survival
advantage in endotoxaemia [67].
Blood
At collection, blood is mixed with a preservative, normally
CPDA-1 [68], providing approximately 17 mEq trivalent
citrate anions per unit, and a small amount of phosphate [69].
The accompanying sodium cation adds about 40 mEq/l to the
effective SID of whole blood. For this reason it is not
surprising that large volume whole blood transfusion
commonly results in a post-transfusion metabolic alkalosis
(following citrate metabolism). With packed red cells, the
standard red cell preparation in most countries, the
preservative load per blood unit is reduced. Nevertheless,
large volume replacement with packed red cells still produces
metabolic alkalosis [69]. Conversely, if liver dysfunction is
severe enough to block or grossly retard citrate metabolism,
then the problem becomes ionized hypocalcaemia and
metabolic acidosis [70].
Conclusion
The principles laid down by the late Peter Stewart have
transformed our ability to understand and predict the
acidbase effects of fluids for infusion. As a result, designing
fluids for specific acidbase outcomes is now much more a
science than an art.
Competing interests
The author declares no competing interests.
Acknowledgements
The authors research in this area has been supported by Research
Grants from the Australian and New Zealand College of Anesthetists
and the Royal Brisbane Hospital Research Foundation.
Table 6
Six colloid solutions
Albumex 4 Haemaccel Gelofusine PENTASPAN HESpan Hextend
[Albumin]
b
40 g/l
[Gelatin urea-linked]
b
35 g/l
[Gelatin succinylated]
b
40 g/l
[Pentastarch] 100 g/l
[Hetastarch] 60 g/l 60 g/l
[Na
+
] 140 145 154 154 154 143
[K
+
] 5.1 3
[Ca
2+
] 12.5 5
[Mg
2+
] 0.8
[Cl
] 128 145 120 154 154 124

[L-lactate] 28
[Glucose] 5.5
[Octanoate] 6.4
Effective SID 12 17.6 34 0 0 26
a
a
Assumes stable plasma lactate concentrations of 2 mmol/L (see text).
b
Weak acid. Electrolyte concentrations are given in mEq/l. SID, strong ion
difference.
210
References
1. Lilley A: The selection of priming fluids for cardiopulmonary
bypass in the UK and Ireland. Perfusion 2002, 17:315-319.
2. Stewart PA: How to understand acidbase. In A Quantitative
Acidbase Primer for Biology and Medicine. Edited by Stewart
PA. New York: Elsevier; 1981:1-286.
Crit Care 2000, 4:6-14.
5. Wooten EW: Science review: Quantitative acidbase physiol-
ogy using the Stewart model. Crit Care 2004, 8:in press.
acidosis with sodium bicarbonate or tris-hydroxymethyl
Analg 2003, 96:1201-1208.
Physiol 1986, 61:2260-2265.
8. Siggaard-Andersen O: The Van Slyke equation. Scand J Clin
Lab Invest 1977, Suppl 146:15-20.
base (strong ion difference) as measure of a non-respiratory
acid-base disturbance. Acta Anesth Scand 1995, Suppl 107:
123-128.
10. Morgan TJ, Clark C, Endre ZH: The accuracy of base excess: an
in vitro evaluation of the Van Slyke equation. Crit Care Med
2000, 28:2932-2936.
11. Schlichtig R, Grogono AW, Severinghaus JW: Current status of
acid-base quantitation in physiology and medicine. Anesthesiol
Clin North Am 1998, 16:211-233.
2
and standard base excess compensation for acid-base imbal-
ance. Crit Care Med 1998, 26:1173-1179.
13. LeBlanc M, Kellum J: Biochemical and biophysical principles of
hydrogen ion regulation. In Critical Care Nephrology. Edited by
Ronco C, Bellomo R. Dordrecht: Kluwer Academic Publishers;
1998:261-277.
14. Kraut JA, Kurtz I: Use of base in the treatment of severe aci-
demic states. Am J Kidney Dis 2001, 38:703-727.
16. Gehlbach BK, Schmidt GA: Bench-to-bedside review: Treating
acidbase abnormalities in the intensive care unit the role of
buffers. Crit Care 2004, 8:259-265.
17. Mathieu D, Neviere R, Billard V, Fleyfel M, Wattel F: Effects of
bicarbonate therapy on hemodynamics and tissue oxygena-
tion in patients with lactic acidosis: a prospective controlled
study. Crit Care Med 1991, 19:1352-1356.
18. Cooper DJ, Walley KR, Wiggs BR, Russell JA: Bicarbonate does
not improve hemodynamics in critically ill patients who have
lactic acidosis: a prospective controlled clinical study. Ann
Intern Med 1990, 112:492-498.
1270.
20. McFarlane C, Lee A: A comparison of Plasmalyte 148 and 0.9%
saline for intra-operative fluid replacement. Anaesthesia 1994,
49:779-781.
21. Prough DS, Bidani A: Hyperchloremic metabolic acidosis is a
predictable consequence of intraoperative infusion of 0.9%
saline. Anesthesiology 1999, 90:1247-1249.
Finsterer U: Acidbase changes caused by 5% albumin versus
normovolemic hemodilution: a randomized prospective study.
Anesthesiology 2000, 93:1174-1183.
23. Morgan TJ, Venkatesh B, Hall J: Crystalloid strong ion difference
determines metabolic acid-base change during acute normov-
olemic hemodilution. Intensive Care Med 2004, 30:1432-1437.
24. Hayhoe M, Bellomo R, Lin G, McNicol L, Buxton B: The aetiology
Care Med 1999, 25:680-685.
25. Liskaser FJ, Bellomo R, Hayhoe M, Story D, Poustie S, Smith B,
Letis A, Bennett M: Role of pump prime in the etiology and
pathogenesis of cardiopulmonary bypass-associated acido-
sis. Anesthesiology 2000, 93:1170-1173.
26. Himpe D, Neels H, De Hert S, Van Cauwelaert P: Adding lactate
to the prime solution during hypothermic cardiopulmonary
bypass: a quantitative acidbase analysis. Br J Anaesth 2003,
90:440-445.
27. Mathes DD, Morell RC, Rohr MS: Dilutional acidosis: Is it a real
clinical entity? Anesthesiology 1997, 86:501-503.
28. Miller LR, Waters JH: Mechanism of hyperchloremic nonanion
gap acidosis. Anesthesiology 1997, 87:1009-1010.
29. Storey DA: Intravenous fluid administration and controversies
in acidbase. Crit Care Resusc 1999, 1:151-156.
Care Med 2000, 162:2246-2251.
strong ion difference and strong ion gap predict outcome from
major vascular injury. Crit Care Med 2004, 32:1120-1124.
1992, 152:1625-1629.
35. Iberti TJ, Leibowitz AB, Papadakos PJ, Fischer EP: Low sensitiv-
ity of the anion gap as a screen to detect hyperlactaemia in
critically ill patients. Crit Care Med 1991, 19:130-131.
36. Makoff DL, da Silva JA, Rosenbaum BJ, Levy SE, Maxwell MH:
Hypertonic expansion: acid-base and electrolyte changes. Am
J Physiol 1970, 218:1201-1207.
37. Narins RG, Gardner LB: Simple acid-base disturbances. Med
Clin North Am 1981, 65:321-360.
38. Adrogue HJ, Madias NE: Medical progress: management of
life-threatening acid-base disorders: second of two parts. N
Engl J Med 1998, 338:107-111.
39. Worthley LIG: Acidbase balance and disorders. In Ohs Inten-
sive Care Manual. Edited by Bersten AD, Soni N. Edinburgh: But-
terworth Heinemann; 2003:873-883.
40. Morgan TJ: Haemodynamic monitoring. In Ohs Intensive Care
Manual. Edited by Bersten AD, Soni N. Edinburgh: Butterworth
Heinemann; 2003:79-94.
41. Bellomo R: Acute renal failure. In Ohs Intensive Care Manual.
Edited by Bersten AD, Soni N. Edinburgh: Butterworth Heine-
mann; 2003:453-458.
42. Gluck S: Acidbase. Lancet 1998, 352:474-479.
43. Morgan TJ, Venkatesh B, Hall J: Crystalloid strong ion differ-
ence determines metabolic acidbase change during in vitro
haemodilution. Crit Care Med 2002, 30:157-160.
44. McLean AG, Davenport A, Cox D, Sweny P: Effects of lactate-
buffered and lactate-free dialysate in CAVHD patients with
and without liver dysfunction. Kidney Int 2000, 58:1765-1772.
45. Kierdorf HP, Leue C, Arns S: Lactate- or bicarbonate-buffered
solutions in continuous extracorporeal renal replacement
therapies. Kidney Int 1999, Suppl 72:S32-S36.
46. Trissel LA: Thiopental sodium. In Handbook of Injectable Drugs,
9th ed. Bethesda, MD: American Society of Health-System Phar-
macists; 1996:1032-1038.
47. Leung JM, Landow L, Franks M, Soja-Strzepa D, Heard SO, Arieff
AI, Mangano DT: Safety and efficacy of intravenous Carbicarb
in patients undergoing surgery: Comparison with sodium
bicarbonate in the treatment of mild metabolic acidosis. Crit
Care Med 1994, 22:1540-1549.
48. Shapiro JI, Elkins N, Logan J, Ferstenberg LB, Repine JE: Effects
of sodium bicarbonate, disodium carbonate, and a sodium
bicarbonate/carbonate mixture on the PCO
2
of blood in a
closed system. J Lab Clin Med 1995, 126:65-69.
49. Hartmann AF, Senn MJ: Studies in the metabolism of sodium r-
lactate. 1. Response of normal human subjects to the intra-
venous injection of sodium r-lactate. J Clin Invest 1932, 11:
337-344.
50. Traverso LW, Lee WP, Langford MJ: Fluid resuscitation after an
otherwise fatal hemorrhage: 1. Crystalloids solutions. J Trauma
1986, 26:168-175
effect of intravenous lactated Ringers solution versus 0.9%
sodium chloride on serum osmolality in human volunteers.
Anesth Analg 1999, 88:999-1003.
211
52. Reid F, Lobo DN, Williams RN, Rowlands BJ, Allison SP:
(Ab)normal saline and physiological Hartmanns solution: a
randomized double-blind cross over study. Clin Sci (Lond)
2003, 104:17-24.
53. Waters JH, Gottleib A, Schoenwald P, Popovich MJ, Sprung J,
Nelson DR: Normal saline versus lactated Ringers solution for
intraoperative fluid management in patients undergoing
abdominal aortic aneurysm repair: an outcome study. Anesth
Analg 2001, 93:817-822.
54. Takil A, Eti Z, Irmak P, Yilmaz Gogus F: Early postoperative res-
piratory acidosis after large intravascular volume infusion of
lactated Ringers solution during major spine surgery. Anesth
Analg 2002, 95:294-298.
55. Myburgh JA: Severe head injury. In Ohs Intensive Care Manual.
mann; 2003:689-709.
56. Cooper DJ, Myles PS, McDermott FT, Murray LJ, Laidlaw J,
Cooper G, Tremayne AB, Bernard SS, Ponsford J; HTS Study
Investigators: Prehospital hypertonic saline resuscitation of
patients with hypotension and severe traumatic brain injury: a
randomized controlled trial. JAMA 2004, 291:1350-1357.
57. Keays R: Diabetic emergencies. In Ohs Intensive Care Manual.
mann; 2003:551-558.
58. Hillman K: Fluid resuscitation in diabetic emergencies: a reap-
praisal. Intensive Care Med 1987, 13:4-8.
59. Harris GD, Fiordalisi I, Harris WL, Mosovich LL, Finberg L: Mini-
mizing the risk of brain herniation during treatment of dia-
betic ketoacidemia: a retrospective and prospective study. J
Pediatr 1990, 117:22-31.
60. Rother KI, Schwenk WF: Effect of rehydration fluid with 75
mmol/L of sodium concentration and serum osmolality in
young patients with diabetic ketoacidosis. Mayo Clin Proc
1994, 69:1149-1153.
61. Linares MY, Schunk JE, Lindsay R: Laboratory presentation in
diabetic ketoacidosis and duration of therapy. Pediatr Emerg
Care 1996, 12:347-351.
62. Gunnerson KJ, Kellum JA: Acidbase and electrolyte analysis in
critically ill patients: are we ready for the new millenium? Curr
Opin Crit Care 2003, 9:468-473.
63. Finfer S, Bellomo R, Boyce N, French J, Myburgh J, Norton R, for
the SAFE Study Investigators: A comparison of albumin and
saline for fluid resuscitation in the intensive care unit. N Engl J
Med 2004, 350:2247-2256.
64. Liskaser F, Story DA: The acidbase physiology of colloid solu-
tions. Curr Opin Crit Care 1999, 5:440-442.
65. Gan TJ, Bennett-Guerrero E, Phillips-Bute B, Wakeling H,
Moskowitz DM, Olufolabi Y, Konstadt SN, Bradford C, Glass PS,
Machin SJ, et al.: Hextend, a physiologically balanced plasma
expander for large volume use in major surgery: a ran-
domised phase III clinical trial. Hextend Study Group. Anesth
Analg 1999, 88:992-998.
66. Wilkes NJ, Woolf R, Mutch M, Mallett SV, Peachey T, Stephens R,
Mythen MG: The effects of balanced versus saline-based het-
astarch and crystalloid solutions on acid-base and electrolyte
status and gastric mucosal perfusion in elderly surgical
patients. Anesth Analg 2001, 93:811-816.
Med 2002, 30:300-305.
68. Mollison PL, Engelfreit CP, Contreras M: The transfusion of red
cells. In Blood Transfusion in Clinical Medicine. Oxford: Black-
well Science; 1997:278-314.
69. Driscoll DF, Bistrian BR, Jenkins RL, Randall S, Dzik WH, Gerson
B, Blackburn GL: Development of metabolic alkalosis after
massive transfusion during orthotopic liver transplantation.
Crit Care Med 1987, 15:905-908.
70. Mollison PL, Engelfreit CP, Contreras M: Some unfavourable
effects of transfusion. In: Blood Transfusion in Clinical Medi-
cine. Oxford: Blackwell Science; 1997:487-508.
573
A
TOT
= total concentration of weak acids; CA = carbonic anhydrase; CD = collecting duct; DCT = distal convoluted tubule; dRTA = distal renal
tubular acidosis; kNBC = kidney Na
+
/HCO
3
cotransporter; NAE = net acid excretion; PCO

2
= partial CO
2
tension; PHA = pseudohypoaldostero-
nism; ROMK = renal outer medullar K
+
channel; RTA = renal tubular acidosis; SID = strong ion difference; SLC = solute carrier; TSC = thiazide-
sensitive cotransporter.
Abstract
The Canadian physiologist PA Stewart advanced the theory that
the proton concentration, and hence pH, in any compartment is
dependent on the charges of fully ionized and partly ionized
species, and on the prevailing CO
2
tension, all of which he dubbed
independent variables. Because the kidneys regulate the
concentrations of the most important fully ionized species ([K
+
],
[Na
+
], and [Cl
]) but neither CO
2
nor weak acids, the implication is
that it should be possible to ascertain the renal contribution to
acidbase homeostasis based on the excretion of these ions. One
further corollary of Stewarts theory is that, because pH is solely
dependent on the named independent variables, transport of
protons to and from a compartment by itself will not influence pH.
This is apparently in great contrast to models of proton pumps and
bicarbonate transporters currently being examined in great
molecular detail. Failure of these pumps and cotransporters is at
the root of disorders called renal tubular acidoses. The
unquestionable relation between malfunction of proton
transporters and renal tubular acidosis represents a problem for
Stewart theory. This review shows that the dilemma for Stewart
theory is only apparent because transport of acidbase equivalents
is accompanied by electrolytes. We suggest that Stewart theory
may lead to new questions that must be investigated
experimentally. Also, recent evidence from physiology that pH may
not regulate acidbase transport is in accordance with the
concepts presented by Stewart.
Introduction
Renal tubular acidoses (RTAs) are forms of metabolic
acidoses that are thought to arise from a lack of urine
excretion of protons or loss of bicarbonate (HCO
3
) due to a
variety of tubular disorders. Characteristically, this causes a
hyperchloraemic (non-anion gap) acidosis without impaired
glomerular filtration. Molecular studies have identified genetic
or acquired defects in transporters of protons and HCO
3
in
many forms of RTA. However, at the same time these trans-
porters have been found also to be involved in transport of Cl
and Na
+
. Furthermore, in a few cases RTA has been associa-
ted with primary defects in electrolyte transporters alone.
The core of Stewart theory is that transport of protons as
such is unimportant to regulation of pH. In contrast, the
theory states that acidbase homeostasis is directly
regulated by electrolyte transport in the renal tubules. H
+
is
effectively a balancing requirement imposed by physical
chemistry. Accounting for how this occurs will probably lead
to an improved understanding of homeostasis.
We begin the review by describing the classical formulation
of the renal regulation of acidbase homeostasis. We then
describe the quantitative physical chemistry notion of
acidbase as described by Stewart (henceforth called the
physicochemical approach). On this basis we analyze some
of the mechanisms that are active in RTA. We show that the
physicochemical approach may lead to new questions that
can be pursued experimentally to supplement insights already
gained with classical theory. Several authors have suggested
that the physicochemical approach could be used to the
benefit of our understanding of RTA [1,2].
The kidney as regulator of acidbase balance
According to traditional concepts [3], daily acid production is
calculated as the combined excretion of sulphate anion
(SO
4
2
) and organic anions in the urine, whereas renal
elimination of acid equivalents is computed as the combined
titrable acidity + ammonium excreted HCO
3
, called net
acid excretion (NAE). Cohen and coworkers [4] reviewed
Review
Clinical review: Renal tubular acidosis a physicochemical
approach
Troels Ring
1
, Sebastian Frische
2
and Sren Nielsen
3
1
Consultant, Department of Nephrology, Aalborg Hospital, Aalborg, Denmark
2
Assistant Professor, The Water and Salt Research Center, Institute of Anatomy, University of Aarhus, Aarhus, Denmark
3
Professor of Cell Biology and Pathophysiology, Director, The Water and Salt Research Center, Institute of Anatomy, University of Aarhus, Aarhus, Denmark
Corresponding author: Troels Ring, tring@gvdnet.dk
574
Critical Care December 2005 Vol 9 No 6 Ring et al.
evidence indicating that the traditional view may be incon-
sistent with observations in patients in renal failure and in a
number of experimental studies. In one of the studies
assessed, Halperin and coworkers [5] examined rats loaded
with extra alkali on top of already basic ordinary rat chow.
Amazingly, increasing unmeasured organic anions had a 10-
fold greater effect on alkali disposal than did changes in NAE,
as traditionally computed. Similar findings had already been
reported by Knepper and coworkers [6] in 1989. That
acidbase balance is always accounted for by standard
measurements may therefore be disputed. Although fervently
rejected [3], this has given rise to a proposal of a new
classification system for NAE that includes the regulation of
loss of organic anions or potential HCO
3
[7].
Difficulties in measuring titrable acidity and organic anions
are one main source of disagreement with regard to
acidbase homeostasis [4] both in normal persons and in
those with renal impairment [8]. A recent Danish study [9]
reinforced the concept from studies of healthy humans
exposed to acid loads that nonmetabolizable base excretion
is important to renal regulation of acidbase homeostasis.
Central to renal acidbase physiology is excretion of
ammonium. One view [10] is that ammonium is produced as
NH
4
+
in large quantities from hydrolysis of peptide bonds,
and its excretion in urine has no bearing on acidbase
chemistry except for the fact that for nitrogen balance it
would otherwise have to be converted to urea a process
seen to consume bicarbonate. Exactly this argument was
used again by Nagami [11] in an authoritative review of renal
ammonia production and excretion. Most recently a study of
normal individuals [12] showed that ureagenesis increased
during experimental acidosis produced by CaCl
2
. This
contrasted with the authors expectations because urea-
genesis was supposed to cost alkali.
However, the traditional view is that NH
4
+
excretion is one of
the most important mechanisms for eliminating metabolic
acid equivalents because the leftover from deamination of
glutamine is effectively bicarbonate and the process comes
to a halt if NH
4
+
is not eliminated [13]. As stated in recent
accounts, this view also accounts for the bicarbonate toll of
ureagenesis [14] but the details of regulation and overall
stoichiometry are still debated. However, it seems that the
handling of NH
4
+
in the kidney is of great importance
because a complicated network of transport mechanisms
have evolved [11]. Most recently, a new group of putative
NH
4
+
(and NH
3
?) transporters related to the rhesus group of
proteins has been described [15]. As far as we know, the
result of missing one or more of these transporters on
acidbase balance is not yet known, and because of
redundancy it could be limited. Finally, apart from being a
transported quantity that is of importance per se, NH
4
+
has
also been found to influence a number of other tubular
processes that are involved in acidbase regulation [16,17].
Hence, although there can be no doubt that excretion of
NH
4
+
is important to acidbase homeostasis, it is not entirely
clear why this is so. We suggest that the physicochemical
approach to acidbase provides a more coherent picture of
the role played by NH
4
+
.
The Stewart approach to acidbase chemistry
Here we consider the approach to acidbase chemistry
proposed by PA Stewart [18,19]. Biological fluids are
dominated by a high concentration of water, approximately
55 mol/l. Physical chemistry determines the dissociation of
water into protons and hydroxyl ions. If the determinants of
that equilibrium are unchanged, then concentration of
protons, and therefore pH, will be as well.
A number of important substances (e.g. many salts)
dissociate completely to ions, when dissolved in water,
whereas water itself dissociates to a very minor degree.
Nonetheless, the dissociation of water into H
+
and OH
provides an inexhaustible source and sink of acidbase

