Concept of acidity

The most commonly used concept of acids and bases is that of Brönsted and Lowry, who stated that an acid is a proton donor and a base a proton acceptor. In the following equation, HA is an acid and A⁻ is a base:

In aqueous solutions, protons or H⁺ ions are normally bound by electrostatic interaction to H₂O, resulting in the formation of hydronium ions, designated H₃O⁺. Conventionally, however, the term hydrogen ion and the symbol H⁺ are used to refer to protons in aqueous solutions.

The acidity of a solution refers to the chemical activity of its constituent H⁺ ions. Chemical activity is related to chemical concentration by the activity coefficient, a factor that varies directly with temperature and inversely with the ionic strength of the solution. Physiologic control of body temperature and osmolality, and the dilute nature of body fluids, result in this factor being near unity, and the difference between activity and concentration is negligible in body fluids.

The concentrations of the most important electrolytes in body fluids (e.g., Na⁺, K⁺, Cl⁻, HCO₃⁻) are in the range of milliequivalents per liter, whereas the concentration of H⁺ is in the range of nanoequivalents per liter. That is, hydrogen ions are present at one-millionth the concentration of other electrolytes. What, then, accounts for the emphasis on hydrogen ions in biology and medicine? The answer lies in the fact that hydrogen ions are highly reactive. The proteins of the body have many dissociable groups. These may gain or lose protons as [H⁺] changes, resulting in alterations in charge and molecular configuration that may adversely affect protein structure and function. The [H⁺] of body fluids must be kept constant so that detrimental changes in enzyme function and cellular structure do not occur. The range of [H⁺] compatible with life is 16 to 160 nEq/L.

Concept of pH

The concept of pH was introduced by Sørensen to allow easier notation for the wide range of [H⁺] found in chemical systems. The term pH is defined as the negative base 10 logarithm of the hydrogen ion concentration expressed in equivalents per liter or the base 10 logarithm of the reciprocal of the hydrogen ion concentration:

Thus, at the normal extracellular fluid (ECF) [H⁺] of 40 nEq/L (4 × 10⁻⁸ Eq/L):

There is an inverse relationship between pH and [H⁺]: the greater the [H⁺], the lower the pH. Furthermore, pH and [H⁺] vary not linearly with one another but exponentially as shown in Figure 9-1. The [H⁺] for a given pH within the physiologic range is given in Table 9-1.

Figure 9-1 Exponential relationship between [H⁺] and pH.

(From Madias NE, Cohen JJ. Acid-base chemistry and buffering. In: Cohen JJ, Kassirer JP, editors. Acid-base. Boston: Little, Brown, 1982: 5.)

Table 9-1 Conversions Between pH and [H⁺]

Law of mass action

The law of mass action states that the velocity of a reaction is proportional to the product of the concentrations of the reactants. For the acid just described, there are two opposing reactions:

The velocity of the first reaction can be written:

and the velocity of the second reaction:

At equilibrium, the rates of the two opposing reactions exactly counterbalance one another and the two velocities are equal:

Rearranging and substituting a new constant, K_a, the ionization, or dissociation, constant for the acid HA:

The ionization, or dissociation, constant for an acid is an indication of the strength of that acid. A large value for K_a means that [H⁺] and [A⁻] are much greater than [HA]; that is, the acid is a strong one and is largely dissociated. A small value for K_a means that [H⁺]and [A⁻] are much smaller than [HA]; that is, the acid is a weak one and little of it is dissociated. Hydrochloric acid (HCl) and sulfuric acid (H₂SO₄) are strong acids and dissociate almost completely in aqueous solutions, whereas NH₄⁺ is a weak acid (i.e., it is a strong base) and dissociates to a small extent.

Taking the base 10 logarithm of both sides of the dissociation equilibrium equation yields:

Multiplying by −1 yields:

Applying the concept of pH to both the hydrogen ion concentration and dissociation constant, K_a:

This is the commonly used Henderson-Hasselbalch form of the dissociation equilibrium equation. Occasionally, the term salt or base is substituted for A⁻ and the term acid for HA:

Concept of buffering

A buffer is a compound that can accept or donate protons (hydrogen ions) and minimize a change in pH. A buffer solution consists of a weak acid and its conjugate salt. When a strong acid is added to a buffer solution containing a weaker acid and its salt, the dissociated protons from the strong acid are donated to the salt of the weak acid and the change in pH is minimized.

Consider an aqueous solution with equal amounts of Na₂HPO₄ and NaH₂PO₄. The pK_a for this buffer pair is 6.8:

If the amounts of Na₂HPO₄ and NaH₂PO₄ are equal, their ratio is 1.0:

Consider adding 1 mmol of HCl to this solution. The protons from the HCl are donated to the salt of the buffer pair (Na₂HPO₄), converting it to its conjugate acid (NaH₂PO₄). If 10 mmol of each phosphate salt was present initially, the new ratio of Na₂HPO₄/NaH₂PO₄ would be 9/11 or 0.82 and:

By contrast, an aqueous solution containing 1 mmol/L HCl (10⁻³Eq/L) would have a pH of 3.0.

By solving the dissociation equilibrium equation for [H⁺], the same can be shown:

For the previously described solution of sodium phosphate:

The K_a for this reaction is 1.6 × 10⁻⁷ Eq/L, and if there are equal amounts of the two phosphate salts present ([NaH₂PO₄] = [Na₂HPO₄]):

After addition of 1 mmol of HCl:

By contrast, an aqueous solution containing 1 mmol/L HCl would have [H⁺] = 0.001 mol/L or 1 million nmol/L. Thus, 99.98% of the added hydrogen ions have been buffered by the sodium phosphate solution.

If the amount of strong acid (e.g., HCl) or base (e.g., NaOH) added to a solution of a weak acid and its salt (i.e., a buffer solution) is plotted against pH, the resulting relationship is called a titration or buffer curve (Fig. 9-2). The curve is sigmoidal, and its slope is greatest in the midregion, over which the curve is approximately linear. In the pH range associated with the midregion of the curve, the change in pH is smallest for a given amount of added acid or base and buffer capacity is greatest at the midpoint of the curve. At this point, there are equal amounts of the weak acid and its conjugate salt, and as shown by the Henderson-Hasselbalch equation, pH = pK_a. The region of best buffer capacity extends approximately 1.0 pH unit on either side of the pK_a. Thus, a buffer is most effective within one pH unit of its pK_a. The pK_a values for some important biologic compounds are listed in Table 9-2.

Figure 9-2 Titration curve for an aqueous solution containing a phosphate buffer.

