Acid-base assessment and treatment of severe abnormalities (Proceedings)

The pH

The pH is a logarithmic representation of the hydrogen ion activity. It is an overall representation of the net effect of all of the acidotic and alkalotic processes in the body. The normal pH is about 7.40 in the dog and cat; values below about 7.35 represent an acidemia while values above about 7.45 represent an alkalemia. The determinants of pH can be broadly divided into respiratory and metabolic components.

The respiratory contribution to acid-base balance

The respiratory component is carbonic acid which is in equilibrium with carbon dioxide on one side and hydrogen and bicarbonate on the other:

Carbon dioxide + water ↔ carbonic acid ↔ hydrogen + bicarbonate

Both reactions are subject to the laws of mass action and dissociation constants; an increase in one component on one side of each reaction will cause the reaction to "flow" toward the other side. The carbon dioxide and water to carbonic acid reaction is facilitated by carbonic anhydrase.

Chemoreceptors regulate alveolar minute ventilation so as to match carbon dioxide elimination with production so as to maintain a normal PaCO2. The central chemoreceptors are actually responsive to cerebral spinal fluid hydrogen ion but carbon dioxide is very diffusible and changes in PaCO2 cause rapid changes in CSF hydrogen ion concentration. Although carbon dioxide content of blood is a balance between production and elimination, the latter is, by far, the more potent determinant. A halving or doubling of carbon dioxide production would be easily accommodated by a barely noticeable change in minute ventilation. It is appropriate to consider the partial pressure of carbon dioxide in arterial blood (PaCO2 ) is entirely attributed to alveolar minute ventilation. The PaCO2 is used to define the respiratory contribution to the acid-base balance. Increases or decreases in PCO2, via the carbonic acid equilibrium, increase or decrease, respectively, the H+/HCO3 - proportionality.

The carbon dioxide dissolved in the plasma exerts a proportional vapor pressure which is known as the partial pressure of carbon dioxide (PCO2). Central chemoreceptors regulate alveolar minute ventilation so as to maintain an arterial PCO2 (PaCO2) of about 40 mmHg (the normal cat may be somewhat lower). A PaCO2 above about 45 represents an accumulation of carbon dioxide (due to hypoventilation) and causes an acidosis of respiratory origin; a PaCO2 below about 35 (due to hyperventilation) represents a respiratory alkalosis. Venous PCO2 (PvCO2) is normally about 5 mmHg higher than PaCO2 and can normally be used as a surrogate estimate of PaCO2. PvCO2 may, however, be variably higher than PaCO2 in low perfusion states, anemia, and with the use of carbonic anhydrase inhibitors. End-tidal PCO2 is normally about 5 mmHg below PaCO2 and can also be used to estimate PaCO2. End-tidal PCO2 can be variably lower than PaCO2 when there is dead space alveolar ventilation (hypovolemia, large tidal volumes, pulmonary thromboembolism).

The direct effect of changes in PCO2 on hydrogen ion balance

Causes of respiratory alkalosis

• Hypotension

• Fever

• Sepsis

• Excitement

• Exercise

• Pain

• Pulmonary thromboembolism

• Early pulmonary parenchymal disease

• Cytokine release in the systemic inflammatory response syndrome

• Inappropriate ventilator settings

• Compensation for metabolic acidosis

Causes of respiratory acidosis

• Rebreathing (mechanical dead space)

• Neuromuscular disease

• Airway obstruction

• Open pneumothorax or flail chest

• Anterior displacement of the diaphragm by abdominal space filling disorders

• Pleural space filling disorder

• Late pulmonary parenchymal disease

• Compensation for metabolic alkalosis

• Carbohydrate-rich IV feeding solutions in debilitated patients

• Bicarbonate therapy in respiratory compromised patients

• Malignant hyperthermia

The metabolic contribution to acid-base balance

The metabolic component is composed of all of the non-carbonic acids. Glucose is normally metabolized to carbon dioxide and water, but impaired metabolism can result in the accumulation of pyruvic and lactic acids. Fatty acids are normally metabolized to carbon dioxide and water, but aberrant metabolism can result in the accumulation of acetoacetic and hydroxybutyric acids. In oliguric renal failure, the inability of the kidneys to excrete the daily metabolic acid load results in a phosphoric and sulfuric acidosis. The inability of the kidneys to reabsorb bicarbonate in the proximal tubules (proximal renal tubular acidosis) or to excrete hydrogen ion in the collecting tubules (distal tubular acidosis) will result in a metabolic acidosis. Gastric loss of hydrochloric acid causes metabolic alkalosis, while gastrointestinal losses of sodium bicarbonate cause metabolic acidosis. Certain ingestions can cause acidosis: aspirin (acetylsalicylic acid), methanol (formic acid), and ethylene glycol (glycolic and oxalic acid). The administration of salts of certain anions can cause alkalosis when these anions are metabolized: sodium bicarbonate, sodium lactate, sodium citrate, sodium acetate, sodium gluconate, while the administration of solutions of other fluids (saline) can cause acidosis.