equivalents. The proton concentration, and hence pH, is
determined by the requirement that positive and negative
charges must balance and by the combined equations that
govern dissociations of involved species. The approach is
formally based on analysis of separate compartments and leads
to the result that [H
+
] in a compartment of physiological fluid is
determined by the concentrations of fully ionized substances
(strong ion difference [SID]), partial CO
2
tension (PCO
2
) and
partly dissociated substances termed weak acids in that
compartment.
In a solution containing only fully dissociated salt (e.g. NaCl)
the requirement for electrical neutrality leads to the following
relation:
(Na
+
+ H
+
) (Cl
+ OH
) = 0 (1)
The water dissociation equilibrium must also be obeyed:
[H
+
] [OH
] = K
w
[H
2
O] K
w
(2)
The SID is defined as the difference between fully
dissociated cations and anions, and in the NaCl solution it is
calculated as follows:
SID = [Na
+
] [Cl
] (3)
Combining Eqns 1, 2 and 3 leads to the following relation:
[H
+
]
2
+ SID [H
+
] K
w
= 0 (4)
The positive solution to this second-degree polynomial yields:
SID
[H
+
] = + [K
w
+ (SID/2)
2
] (5)
2
575
And from Eqn 2:
SID
[OH
] = + [K
w
+ (SID/2)
2
] (6)
2
Hence, in a compartment/solution containing NaCl or similar
salt solution, the proton concentration is simply determined
by SID and the water ion product (K
w
). Addition or removal of
protons or hydroxyl ions may or may not be possible but will
not change pH [20].
It is possible that the development of Stewart concepts to
this extent will suffice for analysis of renal influences on
acidbase homeostasis from a whole body or balance
perspective. However, to present the theory of Stewart in a
more complete form, we may also add weak acids and CO
2
to this framework. A full account of the Stewart approach
with some later adaptations is available in a previous issue of
this journal (see the report by Corey [21]).
Adding a weak acid, specifically a substance that participates
in proton exchanges and hence that has a charge that is
dependent on pH, Stewart showed that Eqn 7 had to be
satisfied.
[H
+
]
3
+ (KA + SID) [H
+
]
2
+ (KA
[SID A
TOT
] Kw) [H
+
] KA Kw = 0 (7)
Where KA is the equilibrium constant and A
TOT
is the total
concentration of weak acids. To arrive at a satisfactory
explanation for acidbase homeostasis from the whole body
perspective, the pervasive effect of continuing production
and transport and pulmonary excretion of CO
2
evidently must
be taken into account. To do this, two more equations were
needed:
[H
+
] [HCO
3
] = KC PCO
2
(8)
[H
+
] [CO
3
2
] = K3 [HCO
3
] (9)
Solving these together, Stewarts model in its most
integrative form is now given by Eqn 10:
[H
+
]
4
+ ([SID] + KA) [H
+
]
3
+
(KA [[SID] [A
TOT
]] KW KC PCO
2
)
[H
+
]
2
(KA [KW + KC PCO
2
] K3 KC PCO
2
)
[H
+
] KA K3 KC PCO
2
= 0 (10)
These equations have explicit entries of constants and
concentrations or tensions, but the practical use of the
framework must be developed with detail sufficient to deal
with the problem at hand. In plasma, other strong ions (e.g.
Ca
2+
and lactate) and weak acids are frequently found but
they are treated on an equal footing.
A number of studies have shown that this algebra yields an
accurate description or prediction of acidbase measure-
ments. More importantly, however, the physicochemical
approach may lead to a better understanding of mechanisms
that are active in disease and treatment. An example of what
may be accomplished is the successful application of the
physicochemical approach to exercise physiology. Here, the
ability of the independent variables to predict measured pH
has been proven (correlation 0.985), but more importantly
changes over time and between the different body
compartments in these independent variables explain how a
range of interventions influence acidbase as a part of
muscle physiology [22].
CO
2
is transported in the body as a number of species and
because the processes involved have variable latency (e.g.
the Cl
/HCO
3
exchanger band3 in red blood cells [23]),

widely differing values of PCO
2
are found in the body [24].
The physicochemical approach, focusing as it does on each
compartment separately and having no special interest in the
quantitatively lesser compartment of arterial blood, is at no
disadvantage relative to conventional concepts in elucidating
this difficult area. Although this is less of a problem when
overall renal regulation of acidbase homeostasis is
considered, notwithstanding that urine CO
2
may be of utility
when diagnosing variants of RTA [25], it is a major problem
with respect to understanding the underlying cellular
transport processes. Further, recent results showing the
complicated organization of transporters together in
physically connected complexes indicate that much work will
be needed if we are to understand the integrated molecular
details of anion transport and CO
2
metabolism in renal
tubules [26].
Whereas the physicochemical approach explains how pH is
determined from independent variables, when applying this to
urine the focus is not on regulation of urine pH but on the
renal regulation of the independent variables that determine
plasma and whole body acidbase balance. These
independent variables are the SID, weak acids, and PCO
2
.
Hence, from the point of view of the physicochemical
approach, assessing urine with the aim of understanding the
renal contribution to acidbase balance amounts to deducing
its effects on the independent variables for a specified body
compartment. It has been reported that the concepts of SID
and weak acids may be blurred. For example, pH may
influence the behaviour of species as either strong ions
(components of SID) or weak acids [27], and this applies, for
instance, to phosphates and proteins. Furthermore, neither
Na
+
nor Ca
2+
is invariably and totally dissociated, as implied
by the common SID construct [28].
One important but thus far undeveloped aspect of the
Stewart approach to whole body acid balance problems is
that the independent variables for the extracellular
compartment normally in focus may be only partly relevant to
576
the much larger intracellular compartment. Excretion of large
amounts of potassium, for example, may be minimally relevant
to SID in the extracellular compartment but may, depending
on the circumstances, be crucial to intracellular SID [29].
It is evident that there will be differences in the approach to
accounting for acidbase balance in the classical compared
with the physicochemical approach. In the classical setting
we must perform difficult titrations [4] and measurements of
NH
4
+
, PCO
2
and pH to compute a [HCO
3
] after correction of
pK for ionic strength. Every part of this is complicated, and
the overall results with regard to our understanding of whole
body balance are not universally accepted [4]. In the
physicochemical approach, renal involvement in acidbase
balance is manifested in its influence on independent
variables nothing more and nothing less. For a first
approximation, this is the urine excretion of SID components,
principally Na
+
and Cl
when extracellular homeostasis alone

is considered. It will be a practical matter to determine the
extent to which the Stewart approach will be complicated by
problems in computing both SID and weak acids in urine.
In the physicochemical approach, the urinary excretion of
NH
4
+
or organic anions will be important for acidbase
balance only to the extent that it influences SID in a body
compartment. Excretion of organic anions is from this
perspective a way to excrete Na
+
without Cl
and thereby
decrease SID in the body. This will result in increasing plasma
H
+
, no matter what the nature of the organic anion is. This
hypothesis can be tested experimentally. On a similar footing,
NH
4
+
excretion could be understood as means to excrete Cl
without Na
+
in order to increase SID in the body. However,
apart from their influence on SID, the excretion of these
substances may convey important information about
underlying pathophysiological processes. Hence, Kellum [30]
has proposed that, when analyzing the mechanism of
hyperchloraemic acidosis, an initial distinction could be made
between states in which the kidney reacted normally (i.e.
increasing the excretion of Cl
relative to Na
+
and K
+
by
augmenting NH
4
+
excretion and so causing urine SID to be
more negative) and situations where, in spite of acidosis, the
kidney continues to decrease whole body SID by excreting
more Na
+
and K
+
than Cl
. This will typically be the case in

distal RTA (dRTA) without increased NH
4
+
excretion during
acidosis.
Overview on renal tubular acidoses
Several types of RTA may be discerned [31]: proximal
(type 2), distal (type 1), mixed (type 3), and a heterogeneous
group of disorders characterized by hyperkalaemia and
acidosis (type 4). RTA is a hyperchloraemic rather than an
anion-gap-type metabolic acidosis. Typically, renal function
(glomerular filtration rate) is unimpaired and the acidosis is
not simply caused by absence of renal clearance. RTA must
be separated from other forms of hyperchloraemic acidosis,
some of which (e.g. the hyperchloraemic acidosis that occurs
following saline infusion) are very important in the intensive
care setting [32,33].
Proximal renal tubular acidosis (type 2)
Proximal RTA is classically characterized by impaired proximal
reclamation of bicarbonate. This may be isolated or combined
with other proximal tubular defects, and it may be congenital
or acquired.
Proximal bicarbonate reabsorption is still incompletely under-
stood [34]. Most of the bicarbonate [35] leaves the tubule
lumen as CO
2
following sodium dependent H
+
secretion via
Na
+
/H
+
exchanger isoforms or (to a minor extent) vacuolar
H
+
-ATPase, apical anion exchange via formate enhanced
Slc26a6, or other mechanisms [36], but some bicarbonate
transport may also be paracellular [37]. The transport
requires both membrane bound carbonic anhydrase (CA)
type 4 and intracellular CA-2.
Among hereditary forms of RTA type 2 [38] is a very rare
autosomal dominant disorder, the mechanism of which is
unknown, but isoform 3 of the Na
+
/H
+
exchanger (solute
carrier [SLC]9A3) is a candidate. More common is an
autosomal recessive form with ocular abnormalities, related to
mutations in kidney Na
+
/HCO
3
cotransporter (kNBC)1
(SLC4A4) gene, which encodes the basolateral, electrogenic
Na
+
/3(HCO
3
) cotransporter. kNBC1 activity leads to a

depolarization of the membrane and to extracellular
accumulation of HCO
3
. A recently identified potassium

channel, named TASK2, recycles K
+
and repolarizes the
potential, and mice that are deficient in this channel had
metabolic acidosis associated with insufficient proximal
bicarbonate reabsorption [39]. Recent studies of the
regulation of kNBC1 and integrated transport in the proximal
tubule have shown that, in addition to a substrate interaction,
there is also a true macromolecular interaction between CA-2
and kNBC1 [40].
Sporadic forms, which are not yet characterized, also occur.
However, most cases of proximal RTA are secondary and a
host of associations have been described. Blockade of CA-4
by acetazolamide leads predictably to proximal RTA.
Important are other genetic diseases that cause a generalized
proximal tubular syndrome (Fanconis; e.g. cystinosis, fructose
intolerance, etc.) and drugs and toxins (e.g. ifosfamide [41],
lead, mercury and cadmium), but light chain disease occurs
among the elderly with proximal RTA. A number of
medications have been related to proximal RTA [42].
Characteristic of proximal RTA is the presence of bicarbona-
turia, with a fractional bicarbonate excretion of more than
15% when bicarbonate is given. Eventually, acidbase
balance and urine acidification is achieved as plasma
bicarbonate drops low enough for reabsorption to keep pace.
Treatment may be difficult because administered base is
often excreted before the desired normalization is achieved.
577
Explaining acidosis in proximal RTA from the conventional
point of view is straightforward because the defining loss of
urinary bicarbonate will inevitably deplete the body and result
in hyperchloraemic acidosis. From the point of view of the
physicochemical approach, the reciprocal retention of Cl
and resulting decline in SID will also explain the findings.

In the conventional notion of acidbase regulation, proximal
bicarbonate reabsorption is thought to be regulated by pH.
However, based on studies of bicarbonate transport in the
perfused rabbit proximal tubules, Boron and coworkers [43]
concluded that the observed regulation would require both a
CO
2
sensor and a HCO
3
sensor. A pH sensor would not be

enough. Stoichiometrically, a HCO
3
sensor transmits the

same information as a hypothetical SID sensor, and the
results thus indicate that the proximal tubule senses the two
important independent variables in the Stewart model. These
quite new results could indicate that the physicochemical
approach is highly relevant to our understanding of the
mechanisms that underlie regulation of acidbase physiology.
Distal renal tubular acidosis (type 1)
dRTA is characterized by impaired ability to acidify the urine
in the distal tubules and it is often accompanied by hypo-
kalaemia, low urinary NH
4
+
and hypocitraturia. In contrast to
proximal RTA, nephrocalcinosis and nephrolithiasis frequently
occur. Clinically, dRTA occurs as a primary (persistent or
transient) or secondary disorder. Secondary dRTA occurs in
a great number of circumstances related to autoimmune
diseases, drugs and toxins, and genetic or structural
disruptions of renal tubules. Treatment of dRTA is simple and
involves substituting about 1 mEq/kg of alkali per day.
The molecular details of some forms of primary dRTA are
being pursued in great detail. -Intercalated cells secrete H
+
by means of a vacuolar-type H-ATPase [44] (and possibly
also a H
+
/K
+
-type ATPase), and bicarbonate is exchanged for
Cl
by means of anion exchanger (AE1) at the basolateral

side. An autosomal dominant form of mutation in 17q21-22 of
SLC4A1 leads to dysfunction of AE1 possibly related to
mistargeting of the protein [45]. Also, AE1 mutations causing
autosomal recessive dRTA and haemolytic anaemia have
been described [46]. Otherwise, recessive forms of dRTA are
related to mutations in the proton pump in -intercalated
cells. Some are accompanied by sensorineural deafness. The
gene involved (ATP6V1B1) is located on chromosome 2, and
encodes the B1-subunit of H
+
-ATPase expressed apically on
-intercalated cells and also in the cochlea. dRTA with less
impaired hearing is related to mutation in ATP6V0A4 on
chromosome 7, which encodes a4, an accessory subunit of
H
+
-ATPase. As far as presently known, the H
+
pumps are
electrogenic and, at least under some circumstances, they
also involve shunting of the potential by Cl
, although reverse
transport of K
+
may also occur [44,47]. The Cl
shunt
pathway has not been elucidated yet nor aligned with any of
the many known Cl
channels [44]. Likewise, functional Cl
channels (CIC5) are necessary to acidify transport vesicles in

Dents disease, pointing to the link between H
+
and Cl
transport [48].
Jentsch and coworkers [49] recently presented a detailed
examination of a mouse model that was knocked out for a
K
+
/Cl
cotransporter, KCC4, which is located in the baso-

lateral membrane in -intercalated cells in the collecting duct.
These animals had metabolic acidosis with alkaline urine, but
electrolyte excretion in urine was unchanged compared with
controls. The investigators measured a high intracellular [Cl
]
and inferred a high intracellular pH also, driven by the basal
HCO
3
/Cl
exchanger AE1. Although intracellular pH was not

actually measured, and the defective cotransporter would be
expected also to result in increased intracellular [K
+
], the
results seem difficult to reconcile with a dominant effect of
intracellular SID to set intracellular pH and with the notion
that urine SID will have to change to explain acidosis in RTA.
Details are awaited for this model; the authors also failed to
document that conventional accounting for acidbase
balance would explain the findings (decreased NAE would
also change electrolyte excretion).
Recently, examination of the dRTA that is sometimes seen in
cyclosporine A treatment has led to deeper insights into the
tubular handling of protons and bicarbonate, but also and
importantly that of Cl
. In a study [50] of perfused rabbit

collecting ducts, cyclosporine A inhibited acidosis induced
downregulation of unidirectional HCO
3
secretory flux in -
intercalated cells and prevented downregulation of the linked
Cl
resorption. Detailed examination of the apical and

basolateral exchanges indicates that, rather than responding
to, for example, intracellular pH, intracellular [Cl
] could be
the regulated entity [51]. If true, this interpretation is
compatible with a Stewart-based perspective.
A number of drugs and chemicals (e.g. amphotericin B [52],
foscarnet and methicillin) have been found occasionally to
cause dRTA [42], although details of the underlying
mechanisms are not available.
Type 3 renal tubular acidosis (carbonic anhydrase
dysfunction)
Type 3 RTA is caused by recessive mutation in the CA-2
gene on 8q22, which encodes carbonic anhydrase type 2
[53]. It is a mixed type RTA that exhibits both impaired
proximal HCO
3
reabsorption and impaired distal acidifi-

cation, and more disturbingly osteopetrosis, cerebral calcifi-
cation and mental retardation. The mechanisms that underlie
the clinical picture in type 3 RTA, apart from much slower
conversion of carbonic acid to and from bicarbonate,
apparently also involve direct interaction between CA and the
Na
+
/HCO
3
cotransporter kNBC1 [54] or Cl
/HCO
3
exchanger SLC26A6 [55]. From the physicochemical inter-

pretation, acidosis is expected under these circumstances
because of impaired transport of SID components.
578
Type 4 (hyperkalaemic) renal tubular acidosis
RTA type 4 or hyperkalaemic RTA is a heterogeneous group of
disorders that is characterized by low urine NH
4
+
, which is
probably caused by the hyperkalaemia or by aldosterone
deficiency or defective signalling. Causes include various types
of adrenal failure or pseudohypoaldosteronism (PHA)1 due to
defects in the mineralocorticoid receptor or the epithelial Na
+
channel, all characterized by salt loss and hypotension. A
similar picture may be seen in obstructive uropathy or drug-
induced interstitial nephritis. Furthermore, a number of drugs
may impair signalling in the reninaldosterone system and
cause hyperkalaemia and metabolic acidosis (e.g. potassium
sparing diuretics, trimethoprim, cyclo-oxygenase inhibitors,
angiotensin converting enzyme inhibitors).
Lately, much interest has been given to a group of rare
autosomal dominant diseases characterized by hyperkalaemia
and acidosis and age-related hypertension [56]. In spite of
hypervolaemia, aldosterone is not low and the disorders have
been collectively termed pseudohypoaldosteronism type 2
(PHA2) [57]. Two of the mutations have been mechanistically
characterized in some detail. Mutations in 17q21 in the
WNK4 gene may change the function of the protein, whereas
a mutation in the intron to the WNK1 gene at 12p increases
transcription of the protein. Briefly, WNK4 normally inhibits
the thiazide-sensitive cotransporter (TSC) in the distal
convolute tubule (DCT), and inhibits the renal outer medullar
K
+
channel (ROMK) in the collecting duct (CD), but
enhances paracellular Cl
transport in both DCT and CD.

Mutations in the WNK4 gene that cause PHA2 are found to
release the normal inhibition of TSC, but at the same time
PHA2 enhances the inhibition of ROMK and enhances the
paracellular Cl
flux (but not Na

+
flux) through claudins.
Hence, the hyperkalaemia is explained both by inhibition of
ROMK and by decreased delivery of Na
+
to CD because of
enhanced absorption in the DCT, and the good effect of
thiazides on the hypertension is readily explained. The normal
explanation for metabolic acidosis is based on the decreased
delivery of Na
+
to CD and thereby inhibition of generation of
lumen negative potential to enhance H
+
secretion in
combination with the decreased delivery of NH
4
+
secondary
to the hyperkalaemia [58].
The effect of the molecular abnormalities on Cl
transport is
barely considered in the explanation of the findings using the
conventional model of acidbase. From the physicochemical
approach it is evident that acidosis is well explained by the
dominant and primary enhancement of Cl
absorption in this
disorder. Even if only the TSC effect were invoked, an
isotonic expansion of body volume with Na
+
and Cl
would
be expected to yield acidosis. In any case, SID in plasma will
decrease and pH will too. Very recently it was described that
WNK1 activates the epithelial Na
+
channel [59], and this was
felt to explain the finding that not all patients with PHA2 are
equally sensitive to thiazides. This would be expected to
relieve the voltage imposed inhibition of H-ATPase in CD and
likewise lessen the degree of hyperkalaemia. Electrolyte and
NAE balance studies across different mutations may help to
clarify how acidbase balance is actually constructed in
these rare diseases.
Diagnosis and differential diagnosis
Traditionally, dRTA is recognized by the inability to decrease
urine pH below 5.5 in spite of metabolic acidosis. These
patients are also characterized by an inability to augment
NH
4
+
excretion [60]. A high urine PCO
2
after bicarbonate
loading has traditionally been the criterion for declaring distal
H
+
secretion to be normal [61], and it was also recently found
to identify patients with confirmed dRTA due to a proton
pump problem [25].
Proximal RTA is characterized by high fractional excretion of
bicarbonate (>15%) during loading, and an ability to achieve
a urine pH below 5.5 during acidosis. Approaches are well
described by Soriano [31] and Smulders and coworkers [62].
When assessing urine to gauge whether the physicochemical
approach or the classical theory is best able to explain the
acidosis in RTA, it is possible that both will do so
successfully. From the physicochemical approach, the lack of
urine NH
4
+
in distal RTA will force excretion of urine with a
relatively high SID and this will explain the acidosis. An old
study did in fact indicate that, in type 1 RTA, Na
+
loss and to
a lesser degree Cl
handling was abnormal in spite of long-

term correction of acidosis [63].
The classical theory also explains the acidosis by a lack of
amplification of NH
4
+
excretion. Likewise, for proximal RTA
bicarbonate loss and high SID excretion will be equivalent. It
was recently suggested that even though it may be difficult
mechanistically to separate the implications of the theories,
by using the physicochemical approach the focus is forced
toward movements of Na
+
and Cl
, and this may lead to a

new understanding [2]. Indeed, analysis of WNK mutations
confirms this expectation.
Conclusion
From the clinical viewpoint, the advantage of employing the
physicochemical approach is that the renal contribution to
acidbase homeostasis, even in complicated settings, can be
ascertained in principle by simple chemical analysis of the
urine. It is possible to explain RTA in general as a
hyperchloraemic form of metabolic acidosis that can be
described as a low SID acidosis, which has focused attention
primarily on the net handling of SID constituents, namely Na
+
,
K
+
, and Cl
. This handling of SID constituents has not had a

central position in our understanding of the various disease
states, and in some cases only seems to be a consequence
of anions necessarily being filled in by Cl
as HCO
3
goes
down and reversely. However, in the future efforts will focus
on which transport mechanism is active (e.g. is Cl
moving
with H
+
or K
+
or against it to shunt the potential generated by
579
the vacuolar H-ATPase [44]) and on which moiety is actually
regulated by the tubular processes. A number of studies have
recently focused on apical anion handling in the collecting
duct via a newly characterized transporter, namely pendrin
[64]. This exchanger seems well poised to react to Cl
balance [65] and could therefore also be sensitive to the