(From Ruch TC, Patton HE (eds): Physiology and biophysics, 20^th edition, Saunders, Philadelphia, 1974.)

Table 9-2 pK_a Values of Biologically Important Compounds*

* Compounds with pK_a values in the range of 6.4-8.4 are most useful as buffers in biologic systems. The pK_a values for the imidazole group of histidine and for α-amino (amino-terminal) amino groups are for those side groups in proteins. The pK_a range for organ phosphates refers to such intracellular compounds as adenosine triphosphate, adenosine diphosphate, and 2,3-diphosphoglycerate.

Isohydric principle

Regardless of the number of buffers present, a solution can have only one [H⁺] and one pH. Using the law of mass action or the Henderson-Hasselbalch equation, the ratio of acid to salt forms of any buffer in the solution can be calculated. This has been called the isohydric principle. The implication of the isohydric principle is that the behavior of any buffer pair in a complex solution can be predicted by knowledge of the dissociation constant and concentrations of any one buffer pair. In clinical practice, the bicarbonate–carbonic acid buffer pair is the one used to monitor acid-base balance in body fluids.

The relative importance of a given buffer in the body is based on its concentration in the relevant body fluid, its pK_a, and the prevailing [H⁺] (40 nmol/L in ECF). The bicarbonate–carbonic acid system is unique among buffers in that carbonic acid is in equilibrium with dissolved CO₂, the concentration of which normally is kept constant by alveolar ventilation.

The bicarbonate–carbonic acid system: physical chemistry

Gaseous CO₂ produced in the tissues is soluble in water, and the concentration of dissolved CO₂ in body fluids is proportional to the partial pressure of CO₂ in the gas phase (Pco₂):

where α is a factor called the solubility coefficient of CO₂. The solubility coefficient of CO₂ has a value of 0.0301 mmol/L/mm in arterial plasma at 37˚ C. Thus:

Dissolved CO₂ combines with water to form carbonic acid:

The uncatalyzed reaction proceeds slowly, but its rate is dramatically increased by the enzyme carbonic anhydrase, which is present in abundance in the body (e.g., red cells, renal tubular cells). In the body, therefore, the hydration of CO₂ to form H₂CO₃ reaches equilibrium almost instantaneously. Normally, the equilibrium is so far to the left that there are approximately 340 molecules of dissolved CO₂ for each molecule of carbonic acid.⁴²

The dissociation of carbonic acid can be expressed using the law of mass action:

K_a for this reaction is 2.72 × 10⁻⁴ mol/L (pK_a=3.57). The ratio of bicarbonate to carbonic acid at the normal [H⁺] of body fluids can be calculated by rearranging this equation:

Thus, at [H⁺] = 40 nmol/L (pH 7.40), there are 6800 bicarbonate ions and 340 molecules of dissolved CO₂ for each molecule of carbonic acid.

The reaction of dissolved CO₂ in aqueous body fluids can be summarized as:

However, the number of carbonic acid molecules is negligible compared with the number of dissolved CO₂ molecules and HCO₃⁻ ions. Therefore, this equation can be simplified as:

The law of mass action for this equilibrium can be expressed as:

The concentration of water in dilute body fluids remains virtually unchanged by this reaction and can be incorporated into K_a to yield another constant, K′_a:

Solving for [H⁺] yields:

In body fluids at 37˚ C, K′_a is approximately equal to 8 × 10⁻⁷ mol/L and p K′_a equals 6.1. An approximate value of 6.1 for this p K′_a is valid at temperatures ranging from 30° to 40° C (86 to 104° F) and pH values ranging from 7.0 to 7.6.³⁷

A formula for [H⁺] in nanomoles per liter or nanoequivalents per liter is obtained by expressing K′_a in nanomoles per liter or nanoequivalents per liter:

Using the solubility coefficient for carbon dioxide yields:

This is the Henderson equation and has been used extensively in the clinical evaluation of acid-base disturbances. It shows clearly that the [H⁺] (and thus pH) of body fluids is determined by the ratio of Pco₂ to HCO₃⁻ concentration. The Henderson-Hasselbalch equation is derived by expressing [H⁺] and K′_a in moles per liter or equivalents per liter and converting the equation to logarithmic form:

Multiplying by −1, we obtain:

Substituting 6.1 for the value of pK′a and applying the solubility coefficient for CO₂, we obtain:

This is the clinically relevant form of the equation and shows that in body fluids, pH is a function of the ratio between HCO₃⁻ concentration and Pco₂.

Body buffers

Body buffers can be divided into bicarbonate, which is the primary buffer system of ECF, and nonbicarbonate buffers (e.g., proteins and inorganic and organic phosphates), which constitute the primary intracellular buffer system. Bone is a prominent source of buffer and can contribute calcium carbonate and, to a lesser extent, calcium phosphate during chronic metabolic acidosis. Bone may even account for up to 40% of the buffering of an acute acid load in the dog.⁹ After administration of NaHCO₃, carbonate can be deposited in bone.

Proteins as buffers

Plasma proteins play a limited role in extracellular buffering, whereas intracellular proteins play an important role in the total buffer response of the body. The buffer effect of proteins is the result of their dissociable side groups. For most proteins, including hemoglobin, the most important of these dissociable groups is the imidazole ring of histidine residues (pK_a, 6.4 to 7.0). Amino-terminal amino groups (pK_a, 7.4 to 7.9) also contribute somewhat to the buffer effect of proteins. Other side groups are relatively unimportant because their pK_a values are either too high or too low to be useful in the normal physiologic range of pH. The pK_a values for the dissociable groups of proteins are listed in Table 9-3.

Table 9-3 pK′a Values for Dissociable Groups Found in Proteins

Dissociable Group (Amino Acid)	pK′a
α-Carboxyl	3.6-3.8
β-Carboxyl (aspartic acid)	≈4.0
γ-Carboxyl (glutamic acid)	≈4.0
Imidazole (histidine)	6.4-7.0
α-Amino	7.4-7.9
Sulfhydryl (cysteine)	≈9.0
ε-Amino (lysine)	9.8-10.6
Phenolic (tyrosine)	8.5-10.9
Guanidino (arginine)	11.9-13.3

From Madias NE, Cohen JJ: Acid-base chemistry and buffering. In Cohen JJ, Kassirer JF, editors: Acid-base, Boston, 1982, Little, Brown & Co., p. 16.

Hemoglobin is responsible for more than 80% of the nonbicarbonate buffering capacity of whole blood, whereas plasma proteins contribute 20%. Of the plasma proteins, albumin is much more important than are the globulins. The buffer value of albumin is 0.12 to 0.14 mmol/g/pH unit, whereas that of globulins is 0 to 0.08 mmol/g/pH unit.^38,69,71 The difference results from a larger number of histidine (Fig. 9-3) residues in albumin.