The magnitude of the metabolic contribution to the acid-base balance is best described by the calculated base deficit/excess which is the titratable acidity of fully oxygenated whole blood at a PCO2 of 40 mmHg. "Actual" base deficit/excess is just the vascular fluid compartment; it is an in vitro calculation; "standard" base deficit/excess represents the entire extracellular fluid compartment; it is an in vivo calculation. No one tries to guess what is happening with intracellular acid-base balance. The normal base deficit/excess is 0 mEq/L (the normal cat may be somewhat more negative); values below -4 represent metabolic acidosis while values above +4 represent metabolic alkalosis. Metabolic acids invariably titrate bicarbonate downwards and the decrement in the calculated bicarbonate concentration has long been used as a measure of the magnitude of the metabolic contribution to the acid-base balance. This approach has been criticized because bicarbonate concentration is not an independent metabolic parameter; it is also affected by changes in carbon dioxide. This effect, however, is minor, and comparisons of base deficit and the decrement in bicarbonate concentration in clinical patients seldom exceeds 2 mEq/L (although may be greater with extreme hypercapnia). When pH and blood gas analyzers are not available, the total carbon dioxide of the plasma sample can be measured. Most of the carbon dioxide in the plasma is in the form of bicarbonate (there is some, but little, carbonic acid and dissolved carbon dioxide) and therefore a total CO2 measurement is virtually the same as the bicarbonate concentration.

Anion gap

There have been several attempts to identify subcategories of the metabolic component. Anion gap is one example and is calculated as: sodium + potassium - chloride - bicarbonate.

In reality, there is never an anion gap; there is always electroneutrality and cations always equal anions. The anion gap calculation is the difference between the commonly measured cations and the commonly measured anions and represents the unmeasured cations and anions. Calcium, magnesium, and phosphorous (which is in equilibrium with phosphate anion) are commonly measured but are seldom included in the formula; they are small players in the game anyway. The major anion represented by the anion gap as calculated is albumin, which, again, is commonly measured but seldom included, but is a big player in the game. The idea of the anion gap calculation is that some acids, e.g. lactic acid, will titrate the bicarbonate downward (bicarbonate being replaced by lactate which was traditional unmeasured), increasing the apparent anion gap as calculated above. The common presumption is that an increased anion gap represents a metabolic acidosis, but this would only be true if the anion was introduced into the system as an acid (the anion being the conjugate base of the acid). If, however, the anion came as a salt (i.e. following a large dose of sodium acetate/sodium gluconate crystalloid), it would not be associated with an acidosis. The loss of hydrochloric acid in vomiting or sodium bicarbonate in diarrhea will not affect the anion gap because these losses primarily change determinants within the formula. So anion gap divided the causes of metabolic acidosis into two broad categories: high anion gap acidoses (the "addition" acidoses: lactic acidosis, ketoacidosis, renal failure acidosis, intoxicant acidosis) and normal anion gap acidoses (vomiting, diarrhea, renal tubular acidosis). The calculation of anion gap never was very useful because the clinician was still left with very broad categories which did not help very much diagnostically; it was used more to corroborate simple versus complex underlying disease processes. The larger problem associated with the calculation of anion gap is the effect of hypoalbuminemia, which is common in critically ill patients. Hypoalbuminemia proportionately decreases the calculated anion gap. Unless this effect is taken into consideration, the anion gap calculation is virtually uninterpretable.

SID

Strong ion difference (the difference between the strongly dissociated cations and anions) is used to identify acid-base disturbances. Ideally, all cations and anions would be measured but they seldom are, and, in fact, SIDapparent is often calculated simply as sodium - chloride (potassium may or may not be included). Calcium and magnesium are commonly measured but seldom included. From the anion gap calculation and discussion, it can be seen that the major contributors to the difference between sodium and chloride are bicarbonate and albumin. A decrease in SID is proposed to be associated with an acidosis (the traditionalists will recognize this as a decrease in bicarbonate) while an increase is associated with an alkalosis (an increase in bicarbonate). Variations in albumin will, as discussed above, perturbate the calculated SIDapparent and, in fact, have the opposite acid-base effect; SID suffers the same hypoalbuminemia-induced confusions as does anion gap.