independent variable in acidbase regulation (i.e. SID) [66].
One defining point in the physicochemical approach that has
an impact on the interpretation of acidbase phenomena is
the concept of [H
+
] as a dependent variable, which tends to
imply that clinical or physiological phenomena might more
fundamentally depend on the baseline independent variables
(e.g. SID, weak acids and PCO
2
). The necessity when
analyzing renal phenomena to differentiate metabolic and
respiratory acidosis may be an indicator that pH as such is
not actually the sensed quantity.
In fact, how derangements in acidbase balance are sensed
by the kidneys remains elusive, although there it is a general
belief that such detection happens there. Quite recently, a
protein, Pyk2, that was sensitive to pH and that regulated
isoform 3 of the Na
+
/H
+
exchanger in the proximal tubules
was described [67]. Furthermore, in experiments identifying
this alleged pH sensor, SID was directly varied but PCO
2
did
not change. Hence, it is not evident that pH was really
sensed, and in an accompanying editorial Gluck [68]
expressed reservations regarding this notion. As explained
above in relation to proximal RTA, recent studies conducted
by Boron and coworkers [43] indicate that that bicarbonate
and PCO
2
are the regulated entities, rather than pH, which is
in accordance with the physicochemical approach to acid
base physiology insofar as bicarbonate and SID are
equivalent.
Finally, if whole body acidbase balance is to be untangled,
then the intracellular domains, which are likely to vary, must
also be understood. In exercise physiology [69] advances
have been made using the Stewart approach in elucidating
plasma acidbase balance as it is perturbed by transfer of
putative independent influences, but modelling cells or whole
organs themselves from this point of view has not been done.
This will entail such difficulties as determining water structure
in cells and small confines [70] and modelling the pH effects
of the structural proteins and nucleic acids as they fold and
integrate. Modelling potassium balance in order to draw
inferences regarding intracellular SID will likewise be
necessary and interesting.
A recent study of patients in acute renal failure [71],
employing state of the art methods, found that almost 80% of
total body water appeared to be extracellular. This indicates
that a great deal of experimental work must be done before
analytical solutions [72] to the whole body multicompartment
system can be derived and applied in clinical practice. We
suggest that the physicochemical approach will prove useful
in formulating hypotheses for future work aimed at developing
a coherent, unpretentious and practical understanding of
mechanisms involved in renal acidbase regulation.
Competing interests
References
1. Corey HE: Stewart and beyond: New models of acid-base
2. Laing CM, Toye AM, Caposso G, Unwin RJ: Renal tubular acido-
sis: developments in our understanding of the molecular
basis. Int J Biochem Cell Biol 2005, 37:1151-1161.
3. Lemann J Jr, Bushinsky DA, Hamm LL: Bone buffering of acid
and base in humans. Am J Physiol 2003, 285:F811-F832.
4. Cohen RM, Feldman GM, Fernandez PC: The balance of acid,
base and charge in health and disease. Kidney Int 1997, 52:
287-293.
5. Cheema-Dhadli S, Lin S-H, Halperin ML: Mechanisms used to
dispose of progressively increasing alkali loads in rats. Am J
Physiol 2002, 282:F1049-F1055.
6. Brown JC, Packer RK, Knepper MA: Role of organic anions in
renal response to dietary acid and base loads. Am J Physiol
1989, 257:F170-F176.
7. Kamel KS, Briceno LF, Sanchez MI, Brenes L, Yorgin P, Kooh SW,
Balfe JW, Halperin ML: A new classification for renal defects in
net acid excretion. Am J Kidney Dis 1997, 29:136-146.
8. Burbea Z-H, Gullans SR, Ben-Yaakov S: Alkalinity: a simple
method to measure cellular net acid-base fluxes. Am J Physiol
1987, 253:C525-C534.
9. Osther PJ, Engel K, Kildeberg P: Renal response to acute acid
loading. Scand J Urol Nephrol 2004, 38:62-68.
10. Atkinson DE, Bourke E: Metabolic aspects of the regulation of
systemic pH. Am J Physiol 1987, 252:F947-F956.
11. Nagami GT: Renal ammonia production and excretion. In The
Kidney, Physiology and Pathophysiology, 3rd ed. Edited by
Seldin DW, Giebisch G. Philadelphia: Lippincott Williams &
Wilkins; 2000:1995-2013.
12. Hosch M, Muser J, Hulter HN, Krapf R: Ureagenesis: evidence
for a lack of hepatic regulation of acid-base equilibrium in
humans. Am J Physiol 2004, 286:F94-F99.
13. Gennari FJ, Maddox DA: Renal regulation of acid-base home-
ostasis: integrated response. In The Kidney, Physiology and
Pathophysiology, 3rd ed. Edited by Seldin DW, Giebisch G.
Philadelphia: Lippincott Williams & Wilkins; 2000:2015-2053.
14. Kurtz I, Dass PD, Cramer S: The importance of renal ammonia
metabolism to whole body acid-base balance: a reanalysis of
the pathophysiology of renal tubular acidosis. Miner Elec-
trolyte Metab 1990, 16:331-340.
15. Weiner ID: The Rh gene family and renal ammonium transport.
Curr Opin Nephrol Hypertens 2004, 13:533-540.
16. Frank AE, Weiner ID: Effects of ammonia on acid-base trans-
port by the B-type intercalated cell. J Am Soc Nephrol 2001,
12:1607-1614.
17. Frank AE, Wingo CS, Andrews PM, Ageloff S, Knepper MA,
Weiner ID: Mechanisms through which ammonia regulates
cortical collecting duct net proton secretion. Am J Physiol
2002, 282:F1120-F1128.
19. Stewart PA: How to understand acidbase. A Quantitative acid
base primer for biology and medicine. London: Edward Arnold;
1981.
20. Stewart PA: Independent and dependent variables of
acidbase control. Respir Physiol 1978, 33:9-26.
21. Corey HE: Bench-to-bedside review: fundamental principles
of acid-base physiology. Crit Care 2004, 8:184-192.
22. Lindinger MI, Heigenhauser GJF, McKelvie RS, Jones NL: Blood
ion regulation during repeated maximal exercise and recovery
in humans. Am J Physiol 1992, 262:R126-R136.
23. Ring T, Andersen PT, Knudsen F, Nielsen FB: Salicylate-induced
hyperventilation [letter]. Lancet 1985, i:1450.
24. Geers C, Gros G: Carbon dioxide transport and carbonic
anhydrase in blood and muscle. Phys Rev 2000, 80:681-715.
580
25. Kim S, Lee JW, Park J, Na KY, Joo KW, Ahn C, Kim S, Lee JS, Kim
GH, Kim J, et al.: The urine-blood PCO2 gradient as a diagnos-
tic index of H+-ATPase defect in distal renal tubular acidosis.
Kidney Int 2004, 66:761-767.
26. Purkerson JM, Schwartz GJ: Expression of membrane-associ-
ated carbonic anhydrase isoforms IV, IX, XII, and XIV in the
rabbit: induction of CA IV and IX during maturation. Am J
Physiol 2005, 288:R1256-R1263.
27. Wooten EW: Analytic calculation of physiological acid-base
parameters in plasma. J Appl Physiol 1999, 86:326-334.
net protein charge and Atot and Ka of nonvolatile buffers in
human plasma. Am J Physiol 2003, 95:620-630.
29. Gowrishankar M, Chen CB, Mallie JP, Halperin ML: What is the
impact of potassium excretion on the intracellular fluid volume:
importance of urine anions. Kidney Int 1996, 50:1490-1495.
Crit Care 2000, 4:6-14.
31. Soriano JR: Renal tubular acidosis: the clinical entity. J Am Soc
Nephrol 2002, 13:2160-2170.
32. Kellum JA: Metabolic acidosis in the critically ill: lessons from
physical chemistry. Kidney Int 1998, Suppl 66:s-81-s-86.
of strong ion acidosis. Anesth Surg 2003, 96:919-922.
34. Weinstein AM: Mathematical models of renal fluid and elec-
trolyte transport: acknowledging our uncertainty. Am J Physiol
2003, 284:F871-F884.
35. Petrovic S, Barone S, Weinstein AM, Soleimani M: Activation of
the apical Na+/H+ exchanger NHE3 by formate: basis of
enhanced fluid and electrolyte reabsorption by formate in the
kidney. Am J Physiol 2004, 287:F336-F346.
36. Wang Z, Wang T, Petrovic S, Tuo B, Riederer B, Lorenz JN, Seidler
U, Aronson PS, Soleimani M: Renal and intestinal transport defects
in Slc26a6-null mice. Am J Physiol 2005, 288:C957-C965.
37. Guo P, Weinstein AM, Weinbaum S: A dual-pathway ultrastruc-
tural model for the tight junction of rat proximal tubule epithe-
lium. Am J Physiol 2003, 285:F241-F257.
38. Igarashi T, Sekine T, Inatomi J, Seki G: Unravelling the molecu-
lar pathogenesis of isolated proximal renal tubular acidosis. J
Am Soc Nephrol 2002, 13:2171-2177.
39. Warth R, Barrire H, Meneton P, Bloch M, Thomas J, Tauc M,
Heitzmann D, Romeo E, Verrey F, Mengual R, et al.: Proximal
renal tubular acidosis in TASK2 K+ channel-deficient mice
reveals a mechanism for stabilizing bicarbonate transport.
Proc Natl Acad Sci USA 2004, 101:8215-8220.
40. Pushkin A, Abuladze N, Gross E, Newman D, Tatishchev S, Lee I,
Fedotoff O, Bondar G, Azimov R, Nguyen M, et al.: Molecular
mechanism of kNBC1-carbonic anhydrase II interaction in
proximal tubule cells. J Physiol 2004, 559:55-65.
41. Skinner R: Chronic ifosfamide nephrotoxicity in children. Med
Pediatr Oncol 2003, 41:190-197.
42. Hemstreet BA: Antimicrobial-associated renal tubular acidosis.
Ann Pharmacother 2004, 38:1031-1038.
43. Zhou Y, Zhao J, Bouyer P, Boron WF: Evidence from renal prox-
imal tubules that HCO3- and solute transport are acutely reg-
ulated not by pH but by basolateral HCO3- and CO2. Proc Natl
Acad Sci USA 2005, 102:3875-3880.
44. Wagner CA, Finberg KE, Breton S, Marshansky V, Brown D,
Geibel JP: Renal vacuolar H+-ATPase. Physiol Rev 2004, 84:
1263-1314.
45. Devonald MAJ, Karet FE: Renal epithelial traffic jams and one-
way streets. J Am Soc Nephrol 2004, 15:1370-1381.
46. Karet FE: Inherited diatal renal tubular acidosis. J Am Soc
Nephrol 2002, 13:2178-2184.
47. Paroutis P, Touret N, Grinstein S: The pH of the secretory
pathway: measurement, determinants, and regulation. Physiol-
ogy 2004, 19:207-215.
48. Jentsch TJ: Chloride transport in the kidney: lessons from
human disease and knockout mice. J Am Soc Nephrol 2005,
16:1549-1561.
49. Boettger T, Hbner CA, Maier H, Rust MB, Beck FX, Jentsch TJ:
Deafness and renal tubular acidosis in mice lacking the K-Cl
cotransporter Kcc4. Nature 2002, 416:874-878.
50. Watanabe S, Tsuruoko S, Vijayakumar S, Fischer G, Zhang Y,
Fujimura A, Al-Awqati Q, Schwartz GJ: Cyclosporin A produces
distal renal tubular acidosis by blocking peptidyl prolyl cis-trans
isomerase activity of cyclophilin. Am J Physiol 2005, 288:F40-F47.
51. Schwartz GJ, Tsuruoka S, Vijayakumar S, Petrovic S, Mian A, Al-
Awqati Q: Acid incubation reverses the polarity of intercalated
cell transporters, and effect mediated by hensin. J Clin Invest
2002, 109:89-99.
52. Goldman RD, Koren G: Amphotericin B nephrotoxicity in chil-
dren. J Pediatr Hematol Oncol 2004, 26:421-426.
53. Shah GN, Bonapace G, Hu PY, Strisciuglio P, Sly WS: Carbonic
anhydrase II deficiency syndrome (osteopetrosis with renal
tubular acidosis and brain calcification): novel mutations in
CA2 indentified by direct sequencing expand the opportunity
for genotype-phenotype correlation. Hum Mutat 2004, 24:272.
54. Pushkin A, Abuladze N, Gross E, Newman D, Tatishchev S, Lee I,
Fedotoff O, Bondar G, Azimov R, Ngyuen M, et al.: Molecular
mechanisms of kNBC1-carbonic anhydrase II interaction in
proximal tubule cells. J Physiol 2004, 559:55-65.
55. Alvarez BV, Vilas GL, Casey JR: Matabolon disruption: a mech-
anism that regulates bicarbonate transport. EMBO J 2005, 24:
2499-2511.
56. Gamba G: Role of WNK kinases in regulating tubular salt and
potassium transport and in the development of hypertension.
Am J Physiol 2005, 288:F245-F252.
57. Kahle KT, Wilson FH, Lifton RP: Regulation of diverse ion trans-
port pathways by WNK4 kinase: a novel molecular switch.
Trends Endocrinol Metab 2005, 20:1-6.
58. DuBose TD, Good DW: Chronic hyperkalemia impairs ammo-
nium transport and accumulation in the inner medulla of the
rat. J Clin Invest 1992, 90:1443-1449.
59. Xu BE, Stippec S, Chu PY, Lazrak A, Li XJ, Lee BH, English JM,
Ortega B, Huang CL, Cobb MH: WNK1 activates SGK1 to regu-
late the epithelial sodium channel. Proc Natl Acad Sci USA
2005, 102:10315-10320.
60. Halperin ML, Richarson RMA, Bear RA, Magner PO, Kamel K,
Ethier J: Urine ammonium: the key to the diagnosis of diatal
renal tubular acidosis. Nephron 1988, 50:1-4.
61. DuBose TD, Caflisch CR: Validation of the difference in urine
and blood carbon dioxide tension during bicarbonate loading
as an index of distal nephron acidification in experimental
models of distal renal tubular acidosis. J Clin Invest 1985, 75:
1116-1123.
62. Smulders YM, Frissen PHJ, Slaats EH, Silberbusch J: Renal
tubular acidosis. Pathophysiology and diagnosis. Arch Intern
Med 1996, 156:1629-1636.
63. Sebastian A, McSherry E, Morris RC Jr: Impaired renal conser-
vation of sodium and chloride during sustained correction of
systemic acidosis in patients with type 1, classic renal tubular
acidosis. J Clin Invest 1976, 58:454-469.
64. Frische S, Kwon T-H, Frkir J, Madsen KM, Nielsen S: Regu-
lated expression of pendrin in rat kidney in response to
chronic NH4Cl or NaHCO3 loading. Am J Physiol 2003, 284:
F584-F593.
65. Wall SM, Kim YH, Stanley L, Glapion DM, Everett LA, Green ED,
Verlander JW: NaCl restriction upregulates renal Slc26a4
through subcellular redistribution. Role in Cl- conservation.
Hypertension 2004, 44:982-987.
66. Quentin F, Chambrey R, Trinh-Trang-Tan MM, Fysekidis M, Cam-
billau M, Paillard M, Aronson PS, Eladari D: The Cl-/HCO3-
exchanger pendrin in the rat kidney is regulated in response
to chronic alterations in chloride balance. Am J Physiol 2004,
287:F1179-F1188.
67. Li S, Sato S, Yang X, Preisig PA, Alpern RJ: Pyk2 activation is
integral to acid stimulation of sodium/hydrogen exchanger 3.
J Clin Invest 2004, 114:1782-1789.
68. Gluck SL: Acid sensing in renal epithelial cells. J Clin Invest
2004, 114:1696-1699.
69. Putman, CT, Jones NL, Heigenhauser GJF: Effects of short-term
training on plasma acid-base balance during incremental
exercise in man. J Physiol 2003, 550:585-603.
70. Truskett TM: The subtleties of water in small spaces. Proc Natl
Acad Sci USA 2003, 100:10139-10140.
71. Ikizler TA, Sezer MT, Flakoll PJ, Hariacher S, Kanagasundaram
NS, Gritter N, Knights S, Shyr Y, Paganini E, Hakim RM, et al.:
Urea space and total body water measurements by stable
isotopes in patients with acute renal failure. Kidney Int 2004,
65:725-732.
72. Wooten EW: Calculation of physiological acid-base parame-
ters in multicompartmental systems with application to
human blood. J Appl Physiol 2003, 95:2333-2344.
500
AG = anion gap; AGc = corrected anion gap; A
TOT
= total weak acids; BE = base excess; PCO
2
= partial carbon dioxide tension; SBE = standard
base excess; SID = strong ion difference; SIG = strong ion gap; Vd = volume of distribution.
Critical Care October 2005 Vol 9 No 5 Kellum
Abstract
Recent advances in acidbase physiology and in the epidemiology
of acidbase disorders have refined our understanding of the basic
control mechanisms that determine blood pH in health and
disease. These refinements have also brought parity between the
newer, quantitative and older, descriptive approaches to acid
base physiology. This review explores how the new and older
approaches to acidbase physiology can be reconciled and
combined to result in a powerful bedside tool. A case based
tutorial is also provided.
Introduction
During the past 5 years, numerous publications have
examined various aspects of acidbase physiology using
modern quantitative acidbase chemistry. These studies have
refined our understanding of the basic control mechanisms
that determine blood pH in health and disease, and have
described the epidemiology and clinical significance of
acidbase imbalances in far more detail than was previously
possible. Furthermore, these refinements have brought into
parity quantitative and descriptive approaches to acidbase
physiology, and permit translation of the old into the new.
Indeed, these advances have established that the modern
(quantitative) and traditional (descriptive) approaches are, in
fact, easily interchangeable at the level of their most basic
elements, with a little mathematical manipulation. This
interchange has in turn resulted in an explication of the
limitations of each approach and has revealed how a
combined approach can be used to achieve a more complete
understanding of clinical acidbase physiology.
These new insights have further called into question some
basic clinical interpretations of acid-base physiology while at
the same time supporting the underlying chemistry. For
example, it is now possible to understand and apply the
variables of strong ion difference (SID) and total weak acids
(A
TOT
) entirely within the context of BronstedLowry
acidbase chemistry [1-5]. However, it remains difficult to
reconcile how alterations in plasma pH can be brought about
by direct manipulations of hydrogen or bicarbonate ions, as
the descriptive approaches suggest (although do not
require), when they are dependent variables according to
quantitative acidbase chemistry. Newer approaches such as
ion equilibrium theory [1,2] can perhaps reconcile these
differences by not requiring independent variables, but it is
likely that advances in our understanding of pathophysiology
will favor one interpretation or the other. For example, the
discovery of genetic polymorphisms that alter the function of
chloride channels being associated with renal tubular
acidosis [6] favors the quantitative explanation. Nevertheless,
observations detailed using descriptive approaches are no
less valid. One way to unify acidbase physiology is merely to
acknowledge that descriptive indices such as standard base
excess (SBE) and the HendersonHasselbalch equation are
useful for describing and classifying acidbase disorders,
whereas quantitative indices such as SID and A
TOT
are more
useful for quantifying these disorders and for generating
hypotheses regarding mechanisms.
This review explores how acidbase reunification is possible
and even desirable, and how a unified approach can be more
powerful than any of its parts. This unified field answers many
stubborn questions and simplifies bedside interpretation to
the point that every practising intensivist should be aware of
its essential components. Finally, a detailed review of a
complex yet typical case is used to reinforce these concepts.
Acidbase reunification
There are three widely used approaches to acidbase
physiology using apparently different variables to assess
changes in acidbase balance (Fig. 1). In fact, each variable
can be derived from a set of master equations and complete
Review
Clinical review: Reunification of acidbase physiology
John A Kellum
The CRISMA (Clinical Research Investigation and Systems Modeling of Acute Illness) Laboratory, Department of Critical Care Medicine, University of
Pittsburgh, Pittsburgh, Pennsylvania, USA
Corresponding author: John A Kellum, kellumja@ccm.upmc.edu
501
parity can be brought to all three acidbase approaches. This
is because acidbase balance in plasma is based upon
thermodynamic equilibrium equations [2]. The total concen-
tration of proton acceptor sites in a solution (C
B
) is given by
the following equation:
C
B
= C +
i
C
i
e
i
D (1)
where C is the total concentration of carbonate species
proton acceptor sites (in mmol/l), C
i
is the concentration of
noncarbonate buffer species i (in mmol/l), e
i
is the average
number of proton acceptor sites per molecule of species i,
and D is Riccis difference function (D = [H
+
] [OH
]). Thus,
Eqn 1 may be regarded as a master equation from which all
other acidbase formulae may be derived [2].
It is no wonder, in terms of describing acidbase
abnormalities and classifying them into various groups, that
the three widely accepted methods yield comparable results
[7]. Importantly, each approach differs only in its assessment
of the metabolic component (i.e. all three treat partial carbon
dioxide tension [PCO
2
] the same). These three methods
quantify the metabolic component by using the relationship
between HCO
3
and PCO
2
(method 1), the SBE (method 2),
or the SID and A
TOT
(method 3). All three yield virtually
identical results when they are used to quantify the acidbase
status of a given blood sample [1,4,8,9], with an increasingly
complex rule set going from method 3 to method 1 [10,11].
In quantitative acidbase chemistry (method 3), a complete
rule set is provided in the form of equilibrium equations
[12,13], so the approach is easily adapted to modern handheld
computer devices [14] and more sophisticated graphical
interfaces [15]. However, this does not in itself necessarily
make the approach any better [4,5], although it is by definition
more transparent and therefore more easily reproduced. The
difficulty with the quantitative approach comes from the fact
that several variables are needed, and when they are absent
and assumed to be normal the approach becomes essentially
indistinguishable from the more traditional descriptive methods.
Of course, this only applies to quantifying and classifying an
acidbase disorder. The quantitative approach has important
implications for our understanding of mechanisms, leading to
conclusions that are at odds with more traditional thinking (e.g.
viewing renal tubular acidosis as chloride channelopathies).
However, in the absence of specific experimental data, the
method can only imply causality it cannot establish it.
Furthermore, all three approaches predict the exact same
changes in all of the relevant variables and, because these
changes occur nearly instantaneously, determining which
variable is causal is extremely difficult. An often used analogy is
that the naked eye can observe the movement of the sun in
reference to the Earth, but without additional observations (via
Galileos telescope) or mathematical models (ala Copernicus)
it is impossible to say which body is in motion [16,17]. In the
case of acidbase physiology multiple variables move, making
the analysis that much more difficult.
In the end, all approaches to acidbase analysis are just
tools. Their usefulness is best evaluated by examining the
predictions that they make and how well they conform to
experimental data. For example, by using only the
HendersonHasselbalch equation a linear relationship
between pH and log PCO
2
should exist, but actual data
demonstrate nonlinear behavior [18]. In order to fit the
HendersonHasselbalch equation to experimental data,
terms for SID and A
TOT
must be added [2,18].
[SID] K
a
[A
TOT
]/[K
a
+ 10
pH
]
pH = pK
1
+ log (2)
SP
CO2
Here, K
1
is the equilibrium constant for the Henderson
Hasselbalch equation, K
a
is the weak acid dissociation
constant, and S is the solubility of CO
2
in plasma. Similarly,
one can predict changes in plasma bicarbonate resulting
from addition of sodium bicarbonate using its estimated
volume of distribution (Vd). Under normal conditions the Vd
for bicarbonate in humans has been estimated to be 4050%
of total body water [19]. However, the calculated Vd for
bicarbonate changes with changes in pH [20], and Vd
changes differently with respiratory versus metabolic
acidbase derangements [21]. Treating bicarbonate as a
dependent variable and predicting the changes with sodium
bicarbonate as a result of the effect on sodium on SID
requires none of these complicating rules and exceptions,
and might therefore be viewed as much simpler.
Updating base excess
As early as the 1940s researchers recognized the limitations
of a purely descriptive approach to acidbase physiology
Figure 1
The continuum of approaches to understanding acidbase physiology.
All three approaches share certain affecter elements and all use
markers and derived variables to describe acidbase imbalance.
A
TOT
, total weak acids; PCO
2
, partial carbon dioxide tension; SBE,
standard base excess; SID, strong ion difference; SIG, strong ion gap.
Henderson-
Hasselbalch
Base Excess
Physical
Chemical
pCO
2
Fixed acids
H
+
pCO
2
Buffer Base
pCO
2
SID
A
TOT
HCO
3
-
Anion Gap
SBE SIG
Markers
& Derived
Variables
Base Excess
Physical
Chemical
pCO
2
Buffer Base
pCO
2
SID
A
TOT
SBE SIG
Base Excess
Physical
Chemical

pCO
2
Buffer Base
pCO
2
SID
A
TOT
SBE SIG
Descriptive Semi-quantitative Quantitative
Affecters
502
[22]. One obvious limitation is that changes in plasma
bicarbonate concentration, although useful in determining the
direction and therefore the type of acidbase abnormality, are
not capable of quantifying the amount of acid or base that
has been added to the plasma unless PCO
2
is held constant.
This observation prompted the development of tools to
standardize bicarbonate or to quantify the metabolic
component of an acidbase abnormality. In 1948, Singer and
Hastings [22] proposed the term buffer base to define the
sum of HCO
3
and the nonvolatile weak acid buffers. A

change in buffer base corresponds to a change in the
metabolic component. The methods for calculating the
change in buffer base were later refined by investigators
[23,24] and refined further by others [25,26] to yield the base
excess (BE) methodology. BE is the quantity of metabolic
acidosis or alkalosis, defined as the amount of acid or base
that must be added to a sample of whole blood in vitro in
order to restore the pH of the sample to 7.40 while the PCO
2
is held at 40 mmHg [24]. Perhaps the most commonly used
formula for calculating BE is the Van Slyke equation [27,28]:
BE = (HCO
3
24.4 + [2.3 Hb + 7.7] [pH 7.4])

(1 0.023 Hb) (3)
where HCO
3
and hemoglobin (Hb) are expressed in mmol/l.