Figure 9-3 The imidazole group of histidine.

(From Madias NE, Cohen JJ. Acid-base chemistry and buffering. In: Cohen JJ, Kassirer JP, editors. Acid-base. Boston: Little, Brown, 1982: 16.)

The isoelectric point (pI) is the pH at which a substance has no tendency to move in an electric field and thus has no net charge. For proteins, this means that the sum of the charges on the negative side groups (e.g., R–COO−) equals the sum of the charges on the positive side groups (e.g., R–NH₃⁺). At physiologic pH (7.4), plasma proteins are polyanions because their pIs range from 5.1 to 5.7. The net negative charge on plasma proteins in mEq/L can be calculated as:³⁸

where [Pr] is the concentration of plasma proteins in grams per liter, β is the buffer value of plasma proteins in millimoles per gram per pH unit, pH is the ECF pH, and pI is the isoelectric point of plasma proteins. Using this formula, it can be calculated that, at a normal plasma protein concentration of 7 g/dL, average buffer value of 0.1 mmol/g/pH unit, and pI range of 5.1 to 5.7, plasma proteins contribute 12 to 16 mEq/L of negative charge. In dogs, the mean contribution of charge by plasma proteins is approximately 16 mEq/L.^16,76

Physiologic lines of defense in acid-base disturbances

An overview of the body buffer response is provided by contrasting the body’s response to a nonvolatile, or fixed, acid (e.g., HCl) and its response to the volatile acid CO₂. The hydrogen ions from a fixed acid load immediately titrate bicarbonate ions in ECF and then titrate intracellular buffers (e.g., proteins, phosphates). This physicochemical response occurs within minutes and protects ECF pH. Alveolar ventilation is stimulated, and Pco₂ is decreased to below normal. This response, which begins immediately and is complete within hours, minimizes the change in pH because the ratio of HCO₃⁻ to Pco₂ is normalized. Finally, the kidneys regenerate titrated HCO₃⁻, pH increases, alveolar ventilation decreases, and Pco₂ returns to normal. The renal response begins within hours but requires 2 to 5 days to reach maximal effectiveness.

The volatile acid CO₂ cannot be buffered by HCO₃⁻, and the hydrogen ions resulting from the dissociation of carbonic acid must titrate intracellular buffers, such as proteins (especially hemoglobin in red cells) and phosphates. Renal adaptation is characterized by increased HCO₃⁻ reabsorption and net acid excretion, mechanisms that require 2 to 5 days to achieve maximal effectiveness. The buffer response of the body to the primary acid-base disorders is considered in more depth in the chapters on those disorders (see Chapters 10 and 11).

Terminology

The terms acidosis and alkalosis refer to the pathophysiologic processes that cause net accumulation of acid or alkali in the body. The terms acidemia and alkalemia refer specifically to the pH of ECF. In acidemia the ECF pH is lower than normal, and the [H⁺] is higher than normal. In alkalemia the ECF pH is higher than normal, and the [H⁺] is lower than normal. The distinction between these terms is important. For example, a patient with chronic respiratory alkalosis may have a blood pH within the normal range because of effective renal compensation in this setting. Such a patient has alkalosis but does not have alkalemia. Patients with mixed acid-base disturbances can have blood pH values within the normal ranges as a result of the presence of two counterbalancing acid-base disturbances (see the following section on Simple and Mixed Acid-Base Disorders).

Primary acid-base disturbances

Acidosis and alkalosis can each be of metabolic or respiratory origin, and as a result, there are four primary acid-base disturbances: metabolic acidosis, respiratory acidosis, metabolic alkalosis, and respiratory alkalosis. The metabolic disturbances refer to a net excess or deficit of nonvolatile, or fixed, acid, whereas the respiratory disturbances refer to the net excess or deficit of volatile acid (dissolved CO₂).

Metabolic acidosis is characterized by a decreased plasma HCO₃⁻ concentration and decreased pH (increased [H⁺]) caused by either HCO₃⁻ loss or buffering of a noncarbonic (nonvolatile or fixed) acid. Metabolic alkalosis is characterized by an increased plasma HCO₃⁻ concentration and increased pH (decreased [H⁺]), usually caused by a disproportionate loss of chloride ions from the body (i.e., loss of fluid with a chloride concentration greater than that of ECF) or hypoalbuminemia (because albumin is a weak acid). In the absence of volume depletion or renal dysfunction, it is extremely difficult to produce metabolic alkalosis by administration of alkali. Respiratory acidosis is characterized by increased Pco₂ (hypercapnia) caused by alveolar hypoventilation. Respiratory alkalosis is characterized by decreased Pco₂ caused by alveolar hyperventilation (hypocapnia). In one study, metabolic acidosis was the most common acid-base disturbance encountered in dogs.¹⁷

Each primary metabolic or respiratory acid-base disturbance is accompanied by a secondary, or adaptive, change in the opposing component of the system (Table 9-4). The adaptive response involves the component opposite the one disturbed and returns the pH of the system toward but not completely to normal. Overcompensation does not occur. For example, metabolic acidosis is accompanied by a secondary or adaptive respiratory alkalosis. Respiratory acidosis is accompanied by a secondary or adaptive metabolic alkalosis.

Table 9-4 Characteristics of Primary Acid-Base Disturbances

Simple and mixed acid-base disorders

An acid-base disorder is said to be simple if it is limited to the primary disorder and the expected secondary, or adaptive, response. The magnitude of the expected responses is considered in detail in the chapters devoted to the primary acid-base disorders (see Chapters 10 and 11). A mixed acid-base disorder is one that is characterized by the presence of at least two separate primary acid-base abnormalities occurring in the same patient. A mixed acid-base disorder should be suspected whenever the secondary, or adaptive, response exceeds or falls short of that expected. In dogs, for example, the expected response to metabolic acidosis is a 0.7-mm Hg decrease in Pco₂ for each 1.0-mEq/L decrement in plasma HCO₃⁻ concentration caused by metabolic acidosis (see Chapter 10 for more details).

Consider a dog with these normal blood gas values: p. 7.39, [H⁺] = 41 nEq/L, [HCO₃⁻] = 21 mEq/L, and Pco₂ = 36 mm Hg. This dog becomes ill and is observed to have the following blood gas values: p. 7.22, [H⁺] = 60 nEq/L, [HCO₃⁻] = 14 mEq/L, and Pco₂ = 35 mm Hg. If the dog had a simple metabolic acidosis, using the rule of thumb described before, we would have expected the following results: p. 7.27, [H⁺] = 53 nEq/L, [HCO₃⁻] = 14 mEq/L, and Pco₂ = 31 mm Hg. Thus, the dog has a mixed acid-base disorder characterized by both metabolic and respiratory acidoses.