SID and Atot

Stewart proposed dividing the metabolic component into SID and Atot (Stewart, 1983). SID being the difference between the strongly dissociated ions: sodium + potassium + calcium + magnesium - chloride - lactate. Bicarbonate, in this proposal, is held to be a dependent variable and is not included in the calculation. The unqualified term "SID" refers to SIDapparent. SIDeffective, the "other" SID, accounts for most of the difference calculated by SIDapparent. SIDeffective is calculated as: bicarbonate + albumin (x 4) + phosphorous (/2). SIDeffective is composed of the incompletely dissociated ions, in other words, Atot. The difference between SIDapparent and SIDeffective represents the contributions of the unmeasured acids or alkalis (strong ion gap).

The approach proposes that changes in SID and Atot cause changes in pH. Experimentally, however, SID can be changed dramatically without changing pH negating any such causal relationship. There is, however, a frequent casual relationship in naturally-occurring diseases wherein changes in SIDapparent do, in fact, track changes in the metabolic component. The calculation of SIDapparent is useful as an approximation of acid-base changes in a patient when pH measurements and base deficit/excess calculations are not available, but no more valuable, for instance, than are total carbon dioxide measurements when a blood gas analyzer is not available. There seems little use for SIDapparent calculations when pH and blood gas measurements are available since these measurements more directly tell the whole story.

The Fencl quantitative approach

Fencl proposed a clinically useful quantitative approach to identifying some of the determinants of the metabolic component (Fencl 1989, Fencl 1993, Leith 1990, deMorais, 1992). There are certain assumptions in these calculations that must be true in order for the calculation to reflect what it is supposed to represent and these will be itemized in the discussion below. Inherent in the following discussion is the absolute necessity to recognize that only a change in hydrogen ion can effect a change in acid-base; changes in sodium, chloride, or any other ion cannot, per se, effect pH. The quantitative approach proposes to calculate the base deficit/excess effect of changes in water concentration (using sodium as the marker), bicarbonate concentration (using chloride as the marker), albumin concentration, lactate concentration, and phosphorous.

Water effect (also called the "sodium effect")

Water is an acid by the equation: H3O+ <=> H+ + H2O. The pH of water is 7.0 at room temperature and 6.8 at body temperature (the dissociation constant is temperature dependent). Compared to the body at a pH of 7.4, water is acid. When water is added to a body and equilibrated to a PCO2 of 40 mmHg, it has a titratable acidity of about 24 mEq/L (it is irrelevant that a crystalloid, in vitro, at room temperature, at a negligible PCO2, has very little titratable acidity [Gaudry, 1972]). When free water is added to a patient (an acidotic effect), the sodium concentration is decreased, and when free water is removed from a patient (an alkalotic effect), the sodium concentration increases (Table 2). The acid-base effect of free water addition or removal is calculated as: Change in sodium from normal x 0.25.

The sodium effect assumes that the underlying cause of the change in sodium concentration is due to free water gain or loss, which is true in most naturally-occurring disease processes. However, hypernatremia caused by the administration of hypertonic sodium solutions (i.e. not caused by a loss of free water) may not be associated with a metabolic alkalosis.

Bicarbonate effect (commonly called the "chloride effect")

Bicarbonate is a very effervescent ion, and can disappear and reappear out of "thin air" in the name of carbon dioxide. It also is a very gregarious ion that is intimately involved with many other buffer systems. It is a little difficult to predict how it is going to react in many situations. In one study, for instance, when water was added to or removed from plasma in vitro in quantities sufficient to cause a 20% dilution or concentration, respectively, the plasma, chloride concentration changed on the average 20% while bicarbonate concentration changed on the average only 6%. Chloride is the marker for the change in bicarbonate that would have occurred if bicarbonate were not so "flighty" and "promiscuous". Many disease processes are associated with the reciprocal handling of chloride and bicarbonate (when chloride goes out, bicarbonate comes in, and vice-versa). Thus, hyperchloremia is often associated with a process in which bicarbonate is lost (diarrhea) or diluted (saline administration) and hypochloremia is often associated with a process in which bicarbonate is retained (gastric vomiting). Terms like hyperchloremic acidosis and hypochloremic alkalosis are unfortunate because they imply that the change in chloride is the cause of the acid-base derangement, however chloride is neither a proton donor nor a proton acceptor and cannot, per se, change acid-base balance. The change in chloride, therefore, bears a casual, but not a causal relationship to the change in the metabolic component. Like the sodium concentration, the chloride concentration is also affected by changes in water concentration. Since the change in water concentration has already been accounted for in the first calculation, it must be adjusted for in the chloride calculation.