However, there is great variability in the equations used for
BE. For example, a commonly used commercially available
arterial blood gas machine calculates BE using a 14 variable
equation. In addition, although BE is quite accurate in vitro,
inaccuracy has always been a problem when applied in vivo in
that BE changes slightly with changes in PCO
2
[29,30]. This
effect is understood to be due to equilibration across the
entire extracellular fluid space (whole blood plus interstitial
fluid). Thus, the BE equation was modified to standardize the
effect of hemoglobin in order to improve the accuracy of BE in
vivo. The term standard base excess (SBE) has been given to
this variable, which better quantifies the change in metabolic
acidbase status in vivo. Again multiple equations exist:
SBE = 0.9287 (HCO
3
24.4 + 14.83 [pH 7.4]) (4)

However, Eqn 4 still yields results that are slightly unstable as
PCO
2
changes (Fig. 2). Furthermore, the equation assumes
normal A
TOT
. When albumin or phosphate is decreased a
common scenario in the critically ill Eqn 4 will result in even
more instability (Fig. 2). Recently, Wooten [4,5] developed a
multicompartment model using quantitative techniques and
suggested a correction for SBE that results in a formula for
SBE that agrees much more closely with experimental data in
humans.
Corrected SBE = (HCO
3
24.4) +
([8.3 albumin 0.15] + [0.29 phosphate 0.32])
(pH 7.4) (5)
Albumin is expressed in g/dl and phosphate in mg/dl.
Thus, the techniques previously developed to calculate
parameters that describe physiological acidbase balance in
single compartments have now been extended to
multicompartment systems. Furthermore, the equations for
multicompartment systems have been shown to possess the
same mathematical inter-relationships as those for single
compartments. Wooten also demonstrated that the
multicompartment form of the Van Slyke equation (Eqn 5) is
related in general form to the traditional form of the Van Slyke
equation (Eqn 3), and that with the multicompartment model
modern quantitative acidbase chemistry is brought into the
same context as the BE method [4].
In this way, SBE can be seen as the quantity of strong acid or
base required to restore the SID to baseline, at which pH is
7.40 and PCO
2
is 40 mmHg. Experimental data have already
borne out this relationship in that the change in SBE is
essentially equal to the change in SID across a vascular bed
(when there is no change in A
TOT
) [8]. If A
TOT
changes then
SBE still quantifies the amount of strong acid or base
required to change the SID to a new equilibrium point at
which pH is 7.40 and PCO
2
is 40 mmHg. This relationship
between SBE and SID is not surprising. Stewarts term SID
refers to the absolute difference between completely (or near
completely) dissociated cations and anions. According to the
principle of electrical neutrality, this difference is balanced by
the weak acids and CO
2
such that SID can be defined either
in terms of strong ions or in terms of the weak acids and CO
2
offsetting it. Of note, the SID defined in terms of weak acids
and CO
2
, which has been subsequently termed the effective
SID [31], is identical to the buffer base term coined by Singer
and Hastings [22] over half a century ago. Thus, changes in
SBE also represent changes in SID [8].
Updating the anion gap
Metabolic acidbase disturbances can be brought about by
changes in strong ions or weak ions. These ions can be
Figure 2
Carbon dioxide titration curves. Computer simulation of in vivo CO
2
titration curves for human plasma using the traditional Van Slyke
equation and various levels of A
TOT
(total weak acids) from normal
(17.2) to 25% of normal. Also shown is the titration curve using the
A
TOT
corrected standard base excess (SBEc).
5
4
3
2
1
0
1
2
3
4
7.7 7.6 7.5 7.4 7.3 7.2 7.1 7.0
pH
B
a
s
e

E
x
c
e
s
s
17.2
8.6
4.6
SBEc
503
routinely measured (e.g. Cl
) or not (e.g. ketones). The ones

not routinely measured are referred to as unmeasured ions.
Many years ago it was impractical to measure certain ions
such as lactate, and it remains impractical to measure others
such as sulfate. Thus, the literature contains a confusing array
of information regarding the magnitude of unmeasured ions
(usually anions) and techniques to estimate them.
Among these techniques, the anion gap (AG) is without
question the most durable. For more than 30 years the AG
has been used by clinicians and it has evolved into a major
tool with which to evaluate acidbase disorders [32]. The AG
is calculated, or rather estimated, from the differences
between the routinely measured concentrations of serum
cations (Na
+
and K
+
) and anions (Cl
and HCO
3
). Normally,
this difference or gap is made up by two components. The
major component is A
(i.e. the charge contributed by

albumin and to a lesser extent by phosphate). The minor
component is made up by strong ions such as sulfate and
lactate, whose net contributions are normally less than
2 mEq/l. However, there are also unmeasured (by the AG)
cations such as Ca
2+
and Mg
2+
, and these tend to offset the
effects of sulfate and lactate except when either is abnormally
increased. Plasma proteins other than albumin can be either
positively or negatively charged, but on aggregate they tend
to be neutral [31] except in rare cases of abnormal
paraproteins, such as in multiple myeloma. In practice the AG
is calculated as follows:
AG = (Na
+
+ K
+
) (Cl
+ HCO
3
) (6)
Because of its low and narrow extracellular concentration, K
+
is often omitted from the calculation. Respective normal
values with relatively wide ranges reported by most
laboratories are 12 4 mEq/l (if K
+
is considered) and
8 4 mEq/l (if K
+
is not considered). The normal AG has
decreased in recent years following the introduction of more
accurate methods for measuring Cl
concentration [33,34].
However, the various measurement techniques available
mandate that each institution reports its own expected
normal AG.
Some authors have raised doubts about the diagnostic value
of the AG in certain situations [35,36]. Salem and Mujais [35]
found routine reliance on the AG to be fraught with
numerous pitfalls. The primary problem with the AG is its
reliance on the use of a normal range produced by albumin
and to a lesser extent by phosphate, as discussed above.
These constituents may be grossly abnormal in patients with
critical illness, leading to a change in the normal range for
these patients. Moreover, because these anions are not
strong anions their charge will be altered by changes in pH.
This has prompted some authors to adjust the normal range
for the AG by the patients albumin and phosphate
concentration. Each 1 g/dl albumin has a charge of 2.8 mEq/l
at pH 7.4 (2.3 mEq/l at 7.0 and 3.0 mEq/l at 7.6), and each
1 mg/dl phosphate has a charge of 0.59 mEq/l at pH 7.4
(0.55 mEq/l at 7.0 and 0.61 mEq/l at 7.6). Thus, in much the
same way that the corrected SBE equation (Eqn 5) updates
BE to allow for changes in A
TOT
, the AG may be corrected to
yield a corrected AG (AGc) [7].
AGc = ([Na
+
+ K
+
] [Cl
+ HCO
3
])
(2[albumin (g/dl)] + 0.5[phosphate (mg/dl)])
or
AGc = [(Na
+
+ K
+
) (Cl
+ HCO
3
)]
(0.2[albumin (g/l)] + 1.5[phosphate (mmol/l)]) (7)
The choice of formula is determined by which units are
desired. Here the AGc should approximate zero. This is
because the terms for albumin and phosphate approximate
A
(the dissociated portion of A

TOT
). When AGc was used to
examine the presence of unmeasured anions in the blood of
critically ill patients, the accuracy improved from 33% with
the routine AG (normal range = 12 mEq/l) to 96% [7]. This
technique should only be used when the pH is less than
7.35, and even then it is only accurate within 5 mEq/l. Note
that some authors have chosen to correct the AG by
increasing the calculated value rather than adjusting its
expected range. Here the same (or slightly simplified
equations) are used to increase the AG toward the traditional
range rather than to decrease it toward zero. Either approach
would be acceptable, but if the objective is to quantify
unmeasured anions then the former may seem unnecessarily
cumbersome because it requires the additional step of
subtracting a normal value.
However, the purpose of the AG is to detect the presence of
unmeasured ions (e.g. ketones, salicylate), and AGc will not
consider abnormalities in other measured ions such as Mg
2+
or Ca
2+
, and the correction for albumin and phosphate is
merely an approximation. To be more exact, one can calculate
the strong ion gap (SIG) [37,38].
SIG = ([Na
+
+ K
+
+ Ca
2+
+ Mg
2+
] [Cl
+ lactate
])
(2.46 10
8
PCO
2
/10
pH
+ [albumin (g/dl)]
[0.123 pH 0.631] + [PO
4
(mmol/l)
(pH 0.469)]) (8)
Importantly, all the strong ions are expressed in mEq/l and
only the ionized portions of Mg
2+
and Ca
2+
are considered
(to convert total to ionized Mg
2+
, multiply by 0.7). Note also
that we do not consider lactate as unmeasured. Because the
concentration of unmeasured anions is expected to be quite
low (<2 mEq/l), the SIG is expected to be quite low.
However, some investigators have found elevations in SIG,
particularly in critically ill patients, even when no acidbase
disorder is apparent [39-42]. By contrast, results from
studies in normal animals [38,43] and values derived from
published data in exercising humans [37] put the normal
SIG near zero. There is even a suggestion that critically ill
patients in different countries might exhibit differences in SIG.
504
In the USA [40,44], Holland [39] and Thailand [45] the SIG
is about 5 mEq/l, whereas studies from England [41] and
Australia [42] report values in excess of 8 mEq/l.
The difference may lie with the use of gelatins in these
countries [46], which are an exogenous source of
unmeasured ions [47]. In this scenario the SIG is likely to be
a mixture of endogenous and exogenous anions. Interestingly,
previous studies that failed to find a correlation between SIG
and mortality were performed in countries that use gelatin
based resuscitation fluids [41,42], whereas studies of
patients not receiving gelatins [40,45,48] or any resuscitation
at all [44] found a positive correlation between SIG and
hospital mortality. Indeed, Kaplan and Kellum [44] recently
reported that preresuscitation SIG predicts mortality in
injured patients better than blood lactate, pH, or injury
severity scores. Similar results were also obtained by
Durward and coworkers [48] in pediatric cardiac surgery
patients. Although that study was done in England, gelatins
were not used. Thus, the predictive value of SIG may exceed
that of the AG, but it may vary from population to population
and even between institutions. As such, estimating the SIG
from the AG, after correcting for albumin and PO
4
, and after
subtracting lactate (i.e. AGc), may be a reasonable substitute
for the long hand calculation [7,39,46].
Together with the updates for SBE discussed above,
conversion between the descriptive approaches to
acidbase balance using HCO
3
or SBE and AG and the

quantitative approach using SID and SIG should be fairly
straightforward; indeed, they are (Table 1).
Quantitative acidbase at the bedside
If acidbase analysis can be reunified and BE and AG
updated, then it should be fairly easy to take the quantitative
approach to the bedside even without a calculator. In fact,
this is the approach that I have been using for several years
but it is now possible to be much more precise, given the
advances of the past few years. To see how this works, let us
consider a complex but all too common case (Table 2). This
patient presented (middle column) with severe metabolic
acidosis, as indicated by the SBE of 20 mEq/l or by the
combination of a low HCO
3
and PCO
2
. However, is this a
pure metabolic disorder or is there a respiratory component
as well? Table 3 shows the typical patterns found in patients
with simple acidbase disorders. A metabolic acidosis should
Table 1
Translator for acidbase variables across traditional and modern approaches
Physical
Traditional chemical
variable variable Comment
pH pH
PCO
2
PCO
2
HCO
3
Total CO
2
Total CO
2
includes dissolved CO
2
, H
2
CO
3
and CO
3
2
in addition to HCO
3
. However, for practical

purposes, at physiologic pH the two variables are very similar
Buffer base SIDe In the absence of unmeasured anions SIDe = SIDa = SID. However, because this rarely happens,
SIDe = SID = SIDa SIG (see text for discussion)
SBE SID
present
For blood plasma in vivo, SBE rather than ABE quantifies the amount of strong acid (or strong base if SBE is
SID
equilibrium
negative) that would be needed to return the SID to its equilibrium point (the point at which pH = 7.4 and
PCO
2
= 40). Note that change in SBE can brought about by a change in A
or SID, but SBE only quantifies

the change in SID required to reach equilibrium. In the case of a change in A
, the new equilibrium for SID

will be different (see text). The version of SBE that corrects for abnormalities in A

(SBEc) is given in Eqn 5
(see text)
Anion gap A
+ X
Virtually all of A
is composed of albumin and phosphate. A
can be approximated by
2(albumin [in g/dl]) + 0.5(phosphate [mg/dl]). The value of X
is the actually the difference between all

unmeasured anions and all unmeasured cations Because unmeasured anions are typically greater than
unmeasured cations, the sign of X
is positive. If a cation gap exists then the convention is to refer to this

as a negative anion gap
Anion gap A
SIG Anion gap A
approximates SIG, except that anion gap does not consider Mg

2+
, Ca
2+
, or lactate. Given
that A
+ X
= anion gap, it is tempting to equate SIG and X
. However, SIG will change if unmeasured weak

acids (A
X
) are present as well, so actually SIG = X
+ A
X
N/A A
TOT
A
TOT
= A
+ AH
Note that the translation from traditional to physical chemical variables is not a one to one exchange. Rather, the variable in the traditional column
corresponds to a similar variable in the physical chemical column (see comments for further explanation). Adapted with permission from Kellum
[10]. A
, nonvolatile weak acid buffers; ABE, actual base excess; AH, nondissociated weak acid; A
TOT
, total weak acids; PCO
2
, partial carbon
dioxide tension; SBE, standard base excess; SID, strong ion difference; SIDa, apparent strong ion difference; SIDe, effective strong ion difference;
SIG, strong ion gap; X
, unmeasured anions unmeasured cations.

505
be accompanied by a PCO
2
that conforms to both formula
([1.5 HCO
3
] + 8) and (40 + SBE), and indeed the PCO

2
of 20 mmHg fits this expectation. So, we can be assured that
this is a pure metabolic acidosis, but what is the cause?
The first step in determining the likely etiology should be to
determine the type of causative anion. Specifically, is the
metabolic acidosis due to measured or unmeasured anions?
The AG is 20 mEq/l so this is a positive AG acidosis, and
lactate is elevated so this is a lactic acidosis. However, are
unmeasured anions also present? Is there a hyperchloremic
acidosis as well? Could there be metabolic alkalosis?
An advantage of quantitative acidbase physiology is its
ability to determine the size of each effect. Using data
obtained 1 month before the current presentation, one can
see that there was already a metabolic acidosis even then,
and that the SID whatever value it was was approximately
8 mEq/l lower than at equilibrium (the point at which pH =
7.4 and PCO
2
= 40). At that time the 8 mEq/l was accounted
for by approximately 4 mEq/l of unmeasured anion (both AGc
and SIG are approximately 4), and the remaining 4 mEq/l
was, by definition, hyperchloremic. Note that the plasma Cl
concentration need not be increased; indeed, in this case the

107 mmol/l is still within the normal range. However, for the
Table 2
Typical case of metabolic acidosis
Parameter 1 month ago At presentation After resuscitation
Na
+
(mmol/l) 130 130 135
K
+
(mmol/l) 3.5 3.0 2.8
Cl
(mmol/l) 107 105 115

HCO
3
(mmol/l) 16 8 6
Creatinine (mg/dl [mol/l]) 2.8 (244) 2.9 (250)
Albumin (g/dl [g/l]) 2.0 (20) 2.3 (23) 1.8 (18)
PO
4
(mg/dl [mmol/l]) 4.5 (1.5) 4.8 (1.6) 4.2 (1.4)
Lactate (mmol/l) 1? 5 3
ABG 7.36/30/70 7.18/20/80 7.06/20/80
SBE (mEq/l) 9 20 23
SBEc (mEq/l) 8 18 20
AG (mEq/l) 10.5 20 17
AGc (mEq/l) 4.2 8 9.3
SIG (mEq/l) 3.8 9.2 10.3
A 55-year-old female with a history of hypertension and chronic renal insufficiency presents with fever, chills and arterial hypotension (blood
pressure 80/40 mmHg). She is resuscitated with approximately 140 ml/kg of 0.9% saline solution. The lactate value from 1 month ago is unknown
and assumed to be normal. Laboratory values are shown in American units (SI units in parentheses). ABG, arterial blood gas (pH/PCO
2
/PO
2
); AG,
anion gap; AGc, corrected anion gap; SBE, standard base excess; SBEc, corrected standard base excess; SIG, strong ion gap.
Table 3
Acidbase patterns observed in humans
Disorder HCO
3
(mEq/l) PCO
2
(mmHg) SBE (mEq/l)
Metabolic acidosis <22 = (1.5 HCO
3
) + 8 = 40 + SBE < 5
Metabolic alkalosis >26 = (0.7 HCO
3
) + 21 = 40 + (0.6 SBE) > +5