Consider a patient with the following blood gas values: p. 7.40, [H⁺] = 40 nEq/L, [HCO₃⁻] = 31 mEq/L, and Pco₂ = 51 mm Hg. This patient is neither alkalemicnoracidemic because blood pH is 7.40; however, based on the Pco₂ and [HCO₃⁻], the patient is not normal. This patient has a mixed disorder characterized by metabolic alkalosis and respiratory acidosis. The two disorders have counterbalancing effects, resulting in a normal pH. Mixed acid-base disorders are considered in detail in Chapter 12.

Compensatory responses for primary acid-base disturbances

The guidelines for secondary or adaptive responses are listed in Table 9-5 for reference. Note that there are single rules of thumb for each of the metabolic acid-base disorders but two rules of thumb (one each for acute and chronic disorders) for the respiratory acid-base disorders. This is a consequence of the fact that the adaptive respiratory response to metabolic disorders begins immediately and is complete within hours. Conversely, the response to respiratory disorders occurs in two phases. In the first phase, there is immediate titration of predominantly intracellular nonbicarbonate buffers, resulting in an initial change in plasma HCO₃⁻ concentration. The second phase is carried out by the kidneys and is characterized by alterations in net acid excretion and bicarbonate reabsorption. This response begins within hours but takes 2 to 5 days to achieve maximal effectiveness. Thus, there are two expected compensatory responses: acute (<24 hours) and chronic (>48 hours). One caution about rules of thumb is that they define the average response and not 95% confidence intervals. Acid-base maps depict 95% confidence intervals and, although more awkward to use, allow the clinician to consider normal variation in response (Fig. 9-4). Thus, a patient should be considered to have a mixed disorder only when the blood gas value in question deviates considerably from the calculated expected value. Guidelines for establishing a diagnosis of mixed acid-base disorder are discussed in Chapter 12.

Table 9-5 Expected Renal and Respiratory Compensations to Primary Acid-Base Disorders in Dogs

Figure 9-4 Acid-base map or template. The shaded areas exemplify the ranges of Paco₂-bicarbonate relationships characteristic of graded degrees of simple acid-base disorders.

(From Harrington JT, Cohen JJ, Kassirer JP. Introduction to the clinical acid-base disturbances. In: Cohen JJ, Kassirer JP, editors. Acid-base. Boston: Little, Brown, 1982: 379.)

Measurement of blood gases

Most blood gas analyzers measure pH and Pco₂. The HCO₃⁻ concentration is calculated. Total CO₂ content is determined by adding a strong acid to plasma or serum and measuring the amount of CO₂ produced according to the following reaction:

The term total CO₂content refers to the fact that this method includes both dissolved CO₂ and HCO₃⁻ present in the sample. As a result, total CO₂ content is greater than HCO₃⁻ concentration in normal individuals by approximately 1 to 2 mEq/L:

If a sample to be analyzed for total CO₂ content is handled aerobically, the dissolved CO₂ is released to the atmosphere, and the value obtained is approximately equal to the HCO₃⁻ concentration.

Total CO₂ concentrations determined by automated chemistry analysis may differ substantially from those obtained by standard blood gas analysis. In one study of normal dogs and cats, factors implicated in this discrepancy included underfilling of blood collection tubes, delays between sampling and analysis, and freshness of laboratory reagents.³⁰ According to the results of this study, values for total CO₂ obtained by routine blood gas analysis may be up to 5 mmol/L higher than those obtained by automated analysis. Another study comparing total CO₂ measurement by three different methods (radiometer blood gas analyzer, Coulter DACOS analyzer [Beckman Coulter, Fullerton, Calif.], and Kodak Ektachem DTE analyzer [Eastman Kodak, Rochester, N.Y.]) found lower than expected agreement among the different methods of analysis.³¹ In this study, sample storage for 7 hours resulted in a decrease of approximately 2 mmol/L in total CO₂ concentration.

CO₂combining power is the total CO₂ content of a plasma sample that has been equilibrated in vitro at 37° C with CO₂ at a partial pressure of 40 mm Hg. This method overestimates total CO₂ content when the patient’s Pco₂ is less than 40 mm Hg and underestimates total CO₂ content when the patient’s Pco₂ is more than 40 mm Hg. It is no longer commonly used in clinical medicine.

Standard bicarbonate is the concentration of bicarbonate in the plasma of fully oxygenated whole blood after equilibration with CO₂ at a partial pressure of 40 mm Hg at 37° C. The base excess (BE) is the amount of strong acid or base required to titrate 1 L of blood to p. 7.40 at 37° C, while Pco₂ is held constant at 40 mm Hg.^5,6,59 It usually is derived from the Siggaard-Andersen alignment nomogram using measurements of pH, Pco₂, and hematocrit. BE is changed only by nonvolatile, or fixed, acids and thus is considered to reflect metabolic acid-base disturbances. In general, a negative value for BE (i.e., a base deficit) indicates metabolic acidosis, whereas a positive value indicates metabolic alkalosis.

One problem with the concept of standard bicarbonate is the assumption that the CO₂ titration curve of a whole blood sample is similar to that of the intact organism. This is not true because in the isolated blood sample, all of the buffering of the CO₂ equilibrated with the sample is done by the hemoglobin and other nonbicarbonate buffers in the sample, and the HCO₃⁻ generated can be distributed only within that sample. In vivo, however, other intracellular buffers are involved, the HCO₃⁻ produced has a larger volume of distribution, and the observed increase in HCO₃⁻ concentration would be less (Fig. 9-5).¹⁴ Another problem with standard bicarbonate and BE determinations is that abnormalities of these values do not necessarily imply the presence of a primary metabolic acid-base disturbance. Rather, the change in HCO₃⁻ concentration may represent the normal adaptive change resulting from renal compensation for a respiratory acid-base disturbance. Debate continues about whether standard bicarbonate and BE are any more useful than bicarbonate in the evaluation of acid-base disturbances.^56,57 Regardless of the approach used, all of the aids devised to facilitate interpretation of blood gas data are merely graphic representations of the classic Henderson-Hasselbalch equation, and there is no substitute for a thorough understanding of the underlying principles of acid-base physiology.⁶⁷

Figure 9-5 Comparison of the CO₂ titration curves for whole blood and whole body using data derived from the dog.

