John W. Devlin and Gary R. Matzke
The kidney plays a central role in the regulation of acid–base homeostasis through the excretion or reabsorption of filtered bicarbonate (HCO3–), the excretion of metabolic fixed acids, and generation of new HCO3–.
Arterial blood gases (ABGs), along with serum electrolytes, physical findings, medical and medication history, and the clinical condition of the patient, are the primary tools to determine the cause of an acid–base disorder and to design and monitor a course of therapy.
Metabolic acidosis and metabolic alkalosis are generated by a primary change in the serum bicarbonate concentration. In metabolic acidosis, bicarbonate is lost or a nonvolatile acid is gained, whereas metabolic alkalosis is characterized by a gain in bicarbonate or a loss of nonvolatile acid.
Renal tubular acidosis (RTA) refers to a group of disorders characterized by impaired tubular renal acid handling despite normal or near-normal glomerular filtration rates. These patients often present with hyperchloremic metabolic acidosis.
Respiratory compensation for a primary metabolic acidosis begins rapidly (within 15 to 30 minutes) but does not reach a steady state for 12 to 24 hours after the onset of metabolic acidosis.
Primary therapy of most acid–base disorders must include treatment or elimination of the underlying cause, not just correction of the pH and electrolyte disturbances.
Potassium supplementation is always necessary for patients with chronic metabolic acidosis, as the bicarbonaturia resulting from alkali therapy increases the renal potassium wasting.
Effective treatment of the underlying cause of some organic acidoses (e.g., ketoacidosis) can result in the regeneration of bicarbonate within hours, thus mitigating the need for alkali therapy.
Loss of gastric acid from vomiting or nasogastric suctioning is often responsible for the development of a metabolic alkalosis, characterized by hypochloremia and hyperbicarbonatemia.
Aggressive diuretic therapy can produce a metabolic alkalosis, and the accompanying hypokalemia can be serious.
The patient’s response to volume replacement can be predicted by the urine chloride concentration and permits the differential diagnosis of metabolic alkalosis.
Management of these disorders usually consists of treatment of the underlying cause of mineralocorticoid excess. In patients in whom the mineralocorticoid excess cannot be corrected, chronic pharmacologic therapy can be required.
In most cases of acute metabolic acidosis, such as following cardiopulmonary arrest, sodium bicarbonate therapy is not indicated and can be detrimental. Blood gas analysis should guide therapy.
Acid–base disorders are common and often serious disturbances that can result in significant morbidity and mortality. This chapter reviews the mechanisms responsible for the maintenance of acid–base balance and the laboratory analyses that aid clinicians in their assessment of acid–base disorders. The pathophysiology of the four primary acid–base disturbances is presented, evidence-based therapeutic options are reviewed, and management guidelines to optimize the outcome of patients with one of these disorders are presented. Given that medications are a frequent cause of acid–base abnormalities and that acid–base abnormalities are often preventable, clinicians must anticipate drug-related problems to avoid or minimize the clinical consequences of acid–base disorders, and when necessary, design appropriate treatment regimens.
An acid (in this equation, hydrochloric acid) is a substance that can donate protons (hydrogen ion [H+]):
(acid) HCl → H+ + chloride ion (Cl–)
A base (in this equation, ammonia [NH3]) is a substance that can accept protons (hydrogen ion [H+]):
Ammonia (NH3) + H+ → NH+4 (base)
The acid–base pairs commonly encountered in clinical practice are listed in Table 37-1.
TABLE 37-1 Acid–Base Pairs
The acidity of body fluids is quantified in terms of the hydrogen ion concentration. By convention, the degree of acidity is expressed as pH, or the negative logarithm (base 10) of the hydrogen ion concentration. Thus, hydrogen ion concentration and pH are inversely related. Normally, the pH of blood is maintained at 7.40 ([H+] of 4 × 10–8 M) with a range of 7.35 to 7.45. A pH of less than 6.7 ([H+] of 2 × 10–7 M), representing a fivefold increase in hydrogen ion concentration, or greater than 7.7 ([H+] of 2 × 10–8 M), representing a 50% decrease in hydrogen ion concentration, is considered incompatible with life.
The hydrogen ion concentration in blood may not be indicative of that in other body compartments. For example, the pH within cells, within the cerebrospinal fluid, or on the surface of bone can all be altered without causing an alteration in blood pH.1 Recognizing this caveat, the acid–base status of the body is usually analyzed based on measurement of blood pH. Alterations in blood pH serve as the basis for the diagnosis of acid–base disorders.