Acute respiratory acidosis = ([PCO
2
40]/10) + 24 >45 = 0
Chronic respiratory acidosis = ([PCO
2
40]/3) + 24 >45 = 0.4 (PCO
2
40)
Acute respiratory alkalosis = 24 ([40 PCO
2
]/5) <35 = 0
Chronic respiratory alkalosis = 24 ([40 PCO
2
]/2) <35 = 0.4 (PCO
2
40)
Adapted with permission from Kellum [7]. PCO
2
, partial carbon dioxide tension; SBE, standard base excess.
506
concentration of Na
+
at that time (130 mmol/l), the Cl
was
certainly increased. The diagnosis of hyperchloremic acidosis
is made by exclusion (i.e. metabolic acidosis not due to
lactate or unmeasured anions).
This combination of hyperchloremic and SIG acidosis is
common in renal failure [49] and, given that this patient has
significant chronic renal insufficiency, it is likely that this is the
cause. At presentation, however, she now has a SBE that is
roughly 10 mEq/l lower than it was 1 month ago. The
decrease appears to have resulted from lactate (increased by
4 mEq/l) and other anions (SIG increased by 5 mEq/l). It is
tempting to attribute the increase in lactate to shock, but
many other etiologies have been identified for
hyperlactatemia that could be responsible for the increase in
this patient [50]. The increase in SIG could be due to a
variety of factors, including poisons (e.g. salicylate, methanol,
etc.), ketones, and other organic acids such as sulfate [7,11].
Under the appropriate clinical conditions, these diagnoses
should be perused. However, sepsis [38] and shock [44]
also appear to increase SIG through unknown mechanisms,
and this may well be the cause in this case. Furthermore, the
SIG before resuscitation appears to correlate (inversely) with
outcome [44,48].
There does not appear to be any evidence of additional
hyperchloremic acidosis because the change in SBE is
almost completely explained by lactate and SIG. Neither is
there evidence of metabolic alkalosis, which would be
manifest by a SBE that was higher (less negative) than
predicted from the SIG and lactate. These complex
acidbase disorders can only be unmasked with the use of
quantitative techniques or, at least, semiquantitative
techniques using SBE, as illustrated here.
Finally, this patient was resuscitated with a large volume of
saline solution (SID = 0). The net effect of this solution on
blood pH is determined by the opposing effects of decreasing
SID (acidifying) and decreasing A
TOT
(alkalinizing). Because
the strong ions have a somewhat greater impact on pH than
do weak acids (which are weak after all), the net effect is an
acidosis [43,51]. Thus, in the final column of Table 2 we have
an SBEc of 20 mEq/l. This increased acidosis is due to an
increase in Cl
relative to Na
+
(approximately 5 mEq/l change)
and an increase in SIG (1 mEq/l). These effects are partially
offset by a decrease in lactate (2 mEq/l) and a decrease in
A
TOT
(approximately equal to a 2 mEq/l decrease). Thus, the
2 mEq/l worsening in SBEc is explained by each of these
components (5 + 1 2 2 = 2).
Conclusion
Recent advances in whole body acidbase physiology as
well as epidemiology have resulted in a much clearer picture
of metabolic acidbase disturbances in the critically ill and
injured. It is now possible to reunify traditional descriptive
approaches to acidbase balance with modern quantitative
techniques. This unified approach is both simple and
transparent and can be easily used at the bedside. It should
also aid in accessing and interpreting the bulk of the clinical
literature. As has already been the trend, newer studies of
acidbase physiology will no doubt take advantage of
quantitative techniques while continuing to report more
traditional variables.
Competing interests
JK has filed a patient disclosure for a software product
related to this field (in general).
References
1. Corey HE: Stewart and beyond: New models of acid-base
balance. Kidney Int 2003, 64:777-787
2. Corey HE: Fundamental principles of acidbase physiology.
Crit Care 2005, 9:184-192.
3. Wooten EW: Analytic claculation of physiological acid-base
5. Wooten EW: Quantitative acid-base physiology using the
Stewart model. Crit Care 2004, 8:448-452.
6. Shayakul C, Alper SL: Defects in processing and trafficking of
the AE1 Cl-/HCO3- exchanger associated with inherited
distal renal tubular acidosis. Clin Exp Nephrol 2004, 8:1-11
Crit Care 2000, 4:6-14.
8. Kellum JA, Bellomo R, Kramer DJ, Pinsky MR: Splanchnic buffer-
ing of metabolic acid during early endotoxemia. J Crit Care
1997, 12:7-12.
2
and standard base excess compensation for acid-base imbal-
ance. Crit Care Med 1998, 26:1173-1179.
10. Kellum JA: Making strong ion difference the Euro for bedside
acid-base analysis. In Yearbook of Intensive Care and Emer-
gency Medicine. Edited by Vincent JL. Berlin: Springer-Verlag;
2005:675-685.
11. Kellum JA: Determinants of plasma acid-base balance. Crit
Care Clin 2005, 21:329-346.
12. Stewart P: Modern quantitative acid-base chemistry. Can J
13. Stewart PA: How to Understand Acid-base: A Quantitative Acid-
base Primer for Biology and Medicine, 1st ed. New York: Else-
vier; 1981.
14. Kellum JA. Acid base pHorum. [http://www.ccm.upmc.edu/edu-
cation/resources/phorum.html]
15. Lloyd P: Strong ion calculator. [http://homepage.mac.com/peter-
lloyd1/FileSharing8.html]
16. Kellum JA: Acid-base physiology in the post-Copernican era.
17. Magder S: Pathophysiology of metabolic acid-base distur-
bances in patients with critical illness. In Critical Care Nephrol-
ogy. Edited by Ronco C, Bellomo R. Dordrecht, The Netherlands:
Kluwer Academic Publishers; 1997:279-296.
equilibria: Application to horse plasma. J Appl Physiol 1997,
83:297-311.
19. Fernandez PC, Cohen RM, Feldman GM: The concept of bicar-
bonate distribution space: the crucial role of body buffers.
Kidney Int 1989, 36:747-752.
20. Garella S, Dana CL, Chazan JA: Severity of metabolic acidosis
as a determinant of bicarbonate requirements. N Engl J Med
1973, 289:121-126.
21. Androgue HJ, Brensilver J, Cohen JJ, Madias NE: Influence of
steady-state alterations in acid-base equilibrium on the fate
of administered bicarbonate in the dog. J Clin Invest 1983, 71:
867-883.
22. Singer RB, Hastings AB: An improved clinical method for the
estimation of disturbances of the acid-base balance of human
blood. Medicine (Baltimore) 1948, 27:223-242.
507
23. Astrup P, Jorgensen K, Siggaard-Andersen O: Acid-base metab-
olism: New approach. Lancet 1960, 1:1035-1039.
24. Siggaard-Andersen O: The pH-log PCO2 blood acid-base
nomogram revised. Scand J Clin Lab Invest 1962, 14:598-604.
25. Grogono AW, Byles PH, Hawke W: An in vivo representation of
acid-base balance. Lancet 1976, 1:499-500.
26. Severinghaus JW: Acid-base balance nomogram a Boston-
Copenhagen dtente. Anesthesiology 1976, 45:539-541.
27. Siggaard-Andersen O: The Acid-base Status of the Blood, 4th
ed. Baltimore, MD: William and Wilkins; 1974,
Lab Invest 1977, 146:15-20.
29. Brackett NC, Cohen JJ, Schwartz WB: Carbon dioxide titration
curve of normal man. N Engl J Med 1965, 272:6-12.
30. Prys-Roberts C, Kelman GR, Nunn JF: Determinants of the in
vivo carbon dioxide titration curve in anesthetized man. Br J
Anesth 1966, 38:500-550.
32. Narins RG, Emmett M: Simple and mixed acid-base disorders:
A practical approach. Medicine (Baltimore) 1980, 59:161-187.
33. Sadjadi SA: A new range for the anion gap. Ann Intern Med
1995, 123:807-808.
34. Winter SD, Pearson R, Gabow PG, Schultz A, Lepoff RB: The fall
of the serum anion gap. Arch Intern Med 1990, 150:3113-3115.
1992, 152:1625-1629.
36. Gilfix BM, Bique M, Magder S: A physical chemical approach to
the analysis of acid-base balance in the clinical setting. J Crit
Care 1993, 8:187-197.
during acute endotoxemia. J Appl Physiol 1995, 78:2212-2217.
39. Moviat M, van Haren F, van der Hoeven H: Conventional or
physicochemical approach in intensive care unit patients with
metabolic acidosis. Crit Care 2003, 7:R41-R45.
anions identified by the Fencl-Stewart method predict mortal-
in the pediatric intensive care unit. Crit Care Med 1999, 27:
1577-1581.
41. Cusack RJ, Rhodes A, Lochhead P, Jordan B, Perry S, Ball JAS,
prognostic value in critically ill patients in a mixed medical/
surgical adult ICU. Intensive Care Med 2002, 28:864-869.
42. Rocktaschel J, Morimatsu H, Uchino S, Bellomo R: Unmeasured
anions in critically ill patients: can they predict mortality? Crit
Care Med 2003, 31:2131-2136.
Shock 1998, 9:364-368.
44. Kaplan L, Kellum JA: Initial pH, base deficit, lactate, anion gap,
strong ion difference, and strong ion gap predict outcome
from major vascular injury. Crit Care Med 2004, 32:1120-1124.
45. Dondorp AM, Chau TT, Phu NH, Mai NT, Loc PP, Chuong LV,
Sinh DX, Taylor A, Hien TT, White NJ, Day NP: Unidentified
acids of strong prognostic significance in severe malaria. Crit
Care Med 2004, 32:1683-1688.
46. Kellum JA: Closing the gap on unmeasured anions. Crit Care
2003, 7:219-220.
Care Med 1999, 25:680-685.
48. Durward A, Tibby SM, Skellett S, Austin C, Anderson D, Murdoch
IA: The strong ion gap predicts mortality in children following
cardiopulmonary bypass surgery. Pediatr Crit Care Med 2005,
6:281-285.
49. Rocktaschel J, Morimatsu H, Uchino S, Goldsmith D, Poustie S,
Story D, Gutteridge G, Bellomo R: Acid-base status of critically
ill patients with acute renal failure: analysis based on Stewart-
Figge methodology. Crit Care 2003, 7:R60-R66.
50. Kellum JA, Kramer DJ, Lee K, Mankad S, Bellomo R, Pinsky MR:
Release of lactate by the lung in acute lung injury. Chest
1997, 111:1301-1305.
ence determines metabolic acid-base change during in vitro
hemodilution. Crit Care Med 2002, 30:157-160.
331
bHS = 6% hetastarch in a balanced electrolyte solution; IL = interleukin; iNOS = inducible nitric oxide synthase; LPS = lipopolysaccharide; LR =
lactated Ringers; MAP = mean arterial pressure; NF-B = nuclear factor-B; NO = nitric oxide; NS = normal (0.9%) saline; pH
i
= intracellular pH;
pH
o
= extracellular pH; SBE = standard base excess; TNF = tumor necrosis factor.
Introduction
Critical illness is exemplified by a state of profound disruption
in normal homeostatic mechanisms. Patients who remain
critically ill may progress to a poorly understood condition
known as multiple organ failure, which is characterized by
widespread alterations in both individual organ function and
integrative function across organs. Although our under-
standing of this condition is extremely limited, numerous
observations suggest that alterations in the immune response
are not only caused by but may also be the cause of ongoing
organ injury, and these alterations may adversely affect
patients ability to recover. Both increased inflammation and
immune suppression have been implicated in the
pathogenesis of multiple organ failure. Little is known about
the influences that therapies have on the immune response.
Emerging evidence suggests that ventilator-associated lung
injury results in increased systemic inflammation [1] and that
systemic inflammation resulting from local tissue injury
appears to have effects on remote organs [2]. Drugs that
appear to modify the course of organ injury such as activated
protein C and corticosteroids appear to have a broad range
of effects on the immune system [3,4]. Abnormalities in
systemic acidbase balance may also induce significant
alterations in the immune response. The clinical significance
of these alterations is not yet known, but their magnitude
suggests that they may play an important role in the
development or maintenance of immune dysfunction. If this is
the case, then they represent attractive targets (or even tools)
for therapy. Extracellular pH (pH
o
) for circulating leukocytes
(i.e. blood pH) is easily altered and thus, for good or bad,
changes in pH may rapidly alter the immune response in
these cells.
Review
Science review: Extracellular acidosis and the immune response:
clinical and physiologic implications
John A Kellum
1
, Mingchen Song
2
and Jinyou Li
3
1
Associate Professor, Critical Care Medicine and Medicine, Co-Director, The MANTRA (Mechanisms And Novel Therapies for Resuscitation and Acute
illness) Laboratory, Department of Critical Care Medicine, University of Pittsburgh School of Medicine, Pittsburgh, Pennsylvania, USA
2
Research Fellow, Department of Critical Care Medicine, The MANTRA Laboratory, Department of Critical Care Medicine, University of Pittsburgh
School of Medicine, Pittsburgh, Pennsylvania, USA
3
Visiting Researcher, Department of Critical Care Medicine, The MANTRA Laboratory, Department of Critical Care Medicine, University of Pittsburgh
School of Medicine, Pittsburgh, Pennsylvania, USA
Corresponding author: John A Kellum, kellumja@ccm.upmc.edu
Published online: 16 June 2004 Critical Care 2004, 8:331-336 (DOI 10.1186/cc2900)
Abstract
Metabolic acidosis is among the most common abnormalities seen in patients suffering from critical
illness. Its etiologies are multiple and treatment of the underlying condition is the mainstay of therapy.
However, growing evidence suggests that acidosis itself has profound effects on the host, particularly
in the area of immune function. Given the central importance of immune function to the outcome of
critical illness, there is renewed interest in elucidating the effects of this all too common condition on
the immune response. In this review we concentrate on the effects of extracellular acids on production
and release of inflammatory mediators, and we demonstrate that different acids produce different
effects despite similar extracellular pH. Finally, we discuss potential clinical implications.
Keywords acidosis, cytokines, immune response, pH, sepsis
332
Critical Care October 2004 Vol 8 No 5 Kellum et al.
Effects of extracellular acidosis on
inflammatory mediator release
There are now several studies documenting the effects of
decreased pH
o
on the synthesis and release of inflammatory
mediators, especially tumor necrosis factor (TNF) and nitric
oxide (NO). Most of these studies were conducted in resident
macrophages or macrophage-like cell lines and yielded
conflicting results (Table 1). However, studies using HCl have
consistently shown proinflammatory effects at the level of
nuclear factor-B (NF-B) DNA binding or TNF synthesis
provided pH
o
was not less than 6.0 [57], although TNF
secretion was reduced even at pH
o
as high as 7.0 [5,7,8].
Studies of nonstimulated resident peritoneal macrophages
[6] and lipopolysaccharide (LPS)-stimulated RAW 264.7
cells [9] have shown increased NO formation at moderately
reduced pH
o
(7.07.2). However, more severely acidic pH
o
reduces NO formation [6,9], and there is an apparent
dissociation between the pH
o
effects on inducible nitric oxide
synthase (iNOS) mRNA, protein, and final NO release [9].
Thus, HCl appears to affect inflammatory mediators
differently at different stages in their synthesis and release.
Little is known about the effects of HCl on other cytokines or
on the kinetics of pH
o
mediated effects.
Lactic acid has been studied in an even more limited way
than HCl. Lactic acid (pH
o
6.75) was shown in one study
[10] to result in increased TNF release in LPS-stimulated
peritoneal macrophages. This finding is surprising in light of
the growing evidence of a protective effect of lactic acid in
neuronal injury [1113]. Several studies have sought to
explore the effect of dialysis solutions on the immune
response [14,15]. These acidic, lactate-based solutions have
been shown to decrease various aspects of the immune
response, including TNF synthesis and release [14,15].
Douvdevani and coworkers [15] also demonstrated a
decrease in LPS-induced NF-B DNA binding in human
blood-derived macrophages when incubated with dialysis
solution. Although these solutions are also hyperosmolar and
have excessive glucose concentrations variables that are
known to influence immune function [14,16] they provide
additional evidence of a potential anti-inflammatory role of
lactate and highlight potential differences between various
acids and their effects on the immune response.
We conducted a series of experiments in LPS-stimulated
RAW 264.7 murine macrophage-like cells in which we
decreased the pH
o
of the medium using different acids.
Remarkably, dramatically different patterns of inflammatory
mediator expression occurred with different acids, despite
normalization to the same pH
o
. In our first set of experiments
[17] we acidified the cell culture medium using HCl and
stimulated the cells with 10 ng/ml LPS (Escherichia coli
0111:B4) for 24 hours. Acidic medium itself barely affected
the release of inflammatory mediators, including NO, IL-6, and
IL-10. However, compared with pH
o
7.4, acidosis (pH
o
7.0)
was associated with significantly increased NO release in
response to LPS stimulation. Interestingly, under more
extreme acidic conditions (pH
o
6.5), NO release decreased in
response to LPS and was again similar to pH
o
7.4 (Table 2).
At pH
o
6.5, release of both IL-6 and IL-10 was significantly
less than at pH
o
7.0 or 7.4. However, IL-10 release was
reduced to a far greater extent than was IL-6, and thus the
ratio of IL-6 to IL-10 increased significantly from 5:1 at pH
o
7.4 to 55:1 at pH
o
6.5.
These findings suggest a proinflammatory effect of HCl,
which is consistent with the existing literature on the effects
of HCl on TNF synthesis [57]. Furthermore, the paradox in
which mild and severe acidosis induced by HCl results in
opposite effects on NO has now been explained. Pedoto and
colleagues [18] first suggested that the optimal intracellular
pH (pH
i
) for iNOS was near 7.0 and that the addition of acid
Table 1
Effects of acids on inflammatory mediators in macrophages
Acid pH
o
Cells LPS Effect Reference
HCl 6.5 Alveolar macrophages (+) TNF mRNA 5
HCl 5.5 Alveolar macrophages (+) TNF mRNA/TNF secretion 5
HCl 5.5 RAW (+) No TNF mRNA/TNF secretion 7
HCl 7.0 Alveolar macrophages (+) TNF secretion 8
HCl 7.0 Peritoneal macrophages () NO, TNF*, NF-B 6
HCl 7.2 RAW (+) NO 9
LA 6.7 Peritoneal macrophages (+) TNF mRNA/TNF secretion 10
DS 6.0 Peritoneal macrophages (+) TNF mRNA/TNF secretion 14
DS 6.5 Human blood-borne macrophages (+) TNF mRNA, NF-B 15
*Tumor necrosis factor (TNF) was not measured directly. DS, lactate-based dialysis solution; LA, lactic acid; LPS, lipopolysaccharide; NF-B,
nuclear factor-B; NO, nitric oxide; NR, not recorded; pH
o
, extracellular pH.
333
would lower the pH
i
toward the optimal value, thus increasing
iNOS activity and NO production. Further addition of acid
would cause pH
i
to fall below the optimal value, leading to
decreased NO production [18]. This hypothesis was recently
tested by Huang and coworkers [9], who demonstrated that
the optimal pH
o
for NO formation by iNOS was 7.2 in RAW
264.7 cells. However, they also noted that alkaline pH
o
favored
expression of iNOS protein but that post-transcriptional
mechanisms predominated, resulting in increased NO release
at slightly acidotic pH
o
.
To clarify the mechanism by which HCl influenced the release
of cytokines from LPS-stimulated cells, we measured NF-B
DNA binding using electrophoretic mobility shift assay after
exposure to different concentrations of HCl [17]. Again,
acidosis (pH
o
7.0) significantly increased LPS-induced NF-B
activation, as compared with pH
o
7.4, whereas more extreme
acidosis (pH
o
6.5) actually attenuated NF-B activation. Thus,
different degrees of hyperchloremic acidosis have differing
effects on inflammatory mediator release as well as on NF-B
activation. Overall, the effects of HCl appear to be
proinflammatory. These findings are in accordance with those
of a study conducted in resident peritoneal macrophages by
Bellocq and colleagues [6]. Those investigators found that
these cells produced more NO when incubated in medium at
pH
o
7.0 than at pH 7.4, and that this effect was associated
with upregulation of iNOS mRNA as well as with activation of
NF-B.
By contrast, our data using lactic acid demonstrates that this
acid is anti-inflammatory to RAW 264.7 cells, as indicated by
decreased cytokine expression and NF-B activation [17]. In
these experiments, increasing concentrations of lactic acid
(030 mmol/l) caused increasing acidification of the media,
and trypan blue exclusion and lactate dehydrogenase release
demonstrated that lactic acid did not reduce cell viability.
However, lactic acid inhibited LPS-induced NF-B DNA
binding (Table 2). Lactic acid also significantly decreased
LPS-induced expression of NO, IL-6, and IL-10, both RNA
and protein, in a dose-dependent manner.
The mechanisms by which these acids exert their effects on
innate immunity are presently unknown. The effects are not
limited to LPS-stimulated cells, however, because the results
have been (preliminarily) reproduced in interferon- stimulated
RAW 264.7 cells [19], suggesting that the effects are not
mediated through pH-induced changes in the LPS molecule
or LPS-binding protein, or at the receptor. The effects may be
partly mediated through NF-B because DNA binding of this
transcription factor is generally consistent with effects on NO
and IL-6 (Table 2). However, extracellular acids also have
effects on IL-10, which is outside the NF-B pathway. What
is apparent is that the effects of extracellular acids are not
limited to the effects on pH
o
because different acids produce
different effects despite similar pH
o
. Whether different effects
can be explained by differences in pH
i
are as yet unknown,
although the patterns of response (Table 2) suggest that this
is likely.
Effects of extracellular acidosis on other
aspects of immune cell function
While this review focuses on the effects of extracellular acids
on inflammatory mediator release, there is evidence that
acidosis influences other aspects of the immune response.
As detailed in the excellent review by Lardner [20],
extracellular acidosis has far reaching effects on the immune
response. For example, leukocyte chemotaxis is impaired at
extreme acidic pH
o
, generally beginning between pH 6.0 and
5.5 [2123] with an additive effect of hypoxia [22,24].
Activation of oxygen burst in neutrophils [25], production of
reactive oxygen species [2628], neutrophil phagocytosis
[25,29], and intracellular killing [30] all appear to be
influenced by pH
o
, as does neutrophil apoptosis [31,32].
Finally, there is evidence that complement activation by C-
reactive protein may be the result of a pH
o
-dependent
conformational change in the protein [33].
Table 2
Summary of effects of lactic acid versus HCl on lipopolysaccharide-stimulated RAW 264.7 cells
Lactic acid (pH 7.0) Lactic acid (pH 6.5) HCl (pH 7.0) HCl (pH 6.5)
NO
iNOS mRNA
IL-6
IL-6 mRNA
IL-10
IL-10 mRNA
IL-6: IL-10 ratio
NF-B
IL, interleukin; iNOS, inducible nitric oxide synthase; NO, nitric oxide. Adapted from Kellum and coworkers [19].
334
Thus, pH
o
, or the effects of the separate ions involved,
appears to influence multiple aspects of the inflammatory
response. In addition, extracellular acidification may exert its
effects by altering pH
i
. Indeed, several studies have identified
a relationship between pH
i
and pH
o
, regardless of which
milieu is altered experimentally [34,35]. For example, when
pH
o
was increased a subsequent increase in pH
i
, mediated
by the N
+
/H
+
exchanger (NHE-1), was observed, along with
augmented leukotriene release by neutrophils [34]. These
events were followed by extracellular acidification. Of note,
studies conducted in bicarbonate-buffered medium [32] have
shown effects on neutrophil function that are at odds with
other literature. Those investigators hypothesized that acid
titration of bicarbonate with generation of CO
2
leads to a
rapid decrease in pH
i
. Alternatively, the CO
2
effect may be
independent from the effect on pH
i
.
In vivo effects of hyperchloremic acidosis
Experiments using cells in culture exposed HCl or lactic acid
provide a highly reproducible but less clinically relevant model
for study. By contrast, saline resuscitation is an extremely
common cause of hyperchloremic acidosis. By using a
mathematical model based on a physicochemical acidbase
analysis, we accurately predicted the serum Cl
concentra-
tion and resulting arterial blood pH changes in healthy dogs
given large volumes of intravenous 0.9% saline [36]. By
applying this model to dogs given an intravenous bolus of
LPS (1 mg/kg) and subsequent large volume saline resuscita-
tion (100 ml/kg over 3 hours), we quantified the effects on
acidbase balance [36]. The total acid load was calculated
from the change in standard base excess (SBE) attributable
to each source. In LPS-treated animals mean arterial pH
decreased from 7.32 to 7.11 (P < 0.01); partial CO
2
tension
and lactate were unchanged. Saline accounted for 38% of
the total acid load. Although serum Na
+
did not change,
serum Cl
increased (128 to 137 mmol/l; P = 0.016). From

these experiments we concluded that saline resuscitation alone
accounts for more than a third of the acidosis seen in this
canine model of acute endotoxemia, whereas lactate accounts
for less than 10%. Furthermore, a large amount of the
unexplained acid load in this model appears to be attributable
to differential Na
+
and Cl
shifts, presumably from extravascular

to vascular or intracellular to extracellular spaces.
In a recent study [37], we found that normal (0.9%) saline
(NS) resuscitation resulted in a decreased survival time and
reduced the SBE by 510 mEq/l as compared with a
balanced colloid solution. In this experiment, we studied 60
rats for 12 hours after intravenous infusion of LPS (20 mg/kg).
We resuscitated to maintain a mean arterial pressure (MAP)
above 60 mmHg using NS, 6% hetastarch in a balanced
electrolyte solution (bHS), or lactated Ringers (LR). We
showed that mean survival time among animals treated with
NS or LR was 45% less than in bHS-treated animals
(P < 0.0001) and that overall survival (at 12 hours) was 0%
with NS or LR versus 20% with bHS (P = 0.05). After
resuscitation with NS, arterial SBE and plasma apparent
strong ion difference were both significantly lower and
plasma Cl
was significantly higher than with bHS.