(Reproduced from Cohen JJ, Brackett NC, Schwartz WB.The nature of the carbon dioxide titration curve in the normal dog. J Clin Invest 1964;43:777, with permission of the American Society for Clinical Investigation.)

Whole-blood buffer base is the sum of the concentrations of all buffer anions contained in whole blood and includes HCO₃⁻, hemoglobin, plasma proteins, phosphates, and any other potential buffer anions.⁶⁰ Its normal value is 40 to 50 mEq/L and is similar to Stewart’s strong ion difference (see Chapter 13).^63,75 Changes in the Pco₂ of the whole blood sample do not change the value of the whole blood buffer base because a change in the nonbicarbonate buffers results in a reciprocal change in HCO₃⁻ concentration. The whole body buffer base decreases with metabolic acidosis and increases with metabolic alkalosis, regardless of changes in Pco₂ in the sample.

The calculations of standard bicarbonate and whole blood buffer base were introduced before the concept of whole body titration was developed²² and represented an attempt to use in vitro titration of whole blood samples to separate the respiratory and metabolic components of acid-base disturbances. These methods do not account for other buffering effects in the body (e.g., intracellular proteins other than hemoglobin, intracellular organic phosphates, bone carbonate).

Normal values

Normal blood gas values for dogs and cats should be established by the laboratory performing the analysis. Extreme care must be taken in obtaining blood samples to establish a normal range because hyperventilation related to fear or pain, increased muscular activity related to struggling, and delays during sample transport and analysis may have effects on the resulting normal range. A review of previously published data provided the following guidelines for normal arterial blood gas values in dogs and cats:²⁵

	Dog	Cat
pH	7.407 (7.351-7.463)	7.386 (7.310-7.462)
Pco2 (mm Hg)	36.8 (30.8-42.8)	31.0 (25.2-36.8)
HCO3− (mEq/L)	22.2 (18.8-25.6)	18.0 (14.4-21.6)
Po2 (mm Hg)	92.1 (80.9-103.3)	106.8 (95.4-118.2)

Studies of normal unanesthetized dogs yielded venous blood gas results as follows: pH 7.397 (7.351 to 7.443), Pco₂ 37.4 (33.6 to 41.2) mm Hg, and HCO₃⁻ 22.5 (20.8 to 24.2) mEq/L.^52,76 In one of these studies, venous Po₂ values were reported to be 52.1 (47.9 to 56.3) mm Hg.⁵² Studies of normal unanesthetized cats indicated venous blood gas values as follows: pH 7.343 (7.277 to 7.409), Pco₂ 38.7 (32.7 to 44.7) mm Hg, and HCO₃⁻ 20.6 (18.0 to 23.2) mEq/L.^11,26,46

When sampling sites were compared using unanesthetized normal dogs, blood gas data from three different venous sites (jugular vein, pulmonary artery, and cephalic vein) were similar, but Pco₂ was higher and pH was lower when venous data were compared with results obtained for the carotid artery.²⁹ The respiratory compensation for metabolic acidosis in these dogs ranged from a 1.1- to 1.3-mm Hg decrement in Pco₂ for each 1-mEq/L decrement in HCO₃⁻, whereas the respiratory compensation for metabolic alkalosis ranged from a 0.4- to 0.6-mm Hg increment in Pco₂ for each 1-mEq/L increment in HCO₃⁻ for arterial, mixed venous, and jugular venous samples. The increment was 1.3 mm Hg per 1-mEq/L increment in HCO₃⁻ for the cephalic samples, which had the highest Pco₂ values, presumably because they were the only samples not collected under free-flowing conditions. Data for the normal dogs in this study are shown in Table 9-6.

Table 9-6 Blood Gas and Acid-Base Measurements (Mean ± Standard Deviation) in Five Normal Unanesthetized Dogs

Aging in humans has been associated with a decrease in Pao₂ and an increase in the alveolar-to-arterial Po₂ gradient, P(A − a)O₂. Mild or no changes in these values were observed in geriatric dogs, and no significant changes in acid-base balance were found in geriatric dogs.^4,32

Anion gap

The major cations of ECF are sodium, potassium, calcium, and magnesium; the major anions are chloride, bicarbonate, plasma proteins, organic acid anions (including lactate), phosphate, and sulfate. The approximate charge contributions of these ions in dogs and cats are listed in Table 9-7. Automated clinical chemistry analyzers provide values for serum sodium, potassium, chloride, and total CO₂ concentrations. Thus, the sum of the concentrations of commonly measured cations exceeds the sum of the concentrations of commonly measured anions, and the difference has been called the anion gap:^18,47

Table 9-7 Approximate Concentrations of Cations and Anions in Plasma in Normal Dogs and Cats (mEq/L)

The serum concentration of potassium varies little, and its charge contribution is small compared with that of sodium. Therefore, the anion gap often is defined as:

From several reported studies, the normal anion gap calculated as (Na⁺ + K⁺) − (Cl⁻+ HCO₃⁻) is approximately 12 to 24 mEq/L in dogs* and 13 to 27 mEq/L in cats.^7,11-13 In one study, the anion gap was significantly increased in aged dogs compared with young dogs (16.7 ± 0.7 vs. 14.3 ± 0.8 mEq/L). The increase in anion gap was attributed to a slight decrease in serum chloride concentration that was balanced by an increase in the net negative charge associated with plasma proteins and phosphate.⁴ In recent studies, the anion gap was calculated to be 18.8 ± 2.9 mEq/L (range, approximately 13 to 25 mEq/L)¹⁶ for dogs and 24.1 ± 3.5 mEq/L (range, approximately 17 to 31 mEq/L) for cats.⁴⁵

In reality, there is no anion gap because the law of electroneutrality must always be satisfied. This can be indicated by including terms for unmeasured cations (UCs) and unmeasured anions (UAs) as follows:

Thus, the anion gap is the difference between UAs and UCs and may be affected by changes in the concentration of either component. However, the magnitude of change in the concentration of any of the UCs (e.g., calcium, magnesium) necessary to cause an appreciable change in the anion gap would probably be incompatible with life.¹⁹ As a result, most discussions of the anion gap focus on changes in UAs.