Because the dissociation of acid–base pairs is an equilibrium reaction, the relationship between hydrogen ion concentration or pH and the relative concentrations of the acid and base can be described mathematically in terms of the dissociation constant for the acid–base buffer pair. When expressed as a logarithmic relationship, where pK is the negative logarithm of the dissociation constant K, this is known as the Henderson–Hasselbalch equation:
pH = pK + log([base]/[acid])
The ability of a weak acid and its corresponding anion (base) to resist change in the pH of a solution on the addition of a strong acid or base is referred to as buffering. An acid–base pair is most efficient in functioning as a buffer at a pH close to its pK. The principal extracellular buffer is the carbonic acid/bicarbonate (H2CO3/HCO3–) system. Other physiologic buffers include plasma proteins, hemoglobin, and phosphates. Because the isohydric principle requires that all buffer systems remain in chemical equilibrium, the complex buffering of biologic fluids can be analyzed based on a single buffer pair.
The carbonic acid/bicarbonate buffer system plays a unique role in acid–base homeostasis. In addition to being the most abundant extracellular buffer, the components of this buffer pair are under dynamic regulation by the body. In the presence of carbonic anhydrase, carbonic acid, [H2CO3], is in equilibrium with carbon dioxide (CO2) gas. Changes in ventilation that alter the partial pressure of CO2 (PCO2) in the blood regulate the carbonic acid level in the blood. The bicarbonate concentration is independently regulated by the kidney. Because the pK for the carbonic acid/bicarbonate system is 6.1, the relationship between pH, carbonic acid, and bicarbonate concentrations can be described by the Henderson–Hasselbalch equation. The concentration of carbonic acid is directly proportional to the amount of CO2dissolved in blood, which is equal to the product of PCO2 and its solubility in physiologic fluids (PCO2 × 0.03 for PCO2 expressed in mm Hg or PCO2 × 0.226 for PCO2 expressed in kPa). This term can, therefore, be substituted into the equation below in place of [H2CO3].
Thus, hydrogen ion concentration and pH are determined not by the absolute amounts of bicarbonate and PCO2, but by their ratio.1 Under normal physiologic conditions, the kidneys maintain the serum bicarbonate at approximately 24 mEq/L (24 mmol/L), whereas the lungs maintain the PCO2 at approximately 40 mm Hg (5.3 kPa). The normal physiologic pH is thus 7.4:
pH = 6.1 + log[24/(0.03 × 40)] (or pH = 6.1 + log[24/(0.226 × 5.3)])
pH = 6.1 + 1.3 = 7.4
If, in response to an acid load, the serum bicarbonate concentration were to decrease to 12 mEq/L (12 mmol/L), the predicted pH would be:
However, the normal respiratory response to an acid load is hyperventilation. As a result, if the PCO2 decreased to approximately 26 mm Hg (3.5 kPa), the change in pH would be less:
Thus, the physiologic regulation of both PCO2 and [HCO3-] permits the carbonic acid/bicarbonate system to provide more effective buffering of the extracellular fluids (ECFs) than could be achieved on the basis of chemical buffering alone.
REGULATION OF ACID–BASE HOMEOSTASIS
Cellular metabolism results in the production of large quantities of hydrogen that need to be excreted to maintain acid–base balance. In addition, small amounts of acid and alkali are also presented to the body through the diet. The bulk of acid production is in the form of CO2, with the average adult producing approximately 15,000 mmol of CO2 each day from the catabolism of carbohydrate, protein, and fat.2 When respiratory function is normal, the amount of CO2 produced metabolically is equal to the amount lost by respiration, and the blood CO2 concentration remains constant.
Digestion of dietary substances and tissue metabolism also result in the production of nonvolatile acids. These acids are derived primarily from the sulfur-containing amino acids cysteine and methionine, as well as from ingested sulfur. In addition, phosphates are generated from the metabolism of proteins and phospholipids. Neutral substances such as glucose can also be incompletely metabolized to intermediates, such as lactic and pyruvic acid, and fatty acids can be incompletely metabolized to acetoacetic acid and β-hydroxybutyric acid. These dietary and metabolic fixed acids are excreted, primarily by the kidney, to maintain acid–base homeostasis. On average, daily fixed acid excretion is approximately 0.8 mEq/kg per day (0.8 mmol/kg per day).3
Three mechanisms, each of which varies in its onset, collectively maintain acid–base balance: extracellular buffering, ventilatory regulation of carbon dioxide elimination, and renal regulation of hydrogen ion and bicarbonate excretion. Extracellular buffering occurs rapidly and is the body’s first defense against a sudden increase in hydrogen ion concentration. Hyperventilation will then result in a decrease in PCO2, returning blood pH toward normal. Finally, the kidney will excrete the excess hydrogen ion, with the resultant return of acid–base balance to normal over a period of day(s).