Resuscitation with LR resulted in a SBE and plasma Cl
between those with NS and bHS. Importantly, we observed

an inverse relationship between the change in serum Cl
and
survival time in these animals (R
2
= 0.37; P < 0.001). From
these data we concluded that, as compared with bHS,
volume resuscitation with NS was associated with more
metabolic acidosis and shorter survival in this experimental
animal model of septic shock. Furthermore, we hypothesized
that hyperchloremia may play a role in reducing short-term
survival, but that other factors must also be involved because
LR-treated rats fared no better than did those treated with
NS, even if they had less hyperchloremia.
Metabolic acidosis might reduce survival from sepsis through
a variety of mechanisms. First, acidosis has been associated
with hemodynamic instability [38], although the association is
not always consistent [39] and the underlying mechanisms
are uncertain. Pedoto and colleagues [18] recently showed
that metabolic acidosis may increase iNOS expression in
animals and that this could exacerbate vasodilation and
shock. Second, acidosis, even in the absence of sepsis or
endotoxemia, is associated with gut barrier dysfunction
[40,41]. Finally, acidosis can lead to oxidative stress by
promoting delocalization of protein-bound iron stores in cells
leading to Fenton-type biochemistry and redox stress [42],
and by causing protonation of the peroxynitrite anion
(ONOO
) and thereby increasing the tendency of this moiety

to behave like the potent free radical hydroxyl (OH
) [43,44].
Pedoto and colleagues demonstrated that hyperchloremic
acidosis increases lung [18] and intestinal injury [45] in
healthy rats.
In order to control for other effects of large-volume
resuscitation (e.g. cell swelling), we next increased serum Cl
concentration by infusing a dilute HCl solution into rats with

sepsis induced by cecal ligation and puncture [46]. Eighteen
hours after cecal ligation and puncture, we randomly assigned
24 rats to three groups. In groups 2 and 3 we began an 8-
hour intravenous infusion of 0.1 N HCl to reduce the SBE by
510 and 1015 mEq/l, respectively. We measured MAP,
arterial blood gases, electrolytes, and plasma nitrate/nitrite
levels at 0, 3, 6 and 8 hours. MAP remained stable in group 1
but decreased in groups 2 and 3 (P < 0.001), such that at
8 hours MAP was much higher in group 1 than in either group
2 or group 3 (Fig. 1). This change in MAP correlated with the
increase in plasma Cl
(R
2
= 0.50; P < 0.0001) and less well
with the decrease in pH (R
2
= 0.24; P < 0.001). After 6 hours
of acidosis plasma nitrite levels were significantly higher in
group 2 animals than in group 1 or group 3 animals
(P < 0.05). We concluded that moderate acidosis, induced by
HCl infusion, worsened blood pressure and increased plasma
nitrate/nitrite levels in septic rats. Some other mechanism is
needed to account for the further reduction in MAP in group 3
335
animals, however, because NO release was not increased in
that group. Our results are in general agreement with reports
by Pedoto and coworkers [18,45] that demonstrated that
metabolic acidosis increased iNOS, leading to vasodilation
and shock in healthy rats. Our study extends these findings by
examining the effects of acidosis in nonshocked, septic
animals. These data are also consistent with our data from
RAW 264.7 cells (presented above), in which a decreased
pH
o
(7.0) resulted in increased NO release but more severe
acidosis (pH
o
= 6.5) did not [17].
Clinical implications
Understanding the effects of acidbase balance on the
inflammatory response is highly relevant to clinical medicine
for a variety of reasons. First, current deficiencies in our
understanding of the effects of acidosis on a wide range of
cellular processes have led to controversy in the way in which
patients are managed in a variety of clinical settings. Most
clinicians tend to ignore the effects of exogenous Cl
on pH
o
,
but many will treat even mild forms of acidemia. In addition, all
forms of metabolic acidosis appear to be associated with
prolonged hospital and intensive care unit length of stay [47].
Because metabolic acidosis is both commonly caused and
treated by clinicians, an understanding of the physiologic
consequences of altered pH
o
is imperative.
Second, our ability to alter acidbase balance as a tool with
which to manipulate cellular processes will be dependent on
an improved understanding of the relationship between pH
o
and the synthesis and release of inflammatory molecules.
Investigators continue to seek means to modulate the
inflammatory response as primary therapy for sepsis and
related conditions. These efforts have focused not only on
reducing proinflammatory mediators in an effort to reduce
tissue injury, but also on the converse augmenting the
inflammatory response to infection. This interest also extends
into other fields, including autoimmune disease and cancer
therapy. For example, decreased lymphocyte function has
been documented with decreased pH
o
in human lymphokine-
activated killer cells [48], human IL-2 stimulated lymphocytes
[49], as well as murine natural killer cells [50]. The
mechanisms responsible for these effects are unknown but
probably do not include energy substrate depletion [50].
Third, even when it is not practical or desirable to manipulate
pH
o
as a primary means of altering the inflammatory response,
an understanding of how pH
o
affects this response is necessary
to interpret data from studies of immunomodulation; to avoid
unintended immunomodulation in clinical and laboratory
settings; and to explore the capacity of pH
o
to improve the
effectiveness of existing treatments. Finally, an understanding of
how pH
o
is involved in the regulation of inflammation by
intracellular signaling pathways or other mechanism might
ultimately lead to other strategies for immunomodulation.
Conclusion
Little is currently known about the effects of acidbase
abnormalities on innate immunity. Acidosis produces
significant effects on immune effector cell function in vitro.
The regulation of NO release and synthesis has been found
to be significantly effected by pH
o
both in vitro and in vivo,
and may be partially responsible for acidosis-associated
hemodynamic instability. Production of inflammatory cyto-
kines, as well as DNA-binding of transcription factors in their
control pathways, appears to be sensitive to pH
o
as well.
However, emerging evidence suggests that different forms of
acidosis (respiratory versus metabolic) and even different
types of metabolic acidosis (lactic versus hyperchloremic)
produce different effects. Overall, lactic acid appears to be
anti-inflammatory whereas HCl is proinflammatory. The extent
to which these effects apply to the clinical situation has yet to
be determined, but given that acidosis is an extremely
common problem in the intensive care unit, and immune
function is of critical importance, efforts to elucidate these
relationships are quite justified.
Competing interests
JAK has received research grants and consulting fees from
Abbott Laboratories.
References
1. Chu EK, Whitehead T, Slutsky AS: Effects of cyclic opening and
closing at low- and high-volume ventilation on bronchoalveo-
lar lavage cytokines. Crit Care Med 2004, 32:168-174.
2. Burne-Taney MJ, Kofler J, Yokota N, Weisfeldt M, Traystman RJ,
Rabb H: Acute renal failure after whole body ischemia is char-
acterized by inflammation and T cell-mediated injury. Am J
Physiol Renal Physiol 2003, 285:F87-F94.
3. Joyce DE, Grinnell BW: Recombinant human activated protein
C attenuates the inflammatory response in endothelium and
monocytes by modulating nuclear factor-kappaB. Crit Care
Med 2002, Suppl:S288-S293.
Figure 1
Mean arterial pressure for septic animals (induced by cecal ligation
and puncture) after infusion of 0.1 N HCl acid to reduce the base
deficit (BD) by 510 mEq/l (white bars) or 1015 mEq/l (black bars).
A control group was given a similar volume of lactated Ringers (gray
bars). Shown are group means (n = 8) SEM. *P < 0.05. Adapted
from Kellum and coworkers [46].
0
20
40
60
80
100
120
140
0 3 6 8
Time (hours)
m
m
H
g
BD 0
BD 510
BD 1015
*
*
336
4. Annane D, Cavaillon JM: Corticosteroids in sepsis: from bench
to bedside? Shock 2003, 20:197-207.
5. Heming TA, Dave SK, Tuazon DM, Chopra AK, Peterson JW,
Bidani A: Effects of extracellular pH on tumour necrosis
factor-alpha production by resident alveolar macrophages.
Clin Sci 2001, 101:267-274.
6. Bellocq A, Suberville S, Philippe C, Bertrand F, Perez J, Fou-
queray B, Cherqui G, Baud L: Low environmental pH is respon-
sible for the induction of nitric-oxide synthase in
macrophages. Evidence for involvement of nuclear factor-
kappaB activation. J Biol Chem 1998, 273:5086-5092.
7. Heming TA, Tuazon DM, Dave SK, Chopra AK, Peterson JW,
Bidani A: Post-transcriptional effects of extracellular pH on
tumour necrosis factor-alpha production in RAW 246.7 and
J774 A.1 cells. Clin Sci 2001, 100:259-266.
8. Bidani A, Wang CZ, Saggi SJ, Heming TA: Evidence for pH sen-
sitivity of tumor necrosis factor-alpha release by alveolar
macrophages. Lung 1998, 176:111-121.
9. Huang CJ, Haque IC, Slovin PN, Nielsen RB, Fang X, Skimming
JW: Environmental pH regulates LPS-induced nitric oxide for-
mation in murine macrophages. Nitric Oxide 2002, 6:73-78.
10. Jensen JC, Buresh C, Norton JA: Lactic acidosis increases
tumor necrosis factor secretion and transcription in vitro. J
Surg Res 1990, 49:350-353.
11. Himmelseher S, Pfenninger E, Georgieff M: Basic fibroblast
growth factor reduces lactic acid-induced neuronal injury in
rat hippocampal neurons. Crit Care Med 1998, 26:2029-2036.
12. Schurr A, Payne RS, Miller JJ, Rigor BM: Brain lactate is an
obligatory aerobic energy substrate for functional recovery
after hypoxia: further in vitro validation. J Neurochem 1997, 69:
423-426.
13. Schurr A, Payne RS, Miller JJ, Rigor BM: Brain lactate, not
glucose, fuels the recovery of synaptic function from hypoxia
upon reoxygenation: an in vitro study. Brain Res 1997, 744:
105-111.
14. Jorres A, Gahl GM, Frei U: In vitro studies on the effect of dialy-
sis solutions on peritoneal leukocytes. Perit Dial Int 1995, 15:
S41-S45.
15. Douvdevani A, Abramson O, Tamir A, Konforty A, Isakov N,
Chaimovitz C: Commercial dialysate inhibits TNF alpha mRNA
expression and NF-kappa B DNA-binding activity in LPS-stim-
ulated macrophages. Kidney Int 1995, 47:1537-1545.
16. AlKharfy KM, Kellum JA, Matzke G: Unintended immunomodula-
tion: part I. Effect of common clinical conditions on cytokine
biosynthesis. Shock 2000, 13:333-345.
17. Kellum JA, Song M, Li J: Lactic, and hydrochloric acids induce
different patterns of inflammatory response in LPS-stimulated
RAW 264.7 cells. Am J Physiol Regul Integr Comp Physiol 2004,
286:R686-R692.
18. Pedoto A, Caruso JE, Nandi J: Acidosis stimulates nitric oxide
production and lung damage in rats. Am J Respir Crit Care
Med 1999, 159:397-402.
19. Lu Y, Song M, Zuo Y, Kellum JA: The inhibitory effect of lactic
acid on NO release by stimulated RAW cells is not specific to
LPS [abstract]. Crit Care Med 2003, Suppl:A45.
20. Lardner A: The effects of extracellular pH on immune function.
J Leukoc Biol 2001, 69:522-530.
21. Nahas GG, Tannieres ML, Lennon JF: Direct measurement of
leukocyte motility: effects of pH and temperature. Proc Soc
Exp Biol Med 1971, 138:350-352.
22. Rotstein OD, Fiegel VD, Simmons RL, Knighton DR: The delete-
rious effect of reduced pH and hypoxia on neutrophil migra-
tion in vitro. J Surg Res 1988, 45:298-303.
23. Rabinovich M, DeStefano MJ, Dziezanowski MA: Neutrophil
migration under agarose: stimulation by lowered medium pH
and osmolality. J Reticuloendothel Society 1980; 27:189-200.
24. Simchowitz L, Cragoe EJ Jr: Regulation of human neutrophil
chemotaxis by intracellular pH. J Biol Chem 1986, 261:6492-
6500.
25. Leblebicioglu B, Lim JS, Cario AC, Beck FM, Walters JD: pH
changes observed in the inflamed gingival crevice modulate
human polymorphonuclear leukocyte activation in vitro. J Peri-
odontol 1996, 67:472-477.
26. Araki A, Inoue T, Cragoe EJ Jr, Sendo F: Na+/H+ exchange
modulates rat neutrophil mediated tumor cytotoxicity. Cancer
Res 1991, 51:3212-3216.
27. Simchowitz L: Intracellular pH modulates the generation of
superoxide radicals by human neutrophils. J Clin Invest 1985,
76:1079-1089.
28. Gabig TG, Bearman SI, Babior BM: Effects of oxygen tension
and pH on the respiratory burst of human neutrophils. Blood
1979, 53:1133-1139.
29. Beachy JC, Weisman LE: Acute asphyxia affects neutrophil
number and function in the rat. Crit Care Med 1993, 21:1929-
1934.
30. Craven N, Williams MR, Field TR, Bunch KJ, Mayer SJ, Bourne FJ:
The influence of extracellular and phagolysosomal pH
changes on the bactericidal activity of bovine neutrophils
against Staphylococcus aureus. Vet Immunol Immunopathol
1986, 13:97-110.
31. Nakagawara A, Nathan CF, Cohn ZA: Hydrogen peroxide
metabolism in human monocytes during differentiation in
vitro. J Clin Invest 1981, 68:1243-1252.
32. Trevani AS, Andonegui G, Giordano M, Lopez DH, Gamberale R,
Minucci F, Geffner JR: Extracellular acidification induces
human neutrophil activation. J Immunol 1999, 162:4849-4857.
33. Miyazawa K, Inoue K: Complement activation induced by
human C-reactive protein in mildly acidic conditions. J
Immunol 1990, 145:650-654.
34. Osaki M, Sumimoto H, Takeshige K, Cragoe EJ, Jr., Hori Y,
Minakami S: Na+/H+ exchange modulates the production of
leukotriene B4 by human neutrophils. Biochem J 1989, 257:
751-758.
35. Grinstein S, Furuya W: Cytoplasmic pH regulation in phorbol
ester-activated human neutrophils. Am J Physiol 1986, 251:
C55-C65.
Shock 1998, 9:364-368.
experimental sepsis: improved survival and acid-base
balance with a synthetic colloid in a balanced electrolyte solu-
tion compared to saline. Crit Care Med 2002, 30:300-305.
38. Opie L: Effect of extracellular pH on function and metabolism
of isolated perfused rat heart. J Appl Physiol 1965, 209:1975-
1980.
39. Cooper D, Herbertson M, Werner H, Walley K: Bicarbonate
does not increase left ventricular contractility during L-lactic
acidemia in pigs. Am Rev Resp Dis 1993, 148:317-322.
40. Salzman AL, Wang H, Wollert PS, Vandermeer TJ, Compton CC,
Denenberg AG, Fink MP: Endotoxin-induced ileal mucosal
hyperpermeability in pigs: role of tissue acidosis. Am J Physiol
1994, 266:G633-G646.
41. Menconi MJ, Salzman AL, Unno N, Ezzell RM, Casey DM, Brown
DA, Tsuji Y, Fink MP: Acidosis induces hyperpermeability in
Caco-2BBe cultured intestinal epithelial monolayers. Am J
Physiol 1997, 272:G1007-G1021.
42. Gonzalez PK, Doctrow SR, Malfroy B, Fink MP: Role of oxidant
stress and iron delocalization in acidosis-induced intestinal
epithelial hyperpermeability. Shock 1997, 8:108-114.
43. Unno N, Menconi MJ, Smith M, Aguirre DE, Fink MP: Hyperperme-
ability of intestinal epithelial monolayers is induced by NO: effect
of low extracellular pH. Am J Physiol 1997, 272:G923-G934.
44. Unno N, Hodin RA, Fink MP: Acidic conditions exacerbate inter-
feron-gamma-induced intestinal epithelial hyperpermeability:
role of peroxynitrous acid. Crit Care Med 1999, 27:1429-1436.
45. Pedoto A, Nandi J, Oler A, Camporesi EM, Hakim TS, Levine RA:
Role of nitric oxide in acidosis-induced intestinal injury in
anesthetized rats. J Lab Clin Med 2001, 138:270-276.
46. Kellum JA, Song M, Venkataraman R: Effects of hyperchloremic
acidosis on arterial pressure and circulating inflammatory
molecules in experimental sepsis. Chest 2004, 125:243-248.
47. Gunnerson KJ, Saul M, Kellum JA: Lactic versus nonlactic meta-
bolic acidosis: outcomes in critically ill patients [abstract]. Crit
Care 2003, Suppl 2:17.
48. Severin T, Muller B, Giese G, Uhl B, Wolf B, Hauschildt S, Kreutz
W: pH-dependent LAK cell cytotoxicity. Tumour Biol 1994, 15:
304-310.
49. Loeffler DA, Juneau PL, Masserant S: Influence of tumour
physico-chemical conditions on interleukin-2-stimulated lym-
phocyte proliferation. Br J Cancer 1992, 66:619-622.
50. Loeffler DA, Juneau PL, Heppner GH: Natural killer-cell activity
under conditions reflective of tumor micro-environment. Int J
Cancer 1991, 48:895-899.
448
BE = base excess; C
Alb
= albumin concentration; C
Phos
= phosphate concentration; PCO
2
= partial CO
2
tension; SBE = standard base excess; SID =
strong ion difference; SIG = strong ion gap.
Critical Care December 2004 Vol 8 No 6 Wooten
Introduction
Acidbase derangements are commonly encountered in the
critical care unit [1], and there is renewed interest in the
precise description of these disorders in critically ill patients
[25]. This new interest has led to a renovation of the
quantitative assessment of physiological acidbase balance,
with increasing use of the Stewart model (strong ion
difference [SID] theory) to calculate acidbase balance in the
critically ill [2,3,6,7]. This method is discussed, particularly as
it pertains to the metabolic component of acidbase
derangements, as one of several approaches that may be
used in the intensive care unit for quantitative evaluation. As
with any mathematical model, a basic understanding of its
principles is useful for proper application and interpretation.
Stewart model
All equilibrium models of acidbase balance utilize the same
basic concept. Under the assumption of equilibrium or a
steady-state approximation to equilibrium, some property of
the system (e.g. proton number, proton binding sites, or
charge, among other possible properties) is enumerated from
the distribution of that property over the various species
comprising the system, according to the energetics of the
system manifested through the relevant equilibrium constants
of the various species under a given set of conditions
[5,812]. This function is calculated at the normal values and
then the abnormal values; from these the degree of change is
obtained to give information about the clinical acidbase
status of the patient. All of the apparently different methods
for assessing acidbase balance arise from this common
framework [5,12].
In the Stewart method, charge is taken as the property of
interest [7,11,13]. Using this property, acidbase status may
be expressed for a single physiologic compartment, such as
separated plasma, as follows [7,10,11,13]:
SID = C
1
z
1
C
2
z
2
C
n
z
n
(1)
Strong ions are those that do not participate in proton
transfer reactions, and the SID is defined as the difference
between the sum of positive charge concentrations and the
sum of negative charge concentrations for those ions that do
not participate in proton transfer reactions. C
n
are the
analytical concentrations of the various buffer species also in
the compartment (e.g. of the buffer amino acid groups on
albumin), and z
n
are the average charges of those various
species. The z
n
can be expressed as functions of pH and
Review
Science review: Quantitative acidbase physiology using the
Stewart model
E Wrenn Wooten
Attending Physician, Radiology Associates, PA, Little Rock, Arkansas, USA
Corresponding author: E Wrenn Wooten, wootenew@msn.com
Published online: 2 July 2004 Critical Care 2004, 8:448-452 (DOI 10.1186/cc2910)
Abstract
There has been renewed interest in quantifying acidbase disorders in the intensive care unit. One of
the methods that has become increasingly used to calculate acidbase balance is the Stewart model.
This model is briefly discussed in terms of its origin, its relationship to other methods such as the base
excess approach, and the information it provides for the assessment and treatment of acidbase
disorders in critically ill patients.
Keywords acidbase, base excess, Stewart model
449
equilibrium constants [11,12], and it is therefore convenient
to calculate SID using Eqn 1 from the pH and the
concentrations of relatively few buffer species, as opposed to
a direct calculation from a measurement of all of the various
strong ion species. In many implementations of the Stewart
method, contributions from the water equilibrium and from
carbonate species other than bicarbonate are neglected,
because these are small under physiologic conditions
[11,14,15]. The first term in Eqn 1 may then be equated with
the bicarbonate concentration, with the remaining terms
referring to other buffer species [11,14].
Plasma physiologic pH is then determined by the
simultaneous solution of Eqn 1 and the Henderson-
Hasselbalch Equation:
pH = pK + log
[HCO
3
]
(2)
S P
CO
2
Where for human plasma pK = 6.103. S = 0.0306 is the
equilibrium constant between aqueous and gas phase CO
2
[16,17]. [HCO
3
] is the concentration of plasma bicarbonate

in mmol/l, and PCO
2
is the partial CO
2
tension in Torr.
The standard technique for acidbase assessment [1,18]
may be recognized as a subset of the Stewart model [14], in
which the series in Eqn 1 is truncated at the first term to give
the following:
SID = [HCO
3
] (3)
In this approach the metabolic component of an acidbase
disorder is quantified as the change in plasma bicarbonate
concentration ([HCO
3
]) [18], which by Eqn 3 is also equal

to SID. This method is often sufficient and has been used
successfully to diagnose and treat countless patients, but it
has also been criticized as not strictly quantitative [19,20].
[HCO
3
] depends upon the PCO