Normal chloride concentration + (change in sodium x 0.75) = expected chloride

Expected chloride concentration - measured chloride concentration = base deficit/excess effect.

Lactate effect

Lactate is commonly measured nowadays and elevations in this measurement are commonly attributed to lactic acidosis secondary to inadequate tissue perfusion. The magnitude of the increase in lactate is directly proportional to the base deficit effect. Lactate measuring devices measure L-lactate but not D-lactate. Some ketoacidotic diabetics generate D-lactate which cannot be measured directly and would only show up as an unmeasured anion. Lactate equates to lactic acidosis when it "comes to the party" with hydrogen ion. When it comes as sodium lactate, it is a salt and has no acid-base effect. If a blood sample is contaminated with lactate ringers solution, the lactate measurement will be bogus.

Albumin effect

Albumin is an acid in the form of HPr <=> H+ + Pr-. Albumin has many anionic sites most of which are associated with sodium or calcium (NaPr and CaPr are salts) but some of which are associated with hydrogen (HPr) (Figge, 1992). Hyperalbuminemia is associated with acidosis, while hypoalbuminemia is associated with an alkalosis: Change in albumin concentration x 3.7 = base deficit/excess effect.

Phosphorous effect

Inorganic phosphate is in equilibrium with organic phosphate in the plasma. About 80% of the organic phosphate exists in the form of H2PO4 - while 20% exists as H2PO4 - . The formula accounts for the proper change in units as well as the proportional distribution of the phosphate: 1.8 [0.3229 (normal phosphorous - measured phosphorous{mg/dl})] = base deficit/excess effect.

Simplified formulas and effect

Unmeasured effect

There are always unmeasured acids and bases, i.e. the sum of these individual determinants seldom add to exactly the calculated base deficit/excess. The difference between the sum of these individuals and the base deficit/excess represent the unknown, unmeasured effects.

Formulas for calculated variables effects on SBE

Acids, bases, and buffers

Acids are generally defined as proton donors while bases are defined as proton acceptors. Strong acids and bases are highly dissociated, while weak acids and bases are only slightly dissociated, at body pH.

Little or no free hydrogen ion exists in an aqueous solution; it is virtually all hydrated in the form of H30+ (hydronium ion) and H502+, H703+, etc. When an acid is added to a solvent such as water, the dissociation of the acid transfers the hydrogen ion to the solvent.

The addition of a base would have the opposite effect.

Buffers cushion the effect of an acid or alkali load on the hydrogen ion concentration. Buffers are usually weak acids or weak bases with a pK within 1 pH unit of the body. The pK of a weak acid and its conjugate base is that pH at which unionized portion is equal to the ionized portion (50:50): HA <=> H+ + A- . A weak acid becomes more unionized as the pH of the solvent becomes more acid. A weak base becomes more ionized as the pH of the solvent becomes more acid. Each pH unit change from the pK of the acid or base results in a 10-fold shift in the relative quantity of each side of the equation.

A buffer with a pK of 7.4 would be ideally positioned to buffer either added acids or base. Compunds with a pK more than 2 pH units from the pH of the fluid, cannot be effective buffers.

When an acid is added, the buffer "scavenges" the hydrogen ion so that it does not increase as much as it otherwise would.

When a base is added, the buffer "scavenges" the hydroxide anion ion so that the pH does not increase as much as it otherwise would.

The primary buffers systems in the extracellular fluid compartment are (in decreasing order of quantitative importance) the carbonic acid/bicarbonate system (pK 6.1): HCO3 - + H+ <=> H2CO3 <=> CO2 + H2O; the imidazole ring of the histidine residue of proteins (principally albumin in the plasma and hemoglobin in the red blood cell) (pK 5.6 to 8.4): R-N + H+ <=> R-NH+; and the inorganic and organic phosphate buffer system (pK 6.8): HPO4 - + H+ <=> H2PO4-.

In spite of having a low pK, the carbonic acid-bicarbonate buffer system is very effective for fixed acids or bases because it operates in an open system - the carbon dioxide that is generated is exhaled. The carbonic acid-bicarbonate system has considerable buffering capacity for fixed acids and bases but has virtually no capacity to buffer carbonic acid. Hemoglobin is a very effective buffer of both carbonic and noncarbonic acids.

The primary buffers in the intracellular fluid compartment are the protein and phosphate buffer systems. Bone is an important buffer in chronic metabolic acidosis.

About 52% of a noncarbonic acid load is buffered intracellularly, while 42% will be buffered by extracellularly fluids, and 6% will be buffered by hemoglobin (Baggot, 1992). About 75% of the extracellular buffering is attributed to the carbonic acid-bicarbonate buffer system.