The body’s buffering system can be divided into three components: bicarbonate/carbonic acid, proteins, and phosphates. The bicarbonate buffer is the most important of the body’s buffers, because (a) there is more bicarbonate present in the ECF than any other buffer component; (b) the supply of carbon dioxide is unlimited; and (c) the acidity of ECF can be regulated by controlling either the bicarbonate concentration or the PCO2.
Carbonic acid represents the respiratory component of the buffer pair because its concentration is directly proportional to the PCO2, which is determined by ventilation. Bicarbonate represents the metabolic component because the kidney may alter its concentration by reabsorption, generating new bicarbonate, or altering elimination.4 The bicarbonate buffer system easily adapts to changes in acid–base status by alterations in ventilatory elimination of acid (PCO2) and/or renal elimination of base (HCO3-).
The phosphate buffer system consists of serum inorganic phosphate (3.5 to 5 mg/dL [1.13 to 1.62 mmol/L]), intracellular organic phosphate, and calcium phosphate in bone. Extracellular phosphate is present only in low concentrations, so its usefulness as a buffer is limited; however, as an intracellular buffer, phosphate is more useful. Calcium phosphate in bone is relatively inaccessible as a buffer, but prolonged metabolic acidosis will result in the release of phosphate from bone.
Intracellular and extracellular proteins also act as buffering systems. The charged side chains of amino acids provide the buffering action. Because the concentration of protein is much greater intracellularly than extracellularly, protein is much more important as an intracellular buffer.
The second mechanism for maintenance of acid–base homeostasis is control of ventilation. Both the rate and depth of ventilation can be varied to allow for excretion of CO2 generated by diet and tissue metabolism. Medullary chemoreceptors in the brainstem sense changes in PCO2 and in pH and modulate the control of breathing. Increasing minute ventilation (the total amount of air exhaled over a 1-minute period), by increasing respiratory rate and/or tidal volume (the amount of air exhaled in one breath), will increase CO2 excretion and decrease the blood PCO2. Conversely, decreasing minute ventilation decreases CO2 excretion and increases blood PCO2. This system rapidly adjusts within minutes to changes in acid–base balance.2
Because bicarbonate is a small ion, it is freely filtered at the glomerulus. The bicarbonate load delivered to the nephron is approximately 4,500 mEq/day (4,500 mmol/day). To maintain acid–base balance, this entire filtered load must be reabsorbed. Bicarbonate reabsorption occurs primarily in the proximal tubule (Fig. 37-1). In the tubular lumen, filtered bicarbonate combines with hydrogen ion secreted by the apical sodium ion (Na+)–H+-exchanger to form carbonic acid. The carbonic acid is rapidly broken down to CO2 and water by carbonic anhydrase located on the luminal surface of the brush border membrane. The CO2 then diffuses into the proximal tubular cell, where it reforms carbonic acid in the presence of intracellular carbonic anhydrase. The carbonic acid dissociates to form hydrogen ions that can again be secreted into the tubular lumen, and bicarbonate that exits the cell across the basolateral membrane and enters the peritubular capillary.
FIGURE 37-1 Proximal tubular bicarbonate reabsorption. In the tubular lumen, filtered bicarbonate (HCO3–) combines with hydrogen ion (H+) secreted by an apical sodium ion (Na+)–H+ exchanger to form carbonic acid (H2CO3). The carbonic acid is rapidly broken down to carbon dioxide (CO2) and water by carbonic anhydrase located on the luminal surface of the brush border membrane. The CO2 then diffuses into the proximal tubular cell, where it reforms carbonic acid in the presence of intracellular carbonic anhydrase. The carbonic acid dissociates the former hydrogen ion that can again be secreted into the tubular lumen, and bicarbonate that exits the cell across the basolateral membrane and enters the peritubular capillary.