2
and does not provide
complete enumeration of all species, because albumin and
phosphate also participate in plasma acidbase reactions
[15,17,20,21].
A more complete calculation may be undertaken for better
approximation by including more terms in the series in Eqn 1.
In addition, although z
n
is a nonlinear function of pH, it can be
approximated over the physiologic range by a more
computationally convenient linear form, such that for plasma
the following explicit expression is obtained [11,12,15]:
SID = [HCO
3
] + C
Alb
(8.0pH 41) + C
Phos
(0.30pH 0.4) (4)
Where C
Alb
and C
Phos
are plasma albumin and phosphate
concentrations, respectively. All concentrations are in mmol/l.
One may multiply albumin in g/dl by 0.15 to obtain albumin in
mmol/l, and phosphate in mg/dl by 0.322 to get phosphate in
mmol/l. The factors 8.0 and 0.30 are the molar buffer values
of albumin and phosphate, respectively. The buffer value is
the change in z
n
of a species for a one unit change in pH
[5,11,17]. Note that the ability of a system to resist pH
change also increases with C
Alb
and C
Phos
[11].
Equation 4 was obtained via a term by term summation over
all of the buffer groups in albumin and of phosphoric acid, as
performed by Figge and coworkers [15,21]. The theoretical
basis for the validity of this approach is well established [8],
and Eqn 4 has been shown to reproduce experimental data
well [11,12,15,21,22]. Some authors have argued that the
effects of plasma globulins should also be considered for
better approximation [17,20,23,24], although other
calculations suggest that the consideration of globulins
would be of little clinical significance in humans [22].
Consideration of the change in SID using Eqn 4 between
normal and abnormal states at constant albumin and
phosphate concentrations gives the following:
SID = [HCO
3
] + (8.0C
Alb
+ 0.30C
Phos
)pH (5)
Which is recognized to be of the same form and numerically
equivalent to the familiar Van Slyke equation for plasma,
yielding the plasma base excess (BE) [5,11,17,25].
Furthermore, Eqn 4 is of the same form as the CO
2
equilibration curve of the BE theory presented by Siggaard-
Andersen [11,17,20,25]. The BE approach and the Stewart
method are equivalent at the same level of approximation
[11,12,26].
Strong ion gap
A widely used concept arising from the Stewart approach is
the strong ion gap (SIG), which was popularized by Kellum
[27] and Constable [28]. This relies upon a direct calculation
of the SID as, for example, the following:
SID
m
= [Na
+
] + [K
+
] + 2[Mg
2+
] + 2[Ca
2+
]
[Cl
] [lactate
] [urate
]
(6)
Where SID
m
is the measured SID [27]. This direct measure-
ment is then compared with that generated via Eqn 4:
SIG = SID
m
SID (7)
This gives a higher level version of the familiar plasma anion
gap [1,18]. Some publications have used the notation SID
a
(for SID apparent) to refer to the variable SID
m
calculated
using Eq. 6, and SID
e
(SID effective) to refer to that
calculated using Eqn 4 [2,3,15,27]. SIG has been shown to
predict the presence of unmeasured ions better than the
conventional anion gap [28], as might be expected, given that
more variables are taken into account. Some unmeasured
ions that are expected to contribute to the SIG are -
hydroxybutyrate, acetoacetate, sulfates, and anions
associated with uremia [6].
450
Changes in noncarbonate buffer
concentration
SID expressed through the relationship of Eqn 5
unambiguously quantifies the nonrespiratory component of
an acidbase disturbance in separated plasma [11,17],
with the total concentrations of amphoteric species such
as albumin and phosphate remaining constant [11,12,17].
An amphoteric substance is one that can act as both an
acid and a base. Stewart and other investigators
[4,7,2933], though, have emphasized the role played by
changes in the noncarbonate buffer concentrations in
acidbase disorders. When the noncarbonate buffer
concentrations change, the situation becomes more
complex, and in general a single parameter such as SID
no longer necessarily quantifies the metabolic component
of an acidbase disorder, and enough variables must be
examined to characterize the disorder unambiguously.
Examples below demonstrate this point when the
concentrations of noncarbonate buffers change, through a
pathologic process or through resuscitation.
Table 1 gives several examples for separated human
plasma, including the normal values of case 1. Case 2
demonstrates a metabolic acidosis with constant
noncarbonate buffer concentrations, in which the SID of
10 mmol/l quantifies the metabolic component of the
acidbase disorder [11], which has been described as a
strong ion acidosis [4]. Case 3 gives values for the fairly
common occurrence of isolated hypoproteinemia. This too
gives a SID of 10 mmol/l, although the total weak acid
and weak base concentrations have both decreased [11].
The physiological interpretation of this condition in terms of
acidbase pathology is the subject of debate
[3,6,12,20,31,34]. Considering this to be an acidbase
disorder, some authors would classify this case as
hypoproteinemic alkalosis with a compensating SID
acidosis [4,6,3032]. More generally, this has been termed
a buffer ion alkalosis with compensating strong ion acidosis
[4]. If the mechanism of hypoalbuminemia is en bloc loss of
charged albumin with counterions in tow, for example in
nephrotic syndrome, then it seems dubious to describe this
process as compensation in the usual physiologic sense.
Also, note that both cases 2 and 3 have the same decrease
in SID, but the individual in case 2 is expected to be quite
sick with acidemia whereas the patient in case 3 is probably
not acutely ill, except for the effects of low oncotic pressure.
Although it has been suggested that alkalosis can result from
hypoproteinemia, with patients without adequate compensation
becoming alkalemic [29,32], the idea of alterations in protein
concentration as acidbase disorders per se has been
questioned [3,20]. The concept of the normal SID changing
as a function of protein concentration has been suggested
[3,11,12]. In such an instance, SID again quantifies the
metabolic component of an acidbase disturbance, essentially
renormalizing the noncarbonate buffer concentrations to the
abnormal values [11,12]. This is basically what has been
advocated in the past for BE [20,34], in which Eqn 5 uses
the abnormal protein and phosphate concentrations for
C
Alb
and C
Phos
[11]. Thus, the SID of 29 mmol/l in case 3 is
said to be normal for the decreased albumin concentration
[3], giving a SID of 0 mmol/l. This individual will, however,
be more susceptible to acidemia or alkalemia for a given
derangement, as expressed through the molar buffer values
and noncarbonate buffer concentrations, than would a
normal individual [5]. If SID is not renormalized as
described above, then BE and SID differ by an added
constant [11,12].
Another interesting issue is raised in the treatment of patients
with intravenous albumin or other amphoteric species. Kellum
previously pointed out that, based on the SID, one might think
that albumin solutions with a SID of 4050 mmol/l would be
alkalinizing to the blood, even though their pH is close to 6.0
[35]. This apparent paradox is resolved by again realizing
that, for amphoteric substances, one is not only changing the
SID but also increasing both the total weak acid and weak
base concentrations by increasing the total protein
concentration [9,11]. This highlights the point made by
Stewart concerning the necessity of considering all variables
in assessing acidbase balance [7,13]. A complete
calculation yields what is intuitively predicted that such a
solution is in fact acidifying to blood (unpublished data). One
might further speculate that the administration of unbuffered
albumin to patients may contribute to the reason why this
treatment has not been more successful in the critically ill
Table 1
Acidbase parameters for a normal and two abnormal cases
Case pH [HCO
3
] (mmol/l) C
Alb
(mmol/l) C
Phos
(mmol/l) PCO
2
(Torr) SID (mmol/l)
1 (normal) 7.40 24.25 0.67 1.16 40.0 39
2 7.30 15.27 0.67 1.16 31.7 29
3 7.40 24.25 0.15 1.16 40.0 29
Case 1 is for a normal individual, case 2 is for a metabolic acidosis at constant noncarbonate buffer concentrations, and case 3 is for
hypoproteinemia. C
Alb
, albumin concentration; C
Phos
, phosphate concentration; PCO
2
, partial CO
2
tension; SID, strong ion difference.
451
[36]. Extensive quantitative discussions regarding the
acidbase balance of administered fluids have typically not
been given in publications on resuscitation with amphoteric
colloids [3639], although this is an issue that should be
examined. Constable [40] recently gave a brief quantitative
discussion of acidbase effects of giving various crystalloids.
Model for whole blood
Several points arise in the comparison of SID with BE, as has
been performed in a number of studies [33,38,4144]. This
is in some respects a misplaced comparison, because BE
represents a difference whereas SID does not [11,26]. The
corresponding variable to SID in the BE formalism is the
concentration of total proton binding sites, while the BE
represents the change in this quantity from the normal value,
and corresponds to SID [11,12,17,26]. More significant,
clinical studies using Stewart theory have calculated the
separated plasma SID, while making comparison with the BE
for whole blood or the standard base excess (SBE)
[33,38,41,42], rather than the corresponding plasma BE.
Furthermore, consideration of only the plasma compartment
creates a potential source of error, because separated
plasma versions of the Stewart method quantify only a portion
of the acidbase disorder [12,17,45]. An equation for the
SID of whole blood has recently been derived, partly to
address this issue [12].
(8)
Where (E) is the hematocrit, C
Hgb
(B) is the hemoglobin
concentration of whole blood, and C
DPG
(E) is the 2,3-
diphosphoglycerate concentration in the erythrocyte. Again,
concentrations are in mmol/l, and one may multiply
hemoglobin in g/dl by 0.155 to obtain hemoglobin in mmol/l.
The normal 2,3-diphosphoglycerate concentration in the
erythrocyte is 6.0 mmol/l [12]. The P, B, and E
designations stand for plasma, whole blood, and erythrocyte
fluid, respectively. The corresponding Van Slyke form has
also been obtained, and is numerically identical to BE for
whole blood [12].
The SBE, as mentioned above, is also widely used
[3,17,20,25]. This parameter reflects the extracellular
acidbase status and approximates the in vivo BE for the
organism [17,20,25]. The Van Slyke equation for SBE
approximates this situation via a 2:1 dilution of whole blood in
its own plasma [17,20,25]. It should be borne in mind,
therefore, that Eqn 4 may prove more concordant with clinical
data than Eqn 8, since the plasma expression may produce
values closer to the in vivo condition because of the
distribution functions of various species across the whole
organism [17].
Stewart theory and mechanism
Finally, the Stewart model is taken by some to be a
mechanistic description of acidbase chemistry in which
changes only occur by alteration in PCO
2
, SID, or
noncarbonate buffer concentrations because these are the
only true independent variables; changes never occur by
addition or removal of H
+
to the system or by changes in
[HCO
3
] because these are dependent variables [7,13]. It is

said that because the Stewart theory provides mechanistic
information, it is superior to the BE approach [3,35,46,47].
Support for this point of view is offered in the form of
philosophic arguments regarding the nature of independence
[7,13], as well as studies showing that the Stewart model
accurately predicts what is observed experimentally
[30,42,44,48]. However, like the BE approach and like any
other method derived from considerations involving the
calculation of interval change via the assessment of initial and
final equilibrium states, the Stewart method does not produce
mechanistic information [8,35]. These are basically
bookkeeping methods. To believe otherwise risks falling prey
to the computo, ergo est (I calculate it, therefore it is) fallacy.
What is thus required for mechanistic understanding is the
collection of actual mechanistic data, perhaps obtainable
through isotopic labeling and kinetics experiments.
Conclusion
Both experimental and theoretical data have shown that the
Stewart method is accurate for describing physiological
acidbase status, and the use of the SIG potentially offers an
improvement over the traditional anion gap, but because the
Stewart method proceeds from the same common framework
as the BE approach, it theoretically offers no quantitative
advantage over BE at corresponding levels of approximation
[11,12,26,35,49]. As such, it remains to be seen whether the
renovation of acidbase assessment afforded by the Stewart
approach constitutes a radical new architecture for
understanding acidbase physiology, or whether it is simply a
new faade.
Competing interests
References
1. Irwin RS, Rippe JM (editors): Intensive Care Medicine. Philadel-
phia, PA: Lipincott, Williams & Wilkins; 2003.
2. Kellum JA: Metabolic acidosis in the critically ill: lessons from
physical chemistry. Kidney Int 1998, Suppl 66:S81-S86.
Crit Care 2000, 4:6-14.
4. Constable PD: Clinical assessment of acid-base status: com-
parison of the Henderson-Hasselbalch and strong ion
approaches. Vet Clin Path 2000, 29:115-128.
5. Corey HE: Stewart and beyond: new models of acidbase
6. Constable PD: Clinical assessment of acid-base status: strong
ion difference theory. Vet Clin N Am 1999, 15:447-471.
8. Tanford C: Physical Chemistry of Macromolecules. New York:
John Wiley & Sons, Inc.; 1961.
P 3
] [HCO ) E ( 49 . 0 1 ) B ( SID

( ) ( ) 0.4 (P) pH .30 0 ) P ( C 41 (P) pH 0 . 8 ) P ( C ) E ( 1 + +
Phos Alb
+ ( ) 5 . 1 7 21 . 0 (P) pH .2 0 1 ) B ( C
Hgb
(
,
\
,
(
j
+ 4 . 0 0.21 (P) pH 70 . 0 ) E ( C ) E (
DPG
452
9. Guenther WB: Unified Equilibrium Calculations. New York: John
Wiley & Sons, Inc.; 1991.
10. Butler JN, Cogley DR: Ionic Equilibrium: Solubility and pH Calcu-
lations. New York: John Wiley & Sons, Inc.; 1998.
11. Wooten EW: Analytic calculation of physiological acid-base
13. Stewart PA: Independent and dependent variables of acid-
base control. Respir Physiol 1978, 33:9-26.
equilibria: application to horse plasma. J Appl Physiol 1997,
83:297-311.
16. Burus CA, Ashwood ER (Editors): Tietz Textbook of Clinical
Chemistry, 2nd ed. Philadelphia, PA: Saunders; 1994.
17. Siggaard-Andersen O: The Acidbase Status of the Blood, 4th
ed. Baltimore, MD: Williams and Wilkins; 1974.
18. Nairns RG, Emmett M: Simple and mixed acid-base disorders:
a practical approach. Medicine (Baltimore) 1980, 59:161-187.
19. Severinghaus JW: Siggaard-Andersen and the Great Trans-
Atlantic Acid-Base Debate. Scand J Lab Invest 1993, Suppl
214:99-104.
base (strong ion difference) as a measure of a non-respira-
tory acidbase disturbance. Acta Anaesthesiol Scand 1995,
Suppl 107:123-128.
21. Figge J, Rossing TH, Fencl V: The role of serum proteins in
acidbase equilibria. J Lab Clin Med 1991, 117:453-467.
22. Watson PD: Modeling the effects of proteins on pH in plasma.
J Appl Physiol 1999, 86:1421-1427.
23. Siggaard-Andersen O, Rorth M, Strickland DAP: The buffer
value of plasma, erythrocyte fluid and whole blood. In Blood
pH, Gases and Electrolytes. Workshop on pH and Blood Gases,
National Bureau of Standards, 1975. National Bureau of Stan-
dards (US), Special Publications 1977, 450:11-19.
net protein charge and A
tot
and K
a
of nonvolatile buffers in
human plasma. J Appl Physiol 2003, 95:620-630.
Lab Invest 1977, Suppl 146:15-20.
26. Schlichtig R: [Base excess] vs. [strong ion difference]. Which
is more helpful? Adv Exp Med Biol 1997, 411:91-95.
55.
28. Constable PD, Hinchcliff KW, Muir WW: Comparison of anion
gap and strong ion gap as predictors of unmeasured strong
ion concentration in plasma and serum from horses. Am J Vet
Res 1998, 59:881-887.
29. McAuliffe JJ, Lind LJ, Lieth DE, Fencl V: Hypoproteinemic alkalo-
sis. Am J Med 1986, 81:86-90.
Physiol 1986, 61:2260-2265.
31. Jabor A, Kazda A: Modelling of acidbase equilibria. Acta
Anaesthesiol Scand 1995, Suppl 107:119-122.
32. Wilkes P: Hypoproteinemia, strong ion difference, and acid-
1740-1748.
33. Rocktaeschel J, Morimatsu H, Uchino S, Goldsmith D, Poustie S,
Story D, Gutteridge G, Bellomo R: Acidbase status of critically
ill patients with acute renal failure: analysis based on Stewart-
Figge methodology. Crit Care 2003, 7:R60-R66.
34. Moon JB: Abnormal base excess curves. Pediatr Res 1967,
1:333-340.
35. Wooten EW: Strong ion difference theory: more lessons from
physical chemistry. Kidney Int 1998, 54:1769-1770.
36. Cochrane Injuries Group Reviewers: Human albumin adminis-
tration in critically ill patients: systematic review of random-
ized controlled trials. BMJ 1998, 317:235-240.
metabolic acidosis using polygeline pump prime. Int Care
Med 1999, 25:680-685.
Finsterer U: Acidbase changes caused by 5% albumin versus
normovolemic hemodilution. Anesthesiol 2000, 93:1174-1183.
acidosis with sodium bicarbonate or tris-hydroxylmethyl
Analg 2003, 96:1201-1208.
40. Constable P: Stewart approach is not always a practical tool.
Anesth Analg 2004, 98:271-272.
Med 2002, 30:300-305.
ence determines metabolic acid-base change during in vitro
hemodilution. Crit Care Med 2002, 30:157-160.
1270.
44. Waters JH, Bernstein CA: Dilutional acidosis following het-
astarch or albumin in healthy volunteers. Anesthesiol 2000,
93:1184-1187.
45. Davenport HW: The ABC of Acid-base Chemistry, 6th ed.
Chicago, IL: University of Chicago Press; 1974.
46. Waters JH, Scanlon TS, Howard RS, Leivers D: Role of minor
electrolytes when applied to Stewarts acidbase approach in
an acidotic rat model. Anesth Analg 1995, 81:1043-1051.
of strong ion acidosis. Anesth Analg 2003, 96:919-922.
48. Alfaro V, Torras R, Ibanez J, Palacios L: A physicalchemical
analysis of the acid-base response to chronic obstructive pul-
monary disease. Can J Physiol Pharmacol 1996, 74:1229-
1235.
49. Cameron JN: Acidbase homeostasis: past and present per-
spectives. Phys Zool 1989, 62:845-865.
Page 1 of 5
(page number not for citation purposes)
Abstract
In the critically ill, metabolic acidosis is a common observation and,
in clinical practice, the cause of this derangement is often multi-
factorial. Various measures are often employed to try and
characterise the aetiology of metabolic acidosis, the most popular
of which is the anion gap. The purpose of the anion gap can be
perceived as a means by which the physician is alerted to the
presence of unmeasured anions in plasma that contribute to the
observed acidosis. In many cases, the causative ion may be easily
identified, such as lactate, but often the causative ion(s) remain
unidentified, even after exclusion of the classic causes. We
describe here the various attempts in the literature that have been
made to address this observation and highlight recent studies that
reveal potential sources of such hitherto unmeasured anions.
Introduction
Metabolic acidosis remains a common problem in acute
medicine and is frequently encountered on the intensive care
unit (ICU) [1-3]. Although many classic causes of metabolic
acidosis are known, including diabetic ketoacidosis, lactic
acidosis and the ingestion of acid-generating poisons, the
origin is often multifactorial and, indeed, often cannot be
ascribed solely to such classic causes or a single causative
anion. In such cases, the source of the acidosis remains
unidentified or unmeasured. For example, given that
hydroxybutyrate is seldom measured, diabetic ketoacidosis is,
strictly speaking, an example of acidosis associated with
large quantities of an unmeasured anion, although in practice
its concentration is regularly inferred. Similarly, it is only in the
past 15 years or so that prompt and repeatable measurement
of arterial blood lactate has become commonplace. Prior to
this, lactic acidosis could also reasonably be considered to
represent the presence of an unmeasured anion.
One of the earliest tools for addressing the potential aetiology
of metabolic acidosis is that of the anion gap, which even in
its simplest form helps to characterise many cases of
metabolic acidosis. This measure has undergone various
refinements over the years but one of its purposes is to alert
the physician to the presence of unmeasured ions in plasma
[4-7]. Those studying critically ill patients with metabolic
acidosis have been aware that such a simple categorisation
is often an inadequate description of the metabolic state of
these patients. In lactic acidosis, for example, there is often a
significant discrepancy between the blood lactate
concentration and the base deficit and, more tellingly, when
calculations are made during bicarbonate-based haemo-
filtration, it is apparent that significant quantities of acid other
than lactic acid are being titrated by the administered
bicarbonate. This has given rise to the concept of the
unmeasured anions as an important component of human
metabolic acidosis. Sometimes these appear to be quanti-
tatively significantly more important than lactic acid itself. But
what is the nature of these unmeasured anions? We discuss
the evidence to date coupled with recent work from our
laboratory that may go some way in elucidating the nature of
these anions.
Identifying unmeasured anions
The presence of unmeasured anions contributing to meta-
bolic acidosis has been recognised for some time and as
early as 1963 Waters and colleagues, whilst discussing
lactic acidosis, hypothesised that under certain conditions
disturbances in acid-base balance may be characterised by
the accumulation of an organic acid other than lactate [8].
Furthermore, studies from Cohens group in London described
a case where hydroxybutyrate contributed significantly to an
observed metabolic acidosis of a non-diabetic patient [9].
The same group also demonstrated an elevation in succinate
levels in both hypoxic patients and perfused hypoxic canine
livers [10]. They proposed that disturbances in the oxidation
of succinate to oxaloacetate could account for this. Interest in
this area was rekindled by studies on critically ill patients in
which elevations in anion gap could not be accounted for
solely by increased lactate levels [11,12]. Further work
Review
Unmeasured anions in metabolic acidosis: unravelling the mystery
Lui G Forni
1,2
, William McKinnon
3
and Philip J Hilton
3
1
Department of Critical Care, Worthing Hospital, Worthing, West Sussex BN11 2DH, UK
2
Brighton and Sussex Medical School, University of Sussex, Brighton, East Sussex BN1 9PX, UK
3
Renal Laboratory, St Thomas Hospital, London SE1 7EH, UK
Corresponding author: Lui G Forni, lui.forni@wash.nhs.uk
Published: 12 July 2006 Critical Care 2006, 10:220 (doi:10.1186/cc4954)
ICU = intensive care unit.
Page 2 of 5
Critical Care Vol 10 No 4 Forni et al.
examining the concentrations of other hitherto unmeasured
ions such as urate and phosphate as well as plasma proteins
could not account for the observed anion gap [13,14]. To try
to elucidate these species further, several workers have
employed animal models.
Animal studies
Some of the earliest studies that attempted to identify the
nature of the unmeasured anions were performed in animal
models. In 1990, Rackow and colleagues [15] assessed the
contribution of such species to the anion gap observed in
rats following caecal perforation. Compared to controls, the
septic animals demonstrated a metabolic acidosis with an
increase in plasma lactate and decrease in bicarbonate
concentrations. Only 15% of the anion gap observed could
be explained by lactate. The concentrations of pyruvate,
-hydroxybutyrate, acetoacetate, citrate as well as some
amino acids were determined. No differences in these anions
could be detected between the study group and sham
animals. However, no detail as to the handling of the samples
was provided. These studies followed earlier work by Gossett
and colleagues [16] on critically ill horses with increased
anion gap acidosis. Again, the unexplained anion gap could
not be accounted for by pyruvate, -hydroxybutyrate, aceto-
acetate, phosphate or albumin.
In other studies on diarrhoeic calves, the observed anion gap
was explained in part, but not completely, by the
accumulation of D-lactate [17]. To date, animal studies have,
therefore, provided little information as to the nature of the
unmeasured anions. Further animal work, employing a canine
model of sepsis, demonstrated that the liver released anions
into the circulation at a rate of 0.12 mEq/minute [18]. This
study also observed that the gut became a consumer of
anions following development of endotoxaemia. Other canine
models have proposed that, in lactic acidosis, impaired
extraction of lactate by the liver coupled with increased
splanchnic production of lactate contributed to the
generation of the metabolic acidosis. Studies with humans,
however, do not support this view [19].
Studies on ICU patients
Pyroglutamic acidaemia
Pyroglutamic acidaemia is an inherited disorder presenting in
infancy due to a deficiency of either 5-oxoprolinase or gluta-
thione synthetase. Several case reports have described this
phenomenon occurring in adults, causing an elevated anion
gap acidosis often in association with drug administration
[20]. An early study of ICU patients described four patients in
whom pyroglutamic acid levels were noted to be elevated
[21]. The authors suggested that patients with this condition
be screened for obvious precipitants. However, a further
study examined pyroglutamic acid levels in 23 ICU patients
with metabolic acidosis and an unexplained increase in ion
gap. They found no correlation between the ion gap and
pyroglutamic acid levels and concluded that, in their
population, pyroglutamic acid could not account for the
unmeasured anions [22].
Krebs cycle intermediates
We recently attempted to identify the missing anions, arguing
that being negatively charged, they should reveal themselves
on negative ion mass spectrometry and should be at least
partially separable by ion exchange chromatography. There
was no predetermined view as to the likely nature of the
anions. Plasma from patients with various forms of metabolic
acidosis was examined. The patients were acidotic with an
average arterial pH of 7.18 (0.11) and a base deficit of
13.4 mmol/l (4.7) [23].
Figure 1 shows an ion exchange chromatogram/negative ion
mass spectrum of a plasma extract from a patient with
metabolic acidosis of unknown aetiology. This shows peaks
of relatively low mass that fitted those of known Krebs cycle
components. Standards of these anions proved to have
identical retention times to the plasma-derived peaks.
Interestingly, no ions attributable to other substances could
be seen apart from urate, which was also seen in control
samples. For comparison, we present the spectrum obtained
from a patient with diabetic ketoacidosis where the large
peaks attributable to acetoacetate and -hydroxybutyrate are
clearly seen [24].
These preliminary results led us to examine the anions of the
Krebs cycle using enzyme assay (we also measured D-
lactate). Table 1 simplifies our results and, as can be seen,
plasma from patients with diabetic ketoacidosis showed
significant increases relative to the control values in -
ketoglutarate, malate and D-lactate levels. However, citrate
and succinate concentrations were not elevated. In lactic
acidosis, increased concentrations of citrate, isocitrate, -
ketoglutarate, succinate, malate and D-lactate were
observed. In patients with an acidosis of unknown origin
(acidosis disproportionate to the blood lactate
concentration), elevations in the concentrations of isocitrate,
-ketoglutarate, succinate, malate and D-lactate were seen.
This observation that plasma concentrations of acids usually
associated with the Krebs tricarboxylic acid cycle are
significantly increased in patients with lactic acidosis as well
as those with unexplained acidosis with normal or near
normal blood lactate concentrations may go some way to
addressing the imbalance in the anion or strong ion gap.
In the main, these anions are effectively fully ionised at the
measured pH but, unlike lactate, they are not all monobasic,
with tribasic acids (citric and isocitric) contributing three
protons, whilst the dibasic acids (-ketoglutaric, malic and
succinic) add two protons to the solution on ionisation. Our
study showed that, on average, the contribution to the
observed anion gap by such anions was regularly in excess of
3 mEq/l and, in some cases, over 5 mEq/l. Therefore, the role
of these anions in generating the anion gap is of much
Page 3 of 5
greater significance than is apparent from their molarity. We
would stress that in data such as these, at least as much
attention should be given to the extreme values as to the
means.
From our preliminary work it became clear that rapid
separation of the plasma from red cells and also from its
proteins through centrifugation and ultrafiltration of the
samples together with prompt assay was vital. Even at 20C
we observed steady degradation of the measured anions. The
most extreme example of the instability of these metabolic
intermediates is oxaloacetate, whose half-life in aqueous
solutions is so short that it is effectively unmeasurable [25].
D-lactate
Although we observed modest elevations in D-lactate
concentration in both diabetic and non-diabetic acidosis, this
never reached levels in these groups that would impact
significantly on the acid-base status of the patients. However,
in the patients with a normal anion gap acidosis, the level of
D-lactate was significantly raised. D-lactate is normally
present at nanomolar concentrations through the metabolism
Figure 1
Ion exchange chromatogram/negative ion mass spectra of plasma from a patient with diabetic ketoacidosis (top) and a patient with acidosis of
unknown aetiology (bottom). Liquid chromatography/electrospray ionisation mass spectrometry was performed on a Hewlett-Packard Series 1100
liquid chromatography system directly coupled to a Series 1100 Mass Spectrometer fitted with electrospray ionisation and operating in negative
ion mode (Agilent Technologies UK Ltd, Wokingham, Berkshire, UK). The extracted ion currents are shown.
Table 1
Relative changes observed in Kreb's cycle intermediates and D-Lactate in patients with differing causes of acidosis
Acid DKA LA AUO NAG
Citrate ?
a