Metabolic acidosis

Metabolic acidosis may be caused by bicarbonate anion loss (e.g., diarrhea or vomiting with duodenal reflux) or hydrogen ion retention. Diarrhea typically has a higher-than-plasma bicarbonate concentration. Most vomition is also associated with the regurgitation of alkaline duodenal fluid and the net loss of a bicarbonate-rich fluid.

Renal tubular acidosis is caused by the renal retention of hydrogen chloride and is heralded by a hyperchloremic, nonanion gap metabolic acidosis which is not due to GI losses. GFR is normal. In contrast, uremic acidosis is caused by a decreased GFR and retention of phosphoric and sulphuric acid, and is associated with a high anion gap, normochloremic metabolic acidosis. Distal tubular acidosis is characterized by a moderate to severe metabolic acidosis with a urine pH > 5.5 (alkaline urine can also be caused by Proteus sp and Staph sp urinary tract infections). The patients may be either hypokalemic is there is a defective H-ATPase pump in or a H+ back leak by the luminal membrane of the intercalated cells in the collecting tubules or hypokalemic if there is defective Na+ reabsorption of K+ excretion by the principle cells. Distal tubular acidosis is often associated with hypercalciuria and hyperphosphaturia. Patients with distal tubular acidosis tolerate an acid load (0.25 g NH4Cl PO q 8 for 1-2 days) poorly (normal patients can excrete this without suffering a metabolic acidosis). Distal tubular acidosis is most often idiopathic or genetic, but may be associated with immune-mediated disease, chronic liver disease, or amphotericin B. Treatment is symptomatic and this disease does not require large amounts of alkalinization therapy to manage.

Proximal tubular acidosis constitutes a reset proximal tubular threshold for bicarbonate reabsorption is characterized by a mild to moderate metabolic acidosis . It may occur alone or as part of a broader proximal tubular dysfunction (Fanconi's syndrome). It is associated with mild hypokalemia, minimal bone resorption, and rare nephrolithiasis. If the plasma bicarbonate is below the renal threshold, the urine will be appropriately acid; if above, inappropriately alkaline (plasma bicarbonate concentrations above the threshold will be rapidly eliminated in the urine). Proximal tubular acidosis is often idiopathic or genetic, and may be caused by multiple myeloma, hypocalcemia, Vitamin D deficiency, aminoglycosides, lead, cadmium, and mercury poisoning, and amyloidosis. Treatment is symptomatic and often require large daily dosages to manage the acidosis.

Intravenous nutrition solutions are high in cationic amino acids (arginine, lysine, histidine) and sulfur-containing amino acids (cysteine and methionine) which generate hydrgoen ion when metabolized. Administration of bicarbonate-free fluids will contribute to a dilutional (decrease in bicarbonate/increase in chloride concentration) acidosis. Ammonium chloride is sometimes administered to acidify the patient and the urine. It is metabolized to hydrochloric acid and urea. Metabolic acidosis may occur as a compensatory response to respiratoy alkalosis

Inorganic acidosis such as the above are not associated with an increase in the calculated anion gap because they directly impact the determinants of that calculation (sodium, chloride, and bicarbonate)(see 4b anion gap). Addition acidoses (organic acidoses) such as lactic acidosis, ketoacidosis, or renal failure acidosis, are associated with an anion gap. For instance, in lactic acidosis: Hlact + NaHCO3 --> NaLact + H2CO3 (Appendix III). Lactate replaces bicarbonate as the sodium salt. If lactate is unmeasured, it becomes an unmeasured (unaccounted) anion in the anion gap equation (see 4b anion gap). Other unmeasured anions could include pyruvate; B-hydroxybutyrate and acetoacetate; and sulfate and phosphate.

Pyruvate is produced in the cytosol by anaerobic glycolysis and then is normally taken up by the mitochondria for conversion to acetyl CoA and incorporated into the tricarboxylic acid cycle. Under anaerobic conditions, when oxidative pathways are disrupted, pyruvate is converted to lactate in order to regenerate NAD+. Hydrolysis of ATP releases H+. Lactic acidosis is most commonly associated with inadequate tissue oxygenation and both skeletal muscle and the gastrointestinal system are the major sources of it. Blood lactate is normally < 1.0 mM/L; it is considered abnormal when above 2.0 mM/L. The magnitude of the elevation, in general, corresponds with the magnitude of the underlying problem. In that severe disease, in general, is associated with a poor prognosis, high blood lactate concentrations may be statistically associated with poor prognosis. Blood lactate does not, however, define the poor outcome, the disease does. If the disease is severe but easily treatable, so will be the high lactate. Lactate measurements should be viewed as a position statement rather than a prognostic indicator. There may also be a poor correlation between blood lactate concentration and the magnitude of the metabolic acidosis: 1) lactate production and H+ production are not linked processes; and 2) there commonly are many other coexisting causes of metabolic acidosis.