Excretion of metabolic fixed acids and generation of new HCO3- is achieved through renal ammoniagenesis and distal tubular hydrogen ion secretion. Ammoniagenesis plays a critical role in acid–base homeostasis, with ammonium (NH4+) excretion comprising approximately 50% of renal net acid excretion. Ammonium is generated from the deamination of glutamine in the proximal tubule. For each ammonium ion excreted in the urine, one bicarbonate ion is regenerated and returned to the circulation.5
Distal tubular hydrogen ion secretion accounts for the remaining 50% of net acid excretion (Fig. 37-2). In the distal tubular cell, CO2 combines with water in the presence of intracellular carbonic anhydrase to form carbonic acid, which dissociates to H+ and HCO3–. The H+ is actively transported into the tubular lumen by a H+–adenosine triphosphatase (ATPase). The bicarbonate exits the cell across the basolateral membrane and enters the circulation.5
FIGURE 37-2 Collecting duct acid excretion. Hydrogen ion (H+) and bicarbonate (HCO3-) are generated intracellularly from carbon dioxide (CO2) and water, in the presence of intracellular carbonic anhydrase. The hydrogen ion is actively secreted into the tubular lumen by H+–ATPase located in the apical (luminal) membrane. Bicarbonate exits the cell across the basolateral membrane and enters the peritubular capillary. (Cl–, chloride ion; Na+, sodium ion.)
Alterations in blood pH are designated by the suffix “-emia”; acidemia is an arterial blood pH <7.35 and alkalemia is an arterial blood pH >7.45. The pathophysiologic processes that result in alterations in blood pH are designated by the suffix “-osis.” These disturbances are classified as either metabolic or respiratory in origin. In metabolic acid–base disorders, the primary disturbance is in the plasma bicarbonate concentration. Metabolic acidosis is characterized by a decrease in the plasma bicarbonate concentration whereas in metabolic alkalosis the plasma bicarbonate concentration is increased. Respiratory acid–base disorders are caused by alterations in alveolar ventilation that produce corresponding changes in the partial pressure of carbon dioxide from arterial blood (PaCO2). In respiratory acidosis, the PaCO2 is elevated; in respiratory alkalosis, it is decreased. Each disturbance has a compensatory (secondary) response that attempts to correct the HCO3--to-PaCO2 ratio toward normal and mitigate the change in pH (Table 37-2). Although the time course of the respiratory compensatory responses to metabolic disturbances is rapid, the metabolic compensation for respiratory disturbances is slow. As a result, respiratory disturbances are characterized as acute (minutes to hours in duration), indicating that there has not been sufficient time for metabolic compensation, or chronic (days), indicating that sufficient time for metabolic compensation has elapsed.
TABLE 37-2 Interpretation of Simple Acid–Base Disorders
CLINICAL ASSESSMENT OF ACID–BASE STATUS
Blood gases are measured to determine the patient’s oxygenation and acid–base status. Under normal circumstances, there is no clinically significant difference in pH between arterial and mixed venous blood. Arterial samples are designated with the letter “a” (e.g., partial pressure of oxygen from arterial blood [PaO2] and PaCO2), whereas mixed venous samples are labeled with the letter “v” or not labeled (e.g., partial pressure of oxygen from venous blood [PvO2] and partial pressure of carbon dioxide from venous blood [PvCO2]). The normal values for arterial and venous blood gases are shown in Table 37-3. Arterial blood reflects how well the blood is being oxygenated by the lungs (an accurate measurement of PaO2), whereas venous blood reflects how much oxygen tissues are using. Arterial blood rather than venous blood should be used whenever possible because venous blood obtained from an extremity can provide misleading information. If metabolism in the extremity is altered by hypoperfusion, exercise, infection, or some other cause, the difference in the amount of dissolved oxygen between arterial and venous blood can be dramatic. The venous pH and PCO2 during cardiopulmonary resuscitation might be significantly lower and higher, respectively, than the arterial pH and arterial PCO2. This indicates a severe tissue acidosis from CO2 accumulation caused by hypoperfusion.
TABLE 37-3 Normal Blood Gas Values
Analysis of Arterial Blood Gas Data
ABGs provide an assessment of the patient’s acid–base status.6 Low pH values (<7.35) indicate an acidemia, whereas high pH values (>7.45) indicate an alkalemia (Fig. 37-3). In a metabolic acidosis, the pH is decreased in association with a decreased serum bicarbonate concentration and a compensatory decrease in PaCO2. In a respiratory acidosis, the pH is decreased; the PaCO2, however, is elevated. The serum bicarbonate concentration is variable, depending on whether it is an acute disturbance (minimal increase in serum bicarbonate) or a chronic respiratory acidosis (substantial increase in serum bicarbonate). In a metabolic alkalosis, the pH is elevated in association with an increased bicarbonate concentration and a compensatory increase in PaCO2. In a respiratory alkalosis, the pH is also elevated; the PaCO2, however, is decreased. As with respiratory acidosis, the metabolic compensation is variable: a minimal decrease in serum bicarbonate is often noted in acute respiratory alkalosis while a larger decrease in [HCO3-] is common with chronic respiratory alkalosis. Although each measurement has a normal range (see Table 37-3), it is often easiest to consider the midpoint of each range as the normal value. This would correlate to a pH of 7.4, PaCO2 of 40 mm Hg (5.3 kPa), and HCO3- of 24 mEq/L (24 mmol/L). Steps in acid–base interpretation are described in Table 37-4.