Isocitrate + +++ +++
-Ketoglutarate +++ +++ +++
Succinate +++ +
Malate +++ +++ +++
D-lactate +++ +++ +++ +++
Dashes represent no significant difference from controls; a plus sign represents p < 0.02; three plus signs represent p < 0.001.
a
This result may
be unreliable since four of the patients in this group had received an infusion of heparin (containing citrate as an anticoagulant) prior to the blood
sample being obtained. AUO, acidosis of unknown origin; DKA, diabetic ketoacidosis; LA lactic acidosis; NAG, normal anion gap acidosis.
of methylglyoxal, although millimolar concentrations can be
observed through excess gastrointestinal metabolism and
elevated levels of D-lactate have been observed in critically ill
patients with intestinal ischaemia [26]. Interestingly, plasma
D-lactate levels have been proposed as an early potential
predictor of reduced 28 day ICU mortality [27] and has been
suggested as a tool for assessing colonic ischaemia in post
operative patients [28]. In rat models, however, D-lactate has
not been confirmed as a reliable marker of gut ischaemia
[29]. However, what is clear is that D-lactate may contribute
to metabolic acidosis and, in some cases, may contribute
significantly to the unmeasured anions.
Hydroxybutyrate
Another anion that does not fit neatly into this concept of
Krebs cycle acidaemia is hydroxybutyrate in non-diabetics.
We detected this anion in concentrations up to 4 mEq/l and,
as such, it could be a significant contributor to the un-
measured anions. We presumed that this was effectively a
marker for the metabolic changes of starvation in the
patients in whom it was demonstrated, in agreement with
earlier studies [9].
Discussion
Many studies have highlighted the presence of unmeasured
anions in critically ill patients with metabolic acidosis,
although few have been successful in addressing their
chemical nature. The prognostic significance of unmeasured
anions is also a source of debate but recent studies seem to
suggest some predictive ability [30,31]. Certainly, the study
from Dondorp and colleagues [30] supports this view,
although the area under receiver operator curve for strong ion
gap toward mortality was just 0.73. However, all other
predictors also had values <0.8. Interestingly, recent studies
on the primary patho-physiological events of malarial infection
in animals revealed up-regulation of transcription of genes
that control host glycolysis [32]. One may speculate that the
unmeasured anions noted in severe malaria may, therefore,
be related to intermediary metabolism, in keeping with our
studies. Other workers have demonstrated the presence of
organic acids commonly associated with intermediary
metabolism under various conditions. Tricarboxylic acids have
been detected in human urine [33] and various organic acids
detected in the haemofiltrate of patients with acute renal
failure where the presence of elevated citrate levels was
loosely associated with a worse prognosis [34]. Furthermore,
citrate, malate and cis-aconitate have been detected in
patients with metabolic acidosis ascribed to salicylate
poisoning [35].
The results obtained from our work suggest that the role of
anions principally associated with the Krebs cycle in the
generation of the anion gap in classic lactic acidosis may be
greater than previously thought and that these anions may
also have a significant role in the generation of the anion gap
in patients with acidosis of unknown cause. Their concentra-
tions did not differ significantly from control values in patients
with normal anion gap acidosis.
The likely source for the generation of these observed anions
is a matter of speculation and we have no direct evidence for
the site of production. Clearly, the mitochondria are one
possible source and the process could reflect mitochondrial
dysfunction, a concept that is currently an area of research in
critical care. It seems unlikely that the acidaemia per se is
responsible for the generation of increased levels of Krebs
intermediates given the normal values found in patients with
normal anion gap acidosis. It may reflect a physiological
response to a limitation in available oxygen supply and recent
work from our group has demonstrated increased levels of
Krebs cycle intermediates in normal subjects following severe
exercise [35].
The Krebs cycle functions not only as a catalytic process in
intermediary metabolism but also as a source of substrates
for other metabolic pathways. For example, during protein
synthesis, -ketoglutarate and oxaloacetate are removed from
the cycle to become aminated to glutamate and aspartate
(cataplerosis). This inevitably results in anaplerotic reactions,
ensuring continued function by replenishing tricarboxylic acid
intermediates. In gluconeogenesis, oxaloacetate is converted
to phosphoenolpyruvate and is lost to the Krebs cycle.
Lipogenesis requires the transfer of citrate from the
mitochondria to the cytosol as that is the site at which the
synthetic process occurs. In disease, the opposite is true;
anaplerotic reactions (those that generate rather than
consume Krebs cycle keto-acids) are likely to predominate.
Excess protein catabolism in particular will give rise to the
component amino acids. These approximately neutral
compounds are rapidly transaminated and/or deaminated to
form oxaloacetic acid, -ketoglutaric and succinyl CoA
(effectively succinic acid), thereby potentially providing an
excess of acidic Krebs cycle components. There are few data
available from the critically ill on these processes. However,
under other conditions of stress, such as prolonged
starvation or extreme exercise [36], the levels of tricarboxylic
acid levels have been measured and it has been shown that
glutamine, for example, undergoes deamination (an ana-
plerotic process) to form -ketoglutarate, which enters the
Krebs cycle and is sequentially converted to malate, which
then leaves the mitochondria. Malate is oxidized in the cytosol
to oxalocetate, which is in turn converted to phospho-
enolpyruvate.
Conclusion
The phenomenon of unexplained metabolic acidosis is well
recognised, as is the generation of unexplained anions. Little
is known as to the nature of these species, although recent
studies suggest that anions usually associated with the Krebs
cycle may contribute to the observed anion or strong-ion
gap. Although these observations go no way to explaining
their genesis, they may provide the first glimpse of the
Critical Care Vol 10 No 4 Forni et al.
Page 4 of 5
underlying derangement in the metabolic acidosis associated
with unmeasured anions.
Competing interests
The authors declare that they have no competing interests.
References
1. Kellum JA: Diagnosis and treatment of acid-base disorders. In
Textbook of Critical Care. Edited by Grenvik A, Ayres SM, Hol-
brook PR, Shoemaker WC. Philadelphia: WB Saunders;
2000:839-853.
2. Gauthier PM, Szerlip HM: Metabolic acidosis in the intensive
care unit. Crit Care Clin 2002, 18:289-308.
Crit Care 2000, 4:6-14.
4. Nairns RG, Emmett M: Simple and mixed acid-base disorders:
A practical approach. Medicine (Baltimore) 1980, 59:161-187.
5. Rossing TH, Maffeo N, Fencl V: Acid-base effects of altering
Physiol 1986, 61:2260-2265.
7. Story DA, Poustie S, Bellomo R, Story DA, Poustie S, Bellomo R:
Estimating unmeasured anions in critically ill patients: anion-
gap, base-deficit, and strong-ion-gap. Anesthesia 2002, 57:
1109-1114.
8. Waters WC, Hall JD, Schwartz WB: Spontaneous lactic acido-
sis. The nature of the acid-base disturbance and considera-
tions in diagnosis and management. Am J Med 1963, 35:
781-793.
9. Barnardo DE, Cohen RD, Iles RA: Idiopathic lactic and -
hydroxybutyric acidosis. BMJ 1970, 4:348-349.
10. Iles RA, Barnett D, Strunin L, Strunin M, Simpson BR, Cohen RD:
The effect of hypoxia on succinate metabolism in man and
the isolated perfused dog liver. Clin Sci 1972, 42:35-45.
11. Mehta K, Kruse JA, Carlson RW: The relationship between
anion gap and elevated lactate. Crit Care Med 1986, 14:405.
12. Mecher C, Rackow EC, Astiz ME, Weil MH: Unidentified anion
during in metabolic acidosis during septic shock. Clin Res
1989, 37:10A.
13. Mecher C, Rackow EC, Astiz ME, Weil MH: Unaccounted for
anion in metabolic acidosis during severe sepsis in humans.
Crit Care Med 1991, 19:705-711.
14. Niwa T: Organic acids and the uraemic syndrome: Protein
metabolic hypothesis in the progression of chronic renal
failure. Semin Nephrol 1996, 16:167-182
15. Rackow EC, Mecher C, Astiz ME, Goldstein C, McKee D, Weil
MH: Unmeasured anion during severe sepsis with metabolic
acidosis. Circ Shock 1990, 30:107-115.
16. Gossett KA, Cleghorn B, Adams R, Church GE, McCoy DJ,
Carakostas MC, Flory W: Contribution of whole blood L-lactate,
pyruvate, D-lactate, acetoacetate and 3-hydroxybutyrate con-
centrations to the plasma anion gap in horses with intestinal
disorders. Am J Vet Res 1987, 1:72-75.
17. Ewaschuk JB, Naylor JM, Zello GA: Anion gap correlates with
serum d- and dl-lactate concentration in diarrheic neonatal
calves. J Vet Intern Med 2003, 17:940-942.
during acute endotoxaemia. J Appl Physiol 1995, 78:2212-
2217.
19. Chrusch C, Bands C, Bose D, Li X, Jacobs H, Duke K, Bautista E,
Eschum G, Light B, Mink SN: Impaired hepatic extraction and
invcreased splanchnic production contribute to lactic acidosis
in canine sepsis. Am J Respir Crit Care Med 2000, 161:517-
526.
20. Tailor P, Raman T, Garganta CL, Njalsson R, Carlssin K, Ristoff E,
Carey HB: Recurrent high anion gap metabolic acidosis sec-
ondary to 5-oxoproline (pyroglutamic acid). Am J Kidney Dis
2005, 46:4-10.
21. Dempsey GA, Lyall HJ, Corke CF, Scheinkestel CD: Pyroglu-
tamic acidemia: A cause of high anion gap metabolic acido-
sis. Crit Care Med 2000, 28:1803-1807.
22. Mizcock BA, Belyaev S, Mecher C: Unexplained metabolic aci-
dosis in critically ill patients: the role of pyroglutamic acid.
Intensive Care Med 2004, 30:502-505.
23. Forni LG, McKinnon W, Lord GA, Treacher DF, Peron J-MR,
Hilton PJ: Circulating anions usually associated with the Krebs
cycle in patients with metabolic acidosis. Crit Care 2005, 9:
R591-R595.
24. McKinnon W, Lord GA, Forni LG, Peron J-MR, Hilton PJ: A rapid
LC-MS method for determination of plasma anion profiles of
acidotic patients. J Chromatogr 2006, B833:179-185.
25. Tsai CS: Spontaneous decarboylation of oxalacetic acid.
Canadian J Chem 1967, 45:873-880.
26. Ewaschuk JB, Naylor JM, Zello GA: D-Lactate in human and
ruminant metabolism. J Nutrition 2005, 135:1619-1626.
27. Sapin V, Nicolet L, Aublet C-B, Sangline F, Roszyk L, Dastugue B,
Gazuy N, Deteix P, Souweine B: Rapid decrease in plasma D-
lactate as an early potential predictor of diminished 28-day
mortality in critically ill septic shock patients. Clin Chem Lab
Med 2006, 44:492-496.
28. Assadian A, Assadian O, Senekowitsch C, Rotter R, Bahrami S,
Frst W, Jaksch W, Hagmller GW, Hbl W: Plasma D-lactate
as a potential early marker for colon ischaemia after open
aortic reconstruction. Eur J Vasc Endovasc Surg 2006, 31:470-
474.
29. Collange O, Tamion F, Chanel S, Hue G, Richard V, Thuilliez C,
Dureuil B, Plissonnier D: D-lactate is not a reliable marker of
gut ischemia-reperfusion in a rat model of supraceliac aortic
clamping. Crit Care Med 2006, 34:1415-1419.
30. Dondorp AM, Chau TTH, Phu NH, Mai NTH, Loc PP, Chuong LV,
Sinh DX, Taylor A, Hien TT, White NJ, Day NPJ: Unidentified ions
of strong prognostic significance in severe malaria. Crit Care
Med 2004, 32:1683-1688.
strong ion difference and strong ion gap predict outcome
1124.
32. Sexton AC, Good RT, Hansen DS, DOmbrain MC, Buckingham
L, Simpson K, Schofield L: Transcriptional profiling reveals sup-
pressed erythropoiesis, up-regulated glycolysis, and inter-
feron-associated responses in murine malaria. J Infect Dis
2004, 189:1245-1256.
33. Zaura DS, Metcoff J: Quantification of seven tricarboxylic acid
cycle and related acids in human urine by gas-liquid chro-
matography. Anal Chem 1969, 41:1781-1787.
34. Guth H-J, Zschiesche M, Panzig E, Rudolph PE, Jager B, Kraatz
G: Which organic acids does haemofiltrate contain in the
presence of acute renal failure? Int J Artificial Organs 1999,
22:805-810.
35. Dienst S, Greer BS: Plasma tricarboxylic acids in salicylate
poisoning. J Maine Med Assoc 1967, 85:11-14.
36. Owen OE, Kalhan SC, Hanson RW: The key role of anaplerosis
and cataplerosis for citric acid cycle function. J Biol Chem
2002, 277:30409-30412.
Page 5 of 5

AcidoBaseDrKellum PDF

Hochgeladen von

Dokumentinformationen

Originalbeschreibung:

Originaltitel

Copyright

Verfügbare Formate

Dieses Dokument teilen

Dokument teilen oder einbetten

Freigabeoptionen

Stufen Sie dieses Dokument als nützlich ein?

Sind diese Inhalte unangemessen?

Copyright:

Verfügbare Formate

AcidoBaseDrKellum PDF

Hochgeladen von

Copyright:

Verfügbare Formate

253

Available online http://ccforum.com/content/8/4/253

Lewis developed a further definition in the 1920s, to

[9,12]. However, all weak acids in a given solution

], plasma bicarbonate concentration; ,

(b) pKa1 + log = pH = pKa2 + log

and lactate), partially dissociated weak acids (such as

]. The study of acidbase balance now becomes

] is the concentration of dissociated

] is the concentration of dissociated weak

]) [10], the SIG is normally equal to 0.

). SIG, strong ion gap (see text).

] = 1000 Kcl PCO

] = 24.25 mmol/l, [albumin] =

= 115 mEq/l, and BE = 10 mEq/l. The

], the buffer value () and

] parameter can be determined from the

. Because they are

, lactate) in the blood. [SID] is so important

] and pH depend simply and quite

is not, but rather reacts with [H

and lactate). This

ratio has been proposed as a simple way to delineate the

of 100 mEq/l, resulting

would increase to 111 mEq/l.

is extracellular fluid, the

, the patient has had a reduction in

) as well as an associated form (AH). When the SID

and lactate. It is important to note

therapeutically reduces the plasma excess positive

while preserving plasma Na

falls; the increased SID

decreases the pH unaccompanied by hypoperfusion and

The group of ions summarized as A

are weak anions (pKa

] varies with pH, A

] + 14.84 (pH 7.4) 24.4}

rather than HCO

]). Large volumes of saline

] rises or falls, forcing acidbase in the direction

] is normally under 20 mmol/l [41].

) stays extracellullar, whereas most of the retained

is left behind in the

] values to cater for varying osmolality requirements. The

] 116 116 116 114

] 128 145 120 154 154 124

cotransporter; NAE = net acid excretion; PCO

provides an inexhaustible source and sink of acidbase

exchanger band3 in red blood cells [23]),

when extracellular homeostasis alone

. This will typically be the case in

) cotransporter. kNBC1 activity leads to a

. A recently identified potassium

and resulting decline in SID will also explain the findings.

sensor. A pH sensor would not be

sensor transmits the

by means of anion exchanger (AE1) at the basolateral

channels [44]. Likewise, functional Cl

channels (CIC5) are necessary to acidify transport vesicles in

cotransporter, KCC4, which is located in the baso-

exchanger AE1. Although intracellular pH was not

. In a study [50] of perfused rabbit