There should be no concern regarding the administration of lactated Ringer's solution (LRS) to an animal with a pre-existing lactic acidosis. The lactate in LRS is sodium lactate, not hydrogen lactate, and will not contribute to the lactic acidosis. It may slightly increase measured blood lactate when administered in large volumes to patients with normal blood lactate measurements. When administered to patients with lactic acidosis, the blood lactate levels routinely and rapidly decrease toward normal. The measured lactate concentrations may, however, be iatrogenically increased if the blood sample is contaminated with lactated Ringer's solution.

Insulin deficiency is associated with increased lipolysis and enhanced delivery of free fatty acid to the liver. Simultaneously glucagon diminishes the formation of malonyl CoA, which inhibits carnitine palmitoyl transferase which catalyzes the formation of fatty acyl carnitine (the carrier state for fatty acyl CoA into the mitochondrial membrane). As a result, hepatic mitochondria become overloaded with fatty acyl CoA, a lot of which is converted to acetoacetate (Rose, 1994). Acetoacetate can then be reduced to B-hydroxybutyrate or acetone. Acetoacetate and B-hydroxybutyrate are organic acids (ketones) while acetone is chemically neutral.

Oliguric renal failure results in metabolic acidosis because of the inability of the kidneys to excrete the daily acid load. It is associated with the accumulation of sulphates, phosphates and organic acid anions.

Ethylene glycol is metabolized to glycolic and glyoxylic acid. Methanol is metabolized to formic acid. Salicylate is metabolized to salicylic acid, but the main reason for metabolic acidosis is the uncoupling of oxidative phosphorylation resulting in the production and accumulation of lactic acid, ketones, and other organic acids. Extensive rhabdomyolysis causes the release of intracellular proteins and amino acids.

Causes of metabolic acidosis

With a Normal Anion Gap

1. Gastrointestinal loss of bicarbonate

• Diarrhea

• Vomiting with reflux from the duodenum

2. Renal loss of bicarbonate

• Proximal tubular acidosis

• Carbonic anhydrase inhibitors

3. Renal hydrogen retention

• Distal tubular acidosis

• Hypomineralocorticism

4. Intravenous nutrition

5. Dilutional acidosis (large-volume saline administration)

6. Compensation for respiratory alkalosis

7. Ammonium chloride

With an Elevated Anion Gap

Lactic and pyruvic acidosis

Ketoacidosis

Phosphate and sulphate acidosis (oliguric renal failure)

Ethylene glycol intoxication

Methanol intoxication

Salicylate poisoning

Extensive rhabdomyolysis

Metabolic Alkalosis

Metabolic alkalosis may be caused by hydrogen loss or bicarbonate retention. Hypochloremia potentiates metabolic alkalosis by limiting the quantity of reabsorbable anion in the proximal tubules. A greater proportion of sodium is therefore reabsorbed in exchange for potassium and hydrogen in the collecting ducts (Figure 2). Hypokalemia limits the amount of potassium available for exchange for sodium and consequently increases the proportion of hydrogen that is exchanged. Furosemide inhibits Na+-K+-Cl- reabsorption in the ascending loop of henle and more sodium is delivered to the collecting tubules where the Na+-K+/H+ exchange mechanism predominates. Steroid therapy, specifically hypermineralocorticism, more specifically aldosterone, promotes the Na+-K+/H+ exchange mechanism in the collecting tubules and thereby promotes a metabolic alkalosis (Figure 2). Some penicillins act as nonreabsorbable anions and consequently retain sodium and water within the tubules. More sodium is delivered to the collecting ducts wherein the Na+-K+/H+ exchange mechanism promotes a metabolic alkalosis.