FIGURE 37-3 Analysis of arterial blood gases. (HCO3-, bicarbonate; PCO2, partial pressure of carbon dioxide.)
TABLE 37-4 Steps in Acid–Base Diagnosis
When ABGs differ significantly from those expected on the basis of the patient’s clinical condition and previous laboratory determinations, additional venous blood samples should be drawn to assess plasma electrolyte concentrations. The bicarbonate calculated from the patient’s PaCO2 and pH of the blood gas should be compared with the measured total CO2 content (the amount of CO2 gas extractable from plasma, consisting of HCO3-, H2CO3, and PCO2). Ordinarily, the blood gas bicarbonate value is approximately 1 to 2 mEq/L (1 to 2 mmol/L) less than total CO2 content.3 If these values do not correspond, the results should be interpreted with caution because the difference can reflect an error in the blood collection or storage of the sample, or in the calibration of the blood gas analyzer.
METABOLIC ACID–BASE DISORDERS
Metabolic acidosis is characterized by a decrease in pH as the result of a primary decrease in serum bicarbonate concentration.
Metabolic acidosis can result from the buffering (consumption of HCO3–) of an exogenous acid, an organic acid accumulating because of a metabolic disturbance (e.g., lactic acid or ketoacids), or the progressive accumulation of endogenous acids secondary to impaired renal function (e.g., phosphates and sulfates).7 The serum HCO3– can also be decreased as the result of a loss of bicarbonate-rich body fluids (e.g., diarrhea, biliary drainage, or pancreatic fistula) or occur secondary to the rapid administration of non-alkali–containing IV fluids (dilutional acidosis).8
The serum anion gap (SAG), as defined below, can be used to infer whether an organic or mineral acidosis is present.
To maintain electroneutrality, the total concentration of cations in the serum must equal the total concentration of anions.
The cation concentration is equal to the sodium concentration plus that of “unmeasured” cations (UCs), predominantly magnesium, calcium, and potassium. The anion concentration is equal to the concentrations of chloride, bicarbonate, and “unmeasured” anions (UAs), including proteins, sulfates, phosphates, and organic anions. Therefore, as the result of the combination of the two equations above, the SAG can be expressed as:
The normal SAG is approximately 9 mEq/L (9 mmol/L), with a range of 3 to 11 mEq/L (3 to 11 mmol/L). This value is lower than the value of 12 mEq/L (12 mmol/L) cited in the literature in the past because of changes in the instrumentation for measurement of serum electrolytes.7 Increases in the anion gap (AG) to values in excess of 17 to 20 mEq/L (17 to 20 mmol/L) are indicative of the accumulation of unmeasured anions in ECF.
These unmeasured anions are generated as the result of the consumption of HCO3– by endogenous organic acids such as lactic acid, acetoacetic acid, or β-hydroxybutyric acid or from the ingestion of toxins such as methanol or ethylene glycol. The degree of elevation in the SAG is dependent on the clearance of the anion, as well as the multiple factors that influence HCO3– concentrations. Thus, the SAG is a relative rather than an absolute indication of the cause of metabolic acidosis. The SAG can also be elevated in the metabolic acidosis because of renal failure, as the result of the accumulation of various organic anions, phosphates, and sulfates.
In hyperchloremic metabolic acidosis, bicarbonate losses from the ECF are replaced by chloride, and the SAG remains normal. This decrease in bicarbonate may be due to GI tract losses, dilution of bicarbonate in the ECF as the result of the addition of sodium chloride solutions or chloride-containing acids. Common causes of metabolic acidosis with an increased or a normal SAG are listed in Table 37-5.