Normally, plasma proteins contribute the majority of UA charge in mEq/L.³⁵ In humans, albumin contributes 2.0 to 2.8 mEq/L for each gram per deciliter, and globulins contribute 1.3 to 1.9 mEq/L for each gram per deciliter.¹⁹ For each 0.1-U increment in pH, there is an approximate 0.1-mEq/L increase in negative charge on plasma proteins.^19,35,69,71 In dogs, net plasma protein charge at a pH of 7.40 is 16 mEq/L and anion gap is approximately 19 mEq/L, and at a pH of 7.40, the anion gap changes 0.42 mEq/L for every 1 g/L change in albumin and 0.25 mEq/L for every 1 g/L change in total plasma proteins.¹⁶

Increases in anion gap are much more common than decreases, and the concept of anion gap is usually used as an aid in differentiating the causes of metabolic acidosis (see Chapter 10). In organic acidoses (e.g., diabetic ketoacidosis, lactic acidosis), HCO₃⁻ is titrated by H⁺ ions from organic acids. Theoretically, the ECF HCO₃⁻ concentration should decrease in reciprocal fashion with the increase in concentration of organic acid anions, and the serum chloride concentration should not change (so-called normochloremic metabolic acidosis). The anion gap in this setting should increase proportionately. In practice, however, the decrement in HCO₃⁻ concentration rarely equals the increment in anion gap for several reasons. For example, buffers other than HCO₃⁻ are titrated by hydrogen ions from the organic acid; the volume of distribution of the organic anion may differ from that of HCO₃⁻; and the prevailing concentration of the organic anion in ECF is affected by its urinary excretion. Furthermore, the patient’s HCO₃⁻ concentration and anion gap before illness are usually not known, and the changes in HCO₃⁻ concentration and anion gap must by necessity be calculated from available normal values.

The anion gap may be useful in identifying mixed acid-base disturbances. For example, consider a mixed disturbance characterized by metabolic alkalosis and lactic acidosis (e.g., chronic vomiting severe enough to have caused hypotension and impaired tissue perfusion). The pH in such a setting could be normal if HCl loss from the stomach was exactly counterbalanced by accumulation of lactic acid from anaerobic metabolism. A markedly increased anion gap suggests the presence of the complicating organic acidosis. The usefulness of the anion gap in this situation is hampered by the fact that alkalemia itself can cause an increase in the anion gap by several mechanisms.^19,35 Alkalemia results in loss of protons from plasma proteins and an increase in their net negative charge. Hemoconcentration related to volume depletion increases the concentration of plasma proteins and the concentration of their net negative charge. Finally, alkalemia increases lactic acid generation by stimulating phosphofructokinase. The net effect is an increase in the concentration of UAs (lactate and anionic plasma proteins) and an increase in anion gap. The utility of the anion gap concept is considered further in Chapter 12.

Acidosis resulting from administration of NH₄Cl causes a decrease in HCO₃⁻ concentration because hydrogen ions are released during ureagenesis. There is a reciprocal increase in serum chloride concentration, and as a result, there is no change in the anion gap (so-called hyperchloremic metabolic acidosis). Gastrointestinal loss of HCO₃⁻ has the same result because the kidneys conserve NaCl in response to volume depletion. The use of the anion gap in the classification of metabolic acidosis is considered further in Chapter 10.

A decreased anion gap may be observed in immunoglobin G (IgG) multiple myeloma because the pI of IgG paraproteins is greater than 7.4. Hypoalbuminemia or dilution of plasma proteins by crystalloid infusion can decrease the anion gap by decreasing the concentration of the net negative charge associated with plasma proteins. Hypoalbuminemia may be the most common cause of a decreased anion gap, and each 1.0-g/dL decrease in albumin is associated with an approximately 2.4- to 3.0-mEq/L decrease in the anion gap.^19,44

The Nontraditional approach to acid-base evaluation

The traditional approach to acid-base evaluation focuses on the relationship between pH, HCO₃⁻, and Pco₂ as described by the Henderson-Hasselbalch equation. In this approach, pH is shown to be a function of HCO₃⁻ concentration and Pco₂. The Pco₂ is viewed as the respiratory component and is determined by alveolar ventilation, whereas the HCO₃⁻ concentration is considered the metabolic (or nonrespiratory) component and is regulated by the kidneys. This approach may lead to the impression that Pco₂ and HCO₃⁻ are independent variables. In reality, only Pco₂ is independent. When a primary increase in Pco₂ occurs, proteins (notably hemoglobin) buffer the hydrogen ions that are produced by dissociation of H₂CO₃, and the HCO₃⁻ concentration increases secondarily. Furthermore, an understanding of the effects of changes in other electrolytes (e.g., Na⁺, K⁺, Cl⁻) and plasma proteins on acid-base balance is not facilitated by the traditional approach. The nontraditional approach allows the clinician to better understand the complexity of the acid-base disturbances in some patients.

Stewart formulated a model of acid-base chemistry in biologic systems governed by three physical laws: (1) maintenance of electroneutrality; (2) satisfaction of dissociation equilibria for incompletely dissociated solutes; and (3) conservation of mass.^63,64 The equations that satisfy these laws were solved simultaneously to identify variables that control [H⁺]. Independent variables are those that may be altered from outside the system, whereas dependent variables are internal to the system and change only in response to changes in independent variables. Simultaneous solution of Stewart’s equations identified three independent variables: strong ion difference (SID), the total concentration of weak acid (HA + A⁻) or [A_tot], and Pco₂.

The SID changes if the difference between the sum of strong cations and the sum of strong anions changes. Ions are considered strong if they are almost completely dissociated at the pH of body fluids. The strong cations consist of sodium, potassium, calcium, and magnesium. Of these, only Na⁺ is present at high enough concentration in ECF that a change in its concentration is likely to have a substantial effect on SID. The strong anions consist of chloride and several other anions that are not routinely measured clinically, and they collectively are referred to as unmeasured strong anions (e.g., lactate, acetoacetate, β-hydroxybutyrate, sulfate). Chloride and some unmeasured strong anions can be sufficiently altered in certain disease states to have a substantial effect on SID. The average concentrations of all cations and anions in the plasma of normal dogs and cats are presented in Table 9-7.

The weak anions in ECF are HCO₃⁻, plasma proteins, and phosphate. Of these, plasma proteins and phosphate constitute the independent variable A_tot, whereas HCO₃⁻ is a dependent variable. Hypoproteinemia has been shown to be associated with metabolic alkalosis in critically ill human patients in whom a decrease in serum albumin concentration of 1 g/dL caused an increase in standard BE of +3.7 mEq/L.⁴⁴ Serum phosphorus concentration (normally approximately 2 mEq/L) cannot decrease enough to cause alkalosis, but hyperphosphatemia in patients with renal failure can make a substantial contribution to A_tot and metabolic acidosis. The nontraditional approach to acid-base evaluation is considered in detail in Chapter 13.