Causes of metabolic alkalosis

1. Gastric losses

• Vomiting due to a pyloric obstruction

• Gastric suctioning

2. Hypochloremia

3. Hypokalemia

4. Furosemide

5. Hypermineralocorticism

6. Citrate and ketone metabolism

7. Carbenicillin and other penicillin derivatives

8. Contraction alkalosis

9. Excessive alkalinization therapy

10. Compensation for respiratory acidosis

Identifying primary problems and secondary compensation

The pH defines whether the patient is alkalemic, acidemic, or within the normal range. The PaCO2 defines the respiratory component: respiratory acidosis, alkalosis, or no significant contribution to the hydrogen ion concentration. The base deficit/excess defines the metabolic component: metabolic acidosis, alkalosis, or no significant contribution. Components that vary in the same direction as the pH disturbance are primarily contributing to the pH disturbance (i.e. an acidemia associated with a respiratory acidosis and a metabolic acidosis). If one of the components varies in the opposite direction, it may be compensatory (i.e. an acidemia associated with a metabolic acidosis and a respiratory alkalosis). It would not, however, be appropriate to presume that the "opposite-directioned" component is compensatory until other causes have been ruled out. A respiratory alkalosis in an acidemic patient with metabolic acidosis could also be caused by hypotension, hyperthermia or pain. If a primary cause cannot be identified, then it may be appropriate to attribute the respiratory alkalosis to compensation.

Expected magnitude of compensation to a primary event

In singular (metabolic or respiratory), chronic disturbances, the opposite component would be expected to compensate. The magnitude of expected compensation has been established in people and dogs (Table 12-11). A value for the "opposite-directioned" component which falls within the framework of this table does not prove that it is compensatory, but is merely supportive of this conclusion. If it does not fall within an appropriate compensating range, it should be concluded that there are other underlying problems.

There are certain assumptions that must be made about the underlying disease and the patient when applying this calculation: 1) that there has been enough time elapsed since the change in the underlying parameter that the animal has had a chance to compensate (8-12 hours for respiratory compensation; 1-3 days for metabolic compensation); 2) that the underlying disease process has been stable during this period of time. While these assumptions may be applicable to singular, chronic medical diseases, they are often blatantly not the case in multiple, acute emergency and unstable critically ill patients. Although it is a popular trend to perform this calculation and analysis very early in the interpretation process, this author would caution against it for our specialty.

Alkalinization therapy

The treatment of metabolic acidosis should be primarily aimed at correction of the underlying disease process and should be the only therapy necessary if the metabolic acidosis and the pH disturbance is mild to moderate and the underlying disease is readily treatable. If, however, the metabolic acidosis is moderate to severe and the underlying disease is difficult to treat, alkalinization therapy may be indicated. In general, alkalinization therapy should be considered if the base deficit is greater than 10 mEq/L or if the bicarbonate concentration is below 14 mEq/L, or if the metabolic acidosis causes the pH to be below 7.2.

Sodium bicarbonate is the most common agent used to treat metabolic acidosis. The mEq of bicarbonate to administer can be calculated. First pick a conservative treatment goal; a base deficit of, say, -5 mEq/L or a bicarbonate of, say, 18 mEq/L. Determine the quantitative difference between the goal value and the measured value; this represents the base deficit or bicarbonate deficit that you wish to treat. Multiply this number by the anticipated volume of distribution of the sodium bicarbonate which is usually considered to be 0.3 x Bwkg.

[mEq of bicarbonate to administer = base/bicarbonate deficit x 0.3 x BWkg]

The factor, (0.3 x Bwkg ), is an overestimate of the extracellular fluid compartment, which is more like 0.24 x Bwkg. This calculation usually, however, underestimates the apparent bicarbonate space which is variable and much larger (Narins, 1994, p761) for several reasons: 1) some of the administered bicarbonate is redistributed to the intracellular fluid compartment (equilibration time 2-4 hours); 2) some of the bicarbonate "buffered" by the acid that is present in the ECF; and 3) to the extent that the underlying disease has not been stabilized, additional acid is being produced between the time it was last measured and the time it is next measured (you are treating a bigger acidosis than you think). It is not surprising that one does not always obtain the desired alkalinization goal and is the reason why one must always re-measure after treatment and administer additional sodium bicarbonate if necessary.

An "off-the-cuff" guideline for dosing sodium bicarbonate is 1 to 5 mEq/kg of body weight, for a mild to severe metabolic acidosis, respectively. This is sufficient to treat a base deficit of approximately 5 to 15 mEq/L, respectively.

These dosages of sodium bicarbonate must not be administered at a rate faster than it can be redistributed from the vascular fluid compartment to the interstitial fluid compartment (20 to 30 minutes). Excessive alkalinization, albeit temporary, of the vascular fluid compartment can cause severe hypotension, restlessness, nausea and vomiting, collapse, and death (in which case it is no longer temporary). The mechanism may be associated with a rapid change in hydrogen ion concentration, a decrease in plasma potassium, or a decrease in plasma ionized calcium. Very small dosages (0.5 mEq/kg) can be rapidly intravenously administered every 5 minutes or so, without problems.