TABLE 37-5 Common Causes of Metabolic Acidosis
Hyperchloremic Metabolic Acidosis
Hyperchloremic metabolic acidosis can result from increased GI bicarbonate loss, renal bicarbonate wasting, impaired renal acid excretion, or exogenous acid gain. GI disorders such as diarrhea, biliary, or pancreatic drainage through either a surgical drain or fistula can result in the loss of large volumes of bicarbonate-containing fluids. Severe diarrhea, the most common cause of hyperchloremic metabolic acidosis, can lead to a daily loss of 5 to 10 L of fluid containing 100 to 140 mEq/L (100 to 140 mmol/L) of sodium, 20 to 40 mEq/L (20 to 40 mmol/L) of potassium, 80 to 100 mEq/L (80 to 100 mmol/L) of chloride, and 30 to 50 mEq/L (30 to 50 mmol/L) of bicarbonate.4 Patients who have undergone ureteral diversion into the sigmoid colon or isolated ileal loop can also develop a hyperchloremic metabolic acidosis. This is the result of a net loss of bicarbonate given that chloride is reabsorbed and bicarbonate is secreted by GI epithelial cells in the presence of the urine that is retained in the colon or bowel loop.
Hyperchloremic metabolic acidosis caused by renal bicarbonate wasting is the defining disturbance in proximal RTA and is a complication of therapy with carbonic anhydrase inhibitors, particularly when they are administered for more than 24 to 48 hours. During the treatment of diabetic ketoacidosis, renal loss of β-hydroxybutyrate and acetoacetate, which would otherwise be metabolized to yield bicarbonate, can contribute to the development of hyperchloremic metabolic acidosis. Impaired renal acid excretion that occurs as a result of distal tubular dysfunction in patients with distal RTAs can also occur in patients with moderate to severe renal insufficiency from other causes. The metabolic acidosis of renal insufficiency is initially hyperchloremic but can progress to an anion-gap acidosis as the renal insufficiency worsens and sulfates, phosphates, and other anions accumulate. Hyperchloremic metabolic acidosis can also result from the exogenous administration of acid (hydrochloric acid, ammonium chloride) or the unbuffered administration of acid salts from the amino acids in total parenteral nutrition fluids.
Renal Tubular Acidosis
Renal tubular disorders can involve the proximal tubule, with a resultant failure to reabsorb filtered bicarbonate, or affect acid excretion in the distal tubule. The distal RTAs are the most common, and are all characterized by impaired net acid excretion. The distal RTAs are subdivided into those that are associated with hypokalemia (type I) and those associated with hyperkalemia (type IV). Patients with classic distal (type I) RTA have impaired hydrogen ion secretion and are unable to excrete the daily acid load necessary to maintain acid–base balance.4 These patients are unable to maximally acidify their urine (i.e., attain urine pH <5.5), even in the face of an acid challenge. Type I RTA may be the result of a primary tubular defect or develop secondary to a wide variety of disorders including hypercalcemia, multiple myeloma, systemic lupus erythematosus, Sjögren’s syndrome, sickle-cell disease, and renal transplant rejection, or following the administration of amphotericin B or ingestion of toluene. The primary form of this disorder usually occurs in children and can result in severe acidosis, slowed growth, nephrocalcinosis, and kidney stones.7,9 In adults, clinical complications include osteomalacia, nephrocalcinosis, and recurrent kidney stones. The hypokalemia associated with classic distal (type I) RTA results from secondary hypoaldosteronism associated with volume depletion. The renal potassium wasting decreases considerably if bicarbonate therapy is administered.
The hyperkalemic distal (type IV) RTAs are a heterogeneous group of disorders characterized by hypoaldosteronism or generalized distal tubule defects. The most common form of type IV RTA is hyporeninemic hypoaldosteronism. This syndrome is most commonly associated with mild renal insufficiency caused by diabetic nephropathy, but can also be seen in a variety of other disorders, including chronic interstitial nephritis, sickle-cell disease, human immunodeficiency virus (HIV) nephropathy, and obstructive uropathy. The clinical presentation of this syndrome is often exacerbated by pharmacologic therapy with agents that can interfere with the renin–angiotensin–aldosterone axis, such as β-adrenergic blockers, angiotensin-converting enzyme (ACE) inhibitors, angiotensin receptor blockers, and nonsteroidal antiinflammatory drugs (NSAIDs). Heparin can induce the syndrome by inhibiting adrenal aldosterone biosynthesis. Patients with this form of RTA are able to maximally acidify their urine (urine pH <5.5).9 The primary defect in acid excretion is impaired ammoniagenesis caused by mild renal insufficiency. Hyperaldosteronism predisposes to the development of hyperkalemia, which results in further impairment of ammoniagenesis. Treatment to control the hyperkalemia is usually sufficient to reverse the metabolic acidosis, and mineralocorticoid replacement is frequently unnecessary.