The Concept of external hydrogen ion balance

External balance for hydrogen ions is maintained by renal excretion of a number of hydrogen ions equal to that consumed in the diet and produced each day by metabolic processes. The majority of hydrogen ions originate from metabolic processes, and little fixed acid originates as such from the diet. A small amount of base is lost each day from the gastrointestinal tract (primarily as organic anions), and this is equivalent to a gain of fixed acid. These processes result in a net daily gain of 50 to 100 mEq of hydrogen ions. Bicarbonate ions that have been titrated by these hydrogen ions must be regenerated. The kidneys are the only regulated route for H⁺ loss from the body.

Metabolic processes that convert cationic compounds to neutral products generate hydrogen ions, whereas those that convert anionic compounds to neutral products consume hydrogen ions.^15,21,73 The main sources of acid are oxidation of the sulfur-containing (e.g., cysteine, methionine) and cationic (e.g., lysine, arginine) amino acids and hydrolysis of organic phosphate diesters, such as phospholipids and nucleic acids. Oxidation of the sulfur-containing amino acids is the major source of acid produced each day:

The main sources of base are metabolism of anionic amino acids (e.g., glutamate, aspartate) and the oxidation or use for gluconeogenesis of other organic anions (e.g., lactate, citrate).

Whole-body regulation of acid-base balance

Acid-base balance requires the cooperation of three major organs: liver, kidneys, and lungs. By the process of alveolar ventilation, the lungs remove the tremendous amount of volatile acid (10,000 to 15,000 mmol CO₂) produced each day by metabolic processes. The liver metabolizes amino acids derived from protein catabolism to glucose or triglyceride and releases NH₄⁺ in the process. When urea is synthesized in the liver from NH₄⁺ and CO₂, H⁺ is produced and HCO₃⁻ is titrated. Consequently, the liver produces much of the fixed or nonvolatile acid that must be excreted each day. The kidneys excrete NH₄⁺ in the urine, thus diverting it from ureagenesis and producing a net gain of HCO₃⁻ and net loss of H⁺.

Renal regulation of acid-base balance

The kidneys maintain normal ECF HCO₃⁻ concentration by reabsorbing virtually all filtered HCO₃⁻ and by regenerating HCO₃⁻ that has been titrated during the daily endogenous production of fixed, or nonvolatile, acid. The latter process is accomplished by excretion of titratable acidity (primarily phosphate salts) and ammonium salts. The term net acid excretion is defined as the sum of titratable acidity and ammonium minus HCO₃⁻ in the urine. Normally, there is a negligible amount of HCO₃⁻ in urine.

All three of the functions described above are accomplished by renal tubular secretion of H⁺. Approximately two thirds of hydrogen ion secretion occurs by means of a luminal Na⁺-H⁺ antiporter (NHE3) and approximately one third by a luminal vacuolar or V-type H⁺-adenosinetriphosphatase (H⁺-ATPase).³³ The H⁺, K⁺-ATPase found in the luminal membranes of the type A intercalated cells of the collecting ducts is quantitatively less important for H⁺ secretion but mediates K⁺ reabsorption. These transport mechanisms depend on the presence of carbonic anhydrase in tubular cells. Carbonic anhydrase II is found in the cytoplasm, where it catalyzes the recombination of CO₂ and H₂O into H₂CO₃, and carbonic anhydrase IV is tethered to the luminal membrane where it facilitates conversion of H₂CO₃ to CO₂ and H₂O in tubular fluid (Fig. 9-6). Of the filtered HCO₃⁻, 80% is reabsorbed in the proximal tubule, 10% in the thick ascending limb of Henle’s loop. 6% in the distal convoluted tubule, and 4% in the collecting duct (Fig. 9-7).

Figure 9-6 A and B, Reabsorption of filtered HCO3⁻ by H⁺ ion secretion in the proximal tubule. CA, Carbonic anhydrase.

(Drawing by Tim Vojt.)

Figure 9-7 Segmental reabsorption of bicarbonate along the nephron. The major portion of filtered bicarbonate is reabsorbed proximally. Fine-tuning of bicarbonate reabsorption occurs in distal nephron segments, including the medullary and cortical collecting ducts, as well as the thick ascending limb of Henle’s loop.

(From Kokko JP, Tannen RL. Fluids and electrolytes, 3rd ed. Philadelphia: WB Saunders, 1996: 208.)

If secreted H⁺ titrates filtered HCO₃⁻, HCO₃⁻ is effectively reabsorbed because one HCO₃⁻ is added to ECF for each filtered HCO₃⁻ titrated by a secreted H⁺ (see Fig. 9-6). This process occurs primarily in the proximal tubules. Net acid excretion and generation of “new” HCO₃⁻ occur whenever secreted H⁺ titrates phosphate in tubular fluid or whenever NH₄⁺ is excreted in the urine with Cl⁻or in exchange for Na⁺ (Figs. 9-8 and 9-9). These processes occur primarily in the distal nephron.

Figure 9-8 A and B, Regeneration of new HCO₃⁻ by titration of phosphate by secreted H⁺ ion in renal tubule. CA, Carbonic anhydrase.

(Drawing by Tim Vojt.)

Figure 9-9 A and B, Regeneration of new HCO₃⁻ by ammonium excretion in renal tubules. CA, Carbonic anhydrase.

(Drawing by Tim Vojt.)

Ammonium excretion

Excretion of ammonium by the kidneys is essential for eliminating the daily fixed acid load and regenerating titrated bicarbonate. Most of the ammonium to be excreted is produced from glutamine in the proximal tubule by action of the enzyme glutaminase^20,21:

(1)

(2)

or

(3)

It can be seen from these reactions that two H⁺ are consumed when the α-ketoglutarate produced from glutamine is either oxidized or converted to glucose. This results in the simultaneous generation of two new bicarbonate ions. If the liver uses an equal number of ammonium ions for urea synthesis, two H⁺ are produced, two HCO₃⁻ are titrated, and there is no net gain of HCO₃⁻. If the NH₄⁺ is excreted in the urine along with Cl⁻or in exchange for Na⁺, however, a net gain of HCO₃⁻ occurs.

The classical theory of ammonium excretion by the kidney suggests that NH₃ diffuses passively through the luminal membrane of the tubular cell into tubular fluid. Hydrogen ions derived from the dissociation of carbonic acid could then combine with NH₃ to form NH₄⁺ because the pK_a for this reaction is 9.2. In the pH range of tubular fluid (6.0 to 7.0), only 0.1% to 1% of this buffer pair would exist as NH₃. Thus, the associated H⁺ is strongly attached to NH₃ by forming NH₄⁺ and does not affect urine pH (Fig. 9-10).