The administration of sodium bicarbonate is also associated with a number of other problems that need to be accommodated. Excessive alkalinization of the patient could be a problem if excessive amounts of sodium bicarbonate are administered; dosages should be calculated carefully and conservatively. The administration of sodium bicarbonate generates carbon dioxide (via carbonic acid) which will result in hypercapnia if the animal is not able to increase its alveolar minute ventilation. Carbon dioxide rapidly diffuses into the intracellular fluid compartment and into the CSF (Pavlin EG, 1975b,c). Once inside, it re-equilibrates across the carbonic acid equilibrium generating an excess of hydrogen ion. Intracellular acidosis may be associated with myocardial and CNS depression (Cingolani, 1975; Berenyi, 1975; Arieff, 1982; Sun, 1996). Animals with normal respiratory responsiveness and capability will eliminate this carbon dioxide within a few breaths, making this a non-issue. Several studies have demonstrated that sodium bicarbonate does not obligatorily increase PCO2 nor cause intracellular or CSF acidosis (Sanders, 1988, Pavlin, 1975a). Conservative dosages and careful monitoring is, however, necessary when sodium bicarbonate is administered to patients with diminished carbon dioxide responsiveness or impaired ventilatory capability.

Sodium bicarbonate administration increases plasma sodium concentration and osmolality. The increase is moderate and similar to that associated with the administration of hypertonic saline, when only one or two dosages are administered. Plasma sodium concentration should, however, be monitored when repeated dosages are administered. Sodium bicarbonate is also very hypertonic (2000 mOsm/kg) and therefore can be irritating to peripheral veins if administered as a continuous infusion. A single dose is unlikely to cause a problem. If the sodium concentration or osmolality of the solution is a concern, it should be diluted with 5% dextrose in water. One part 7.4 % sodium bicarbonate mixed with 6.5 parts 5% dextrose in water will be isonatremic (154 mEq/L) and iso-osmotic (308 mOsm/L). Hypertonic sodium solutions also augment circulating blood volume and cause systemic vasodilation (Kette F, 1991).

These could be desirable or undesirable characteristic, depending upon the patient.

There are alternative alkalinizing agents. Tromethamine THAM)(Abbott) is an organic amine buffer that binds directly with hydrogen ion.

(CH2OH) 3C-NH2 + H+ <=> (CH2OH) 3C-NH3+

Tromethamine thus diminishes carbon dioxide concentration. The unionized portion of tromethamine (about 30% of it) is freely diffusable into the cell, and therefore this agent may be a more effective intracellular buffer than is sodium bicarbonate (Rothe, 82; Robin ED, ). Tromethamine is also an osmotic diuretic. It is a very alkaline solution (a 0.3 M solution has a pH of 10.6) and is irritating to tissues and small veins (it may cause phlebitis and thrombosis if administered undiluted as a continuous infusion into small peripheral veins). It should be administered via large veins or with significant dilution The dosage is calculated by the formula: base/bicarbonate deficit x 0.4 x Bwkg. It is much more expensive than sodium bicarbonate and it is supplied in 500 ml units. Carbicarb is a combination of 0.33 M solution of sodium carbonate and a 0.33 M solution of sodium bicarbonate. This product generates less carbon dioxide than does sodium bicarbonate (Shaprio, 89; Basir, 96; Beech, 94; Leung, 94). There may be some advantages to tromethamine or Carbicarb compared to sodium bicarbonate in certain circumstances, but the problems associated with sodium bicarbonate are not insurmountable and these alternative solutions have not been demonstrated to be superior.

Metabolic alkalosis is usually treated with therapy directed at the underlying disease.

References

Autran deMorais HS, A nontraditional approach to acid-base disorders. In: DiBartola SP, editor, Fluid Therapy in Small Animal Practice, WB Saunders Co; 1992. pp 297-320.

Fencl V, Rossing TH. Acid-base disorders in critical care medicine. Annu Rev Med 1989;40:17-29.

Fencl V, Leith DE. Stewart's quantitative acid-base chemistry: applications in biology and medicine. Respiration Physiology 1993;91:16.

Figge J, Mydosh T, Fencl V. Serum proteins and acid-base equilibria: a follow-up. J Laboratory Clinical Medicine 1992;120:713-19.

Gaudry PL, Duffy C, Bookallil MJ. The pH and titratable acidiy of intravenous infusion solutions. Anaesthesia Intensive Care 1972;1:41-44.

Haskins SC, Rezende ML, Hopper K, The acid-base impact of free water removal from, and addition to, plasma. J Lab Clin Med 2006;147:114-120.

Leith DE. The new acid-base: Power and simplicity. Proc Am Coll Vet Int Med 1990; 449-455.

Stewart PA. Modern quantitative acid-base chemistry. Can J Physiol Pharmacol 1983; 61: 1444-1461