Hyperkalemic distal (type IV) RTA resulting from generalized distal tubule defects is less common than hyporeninemic hypoaldosteronism but is more common than classic distal (type I) RTA. Patients with this defect have impaired tubular potassium secretion in addition to impaired urinary acidification (urine pH >5.5, despite acidemia or acid loading). Urinary obstruction is the most frequent cause of this disorder, which can also be associated with sickle-cell nephropathy, systemic lupus erythematosus, HIV nephropathy, analgesic abuse nephropathy, amyloidosis, renal transplant rejection, and chronic cyclosporine nephrotoxicity.
Proximal (type II) RTA is characterized by defects in proximal tubular reabsorption of bicarbonate. Normally, more than 85% of filtered bicarbonate is reabsorbed in the proximal tubule. Defects in proximal tubular bicarbonate reabsorption result in increased delivery of bicarbonate to the distal nephron, which has a limited capacity for bicarbonate reabsorption. As a result, at a normal serum bicarbonate concentration, the filtered bicarbonate load is incompletely reabsorbed, and is lost in the urine. As the serum bicarbonate concentration decreases, the filtered load of bicarbonate is proportionately decreased. A new equilibrium is established in which the kidney is able to reabsorb the filtered bicarbonate load, albeit at a reduced serum bicarbonate concentration. Thus, patients with proximal RTA present with a chronic, nonprogressive hyperchloremic metabolic acidosis. These patients are able to acidify their urine in response to an acid load, but develop bicarbonaturia at a reduced serum bicarbonate concentration following bicarbonate loading. The impaired bicarbonate reabsorption results in salt wasting and secondary hyperaldosteronism. Hypokalemia, which can be severe, usually develops as a result of the hyperaldosteronism and bicarbonaturia.4,7 Unlike patients with classic distal (type I) RTA, the hyperkalemia if present in proximal RTA is exacerbated by alkali replacement. Proximal RTA can develop as an isolated defect, or it can be associated with generalized proximal tubular dysfunction (Fanconi’s syndrome), with impaired proximal tubular glucose, phosphate, and amino acid reabsorption. Proximal RTA usually presents as an acquired disorder, secondary to a variety of diseases (amyloidosis, multiple myeloma, or nephrotic syndrome) or exposure to toxins (lead, cadmium, mercury, or outdated tetracyclines). Pharmacologic therapy with carbonic anhydrase inhibitors produces an iatrogenic form of proximal RTA.
Elevated Anion Gap Metabolic Acidosis
Metabolic acidosis with an increased SAG commonly results from increased endogenous organic acid production.7 In lactic acidosis, lactic acid accumulates as a by-product of anaerobic metabolism. Accumulation of the ketoacids β-hydroxybutyric acid and acetoacetic acid defines the ketoacidosis of uncontrolled diabetes mellitus, alcohol intoxication, and starvation (see Table 37-5). In advanced renal failure, accumulation of phosphate, sulfate, and organic anions is responsible for the increased SAG, which is usually less than 24 mEq/L (24 mmol/L).7 The severe metabolic acidosis seen in myoglobinuric acute renal failure caused by rhabdomyolysis may be caused by the metabolism of large amounts of sulfur-containing amino acids released from myoglobin.
The presence of mild elevations in the SAG cannot be automatically attributed to the presence of a high SAG metabolic acidosis. Elevations in the SAG are commonly seen in hospitalized patients, especially those who are critically ill.7,10 A variety of factors can contribute to this nonspecific elevation in the SAG, including the presence of alkalemia, which increases the anionic charge of albumin and other plasma proteins. The usefulness of the SAG as a marker of acid–base status is dependent on proper interpretation of a patient’s clinical status.5,11 Despite these limitations, when the SAG exceeds 20 to 25 mEq/L (20 to 25 mmol/L) a significant organic acidosis is likely to be present.
High anion gap metabolic acidosis can develop in many clinical settings, including uncontrolled diabetes mellitus (see Chap. 57), alcohol intoxication (see Chaps. 24 and 49), and starvation (see Chap. 47). Toxic ingestions of methanol and ethylene glycol are also associated with high anion gap metabolic acidosis and can be differentiated from other causes of SAG because of the presence of an elevated osmolar gap. The mechanisms responsible for the development of acidosis in these settings are diverse.7,12
Lactic Acidosis Lactic acidosis is one of the most common causes of high SAG metabolic acidosis and can impact approximately 1% of hospitalized patients. Lactic acid is the end product of anaerobic metabolism of glucose (glycolysis).7 In normal individuals, lactic acid derived from pyruvate enters the circulation in small amounts and is promptly removed by the liver. In the liver, and to a lesser extent in the kidney, lactic acid is reoxidized to pyruvic acid, which is then metabolized to CO2 and H2O. The normal plasma lactate concentration in healthy subjects is approximately 1 mEq/L (1 mmol/L).3,7,12 The diagnosis of lactic acidosis should be considered in all patients with metabolic acidosis associated with an increased SAG. Lactic acidosis is considered to be present when lactate concentrations exceed 4 to 5 mEq/L (4 to 5 mmol/L) in an acidemic patient.
Classically, lactic acidosis has been differentiated into disorders associated with tissue hypoxia (type A lactic acidosis) and disorders associated with deranged oxidative metabolism (type B lactic acidosis), although the distinction between them is blurred (Table 37-6).7,11,12 The etiologies of lactic acidosis can also be categorized on the basis of changes in lactate production and/or utilization.7,12 Metabolic disturbances can result in increased tissue pyruvate production or impaired utilization, with proportional increases in lactate concentrations. Increased lactate production is more commonly associated with alterations in tissue redox state, resulting in preferential conversion of pyruvate to lactate. During anaerobic metabolism, reduced nicotinamide adenine dinucleotide accumulates, driving the conversion of pyruvate to lactate and increasing the lactate-to-pyruvate ratio. States of enhanced metabolic activity (e.g., grand mal seizures, strenuous exercise, or hyperthermia), decreased tissue oxygen delivery (e.g., severe anemia, hypoxia, circulatory shock, or carbon monoxide poisoning), or impaired oxygen utilization (e.g., cyanide toxicity) all are associated with lactic acidosis. Impaired hepatic clearance of lactate, as seen in hypoperfusion states, liver failure, and alcohol intoxication, can also result in lactic acidosis.
TABLE 37-6 Causes of Lactic Acidosis
Cardiovascular and septic shock, with resultant tissue hypoperfusion, are the most common causes of lactic acidosis. Poor tissue perfusion and hypoxia influence enzymatic pyruvate and lactate metabolism to stimulate anaerobic glycolysis and to decrease lactate utilization. This leads to hyperlactatemia and lactic acidosis. The mortality rate of this type of lactic acidosis can be as high as 80% and correlates with the degree of hyperlactatemia.
Lactic acidosis associated with liver disease, toxins, and congenital enzyme deficiency can be caused by deranged oxidative metabolism or impaired lactate clearance.4,7,12 The exact role of diabetes mellitus in the induction of lactic acidosis is not clear. It may involve a decrease in pyruvate dehydrogenase activity, the enzyme responsible for pyruvate metabolism. Lactic acidosis in neoplastic disease is uncommon and reported mostly in patients with myeloproliferative disorders. Leukocytes and neoplastic cells in general have high rates of glycolysis. In the case of a large tumor or tightly packed bone marrow, oxygenation can be decreased, favoring the accumulation of lactate. Lactic acidosis has been reported in patients with massive liver tumors, and it has been postulated that the liver uptake of lactate is decreased in these patients. Lactic acidosis associated with seizures is usually transient and occurs because of excessive muscle activity.13
A number of medications can cause lactic acidosis.13–23 Two of the most common medications associated with the development of lactic acidosis are nucleoside-analog reverse transcriptase inhibitors (NRTIs) (3.9 cases per 1,000 person-years) and metformin (0.03 cases per 1,000 person-years).15–17 The proposed mechanism of NRTI-induced lactic acidosis is the inhibition of the enzyme DNA polymerase gamma that is responsible for mitochondrial DNA synthesis.15 Disruption of this enzyme can inhibit the transport of lactate into the mitochondria, leading to an accumulation in the cytoplasm. Stavudine is the NRTI most frequently associated with lactic acidosis; however, the combination of stavudine and didanosine confers the highest risk.
The primary suspected mechanism for metformin-induced lactic acidosis is inhibition of liver gluconeogenesis as the result of its inhibitory effects on pyruvate carboxylase which is necessary for the conversion of pyruvate to glucose.16,17 Other possible pathways for metformin-associated lactic acidosis include a decrease in both hepatic intracellular pH and cardiac output and an increase in lactate production in the gut and increased renal loss of bicarbonate.13 Risk factors for metformin-induced lactic acidosis include renal insufficiency, liver disease, dehydration, advanced age, alcohol consumption, and supratherapeutic dosing. Metformin should be discontinued during periods of tissue hypoxia (e.g., myocardial infarction, sepsis), for 3 days after contrast media has been administered or 2 days before general anesthesia