Figure 9-10 Explanation of how NH₄⁺ excretion allows removal of acid without affecting urine pH.

(Drawing by Tim Vojt.)

The classical theory of ammonium excretion by the kidneys was based on diffusion trapping of NH₃ in tubular fluid. According to this theory, the lipid-soluble, nonionized NH₃ diffuses passively into tubular fluid, where it is trapped by combination with H⁺ to form less permeant NH₄⁺. This theory dictates that diffusion equilibrium occurs for NH₃ and that renal tubular cells do not transport NH₄⁺.

Several renal transport mechanisms contribute to the ultimate appearance of NH₄⁺ in urine. Ammonium arises from the metabolism of glutamine and glutamate in the proximal tubules. The ammonium ions that are produced can substitute for H⁺ on the luminal Na⁺/H⁺ antiporter (NHE3) and be secreted into the tubular lumen. When the α-ketoglutarate resulting from the deamination of glutamine and glutamate is metabolized either to CO₂ and H₂O via the Krebs cycle or to glucose via gluconeogenesis, there is a net gain of 2HCO₃⁻, and these “regenerated” bicarbonate ions are returned to the interstitial fluid via an electrogenic basolateral 3HCO₃⁻/Na⁺ cotransporter (NBCe1). The secreted NH₄⁺ travels down the lumen of the descending limb of Henle’s loop and is reabsorbed in the thick ascending limb by substituting for K⁺ in the luminal Na⁺-K⁺-2Cl⁻ cotransporter (NKCC2) in this nephron segment. The lumen-positive transepithelial potential difference in the thick ascending limb also may drive some reabsorption of NH₄⁺ by the paracellular route. The cytoplasm of the tubular cells has a higher pH than the tubular fluid, which allows some NH₃ to form that then can diffuse across the basolateral membranes into the medullary interstitium where it reaches a high concentration. The NH₃ does not escape back across the luminal membranes into the tubular fluid because the luminal membranes of the thick ascending limb are impermeable to NH₃. The interstitial NH₃ then diffuses into segments of the nephron that lack luminal carbonic anhydrase and consequently have the lowest luminal pH (i.e., the S3 segment of the proximal tubule, the cortical collecting duct, and most of the medullary collecting duct). In these segments, the low luminal pH facilitates trapping of NH₃ in the lumen as NH₄⁺. In the S3 segment of the proximal tubule, the NH₄⁺ that is formed is reabsorbed (i.e., recycled) in the thick ascending limb of Henle’s loop. The cell membranes of the collecting ducts are highly permeable to NH₃ (but not NH₄⁺), which facilitates diffusion of interstitial NH₃ across the basolateral membranes into the cells and across the luminal membranes into the tubular fluid where it is trapped as NH₄⁺. The NH₄⁺ trapped in the collecting ducts is excreted in urine and represents a major avenue for elimination of hydrogen ions from the body and for regenerating titrated bicarbonate ions. Rhesus (Rh) glycoproteins RhBG and RhCG are expressed in the distal convoluted tubule, connecting tubule, and collecting duct and function as transporters of ammonia.^33,72,74 RhCG (but not RhBG) expression is increased by chronic acidosis reflecting the important role of RhCG in ammonia secretion and net acid excretion.

The ability of the kidneys to excrete an acid load despite their inability to reduce urine pH below 5.0 (in dogs and cats) is explained by the high pK_a (9.2) of the NH₃-NH₄⁺ buffer pair⁶⁸ (see Fig. 9-10). In the normal animal, 30 to 60 of the 50 to 100 (60%) mEq of the fixed or nonvolatile acid produced each day is excreted in urine as ammonium, either as the chloride salt or in exchange for sodium. The more acidic the urine, the greater the proportion of ammonium that exists as NH₄⁺. The kidneys can also increase their production of ammonium from glutamine during acidosis. At any given urine pH, the rate of ammonium salt excretion is higher in the presence of acidosis,⁶⁸ and renal ammonium excretion can increase fivefold to tenfold (from basal ammonium excretion of 30 to 60 mEq to as much as 300 mEq per day) in response to chronic metabolic acidosis (Fig. 9-11).

Figure 9-11 Effect of acidosis on urinary ammonium excretion.

(From Pitts RF: Fed Proc 7:418, 1948.)

Potassium and acid-base balance

The distribution of potassium ions between intracellular fluid and ECF may be affected by acid-base disorders. When HCl was infused acutely into nephrectomized dogs, approximately 50% of the H⁺ load was buffered intracellularly.^55,65 Intracellular sodium and potassium ions entered ECF in exchange for the H⁺ entering cells, and serum potassium concentration increased. These early animal studies and observations in a small number of human patients¹⁰ led to the prediction that metabolic acidosis would be associated with a 0.6-mEq/L increase in serum potassium concentration for each 0.1-U decrease in pH. A review of animal studies demonstrated that the change in serum potassium concentration observed during acute metabolic acidosis caused by mineral acids (e.g., HCl, NH₄Cl) was variable.³ Furthermore, an increase in serum potassium concentration does not occur in acute metabolic acidosis caused by organic acids (e.g., lactic acid, ketoacids).* Acute infusion of β-hydroxybutyrate in normal dogs caused an increase in insulin in portal venous blood and hypokalemia, presumably as a result of potassium uptake by cells.² Acute infusion of HCl led to hyperkalemia and increased portal vein glucagon concentration.² These acute changes in serum potassium concentration are not the result of changes in renal excretion of potassium.^2,49

The hyperkalemia associated with acute metabolic acidosis caused by mineral acids is transient. In a study of acute and chronic metabolic acidosis induced in dogs by administration of HCl or NH₄Cl, hyperkalemia was observed after acute infusion of HCl, but hypokalemia developed after 3 to 5 days of NH₄Cl administration.⁴¹ The observed hypokalemia was associated with inappropriately high urinary excretion of potassium and increased plasma aldosterone concentration.⁴¹ Similar findings in rats with chronic metabolic acidosis induced by NH₄Cl have been reported.⁵³ Acute metabolic acidosis induced by administration of mineral acid decreases renal proximal tubular reabsorption of sodium, leading to volume contraction and increased distal delivery of sodium. Increased Na⁺-H⁺ and Na⁺-K⁺ exchange then occurs in the distal nephron, mediated by increased distal tubular fluid flow and hyperaldosteronism. These findings suggest that mild hypokalemia and potassium depletion are likely to develop during chronic metabolic acidosis caused by administration of a mineral acid. The observation of hyperkalemia during chronic metabolic acidosis should prompt consideration of impaired renal potassium excretion or some other cause of hyperkalemia (see Chapter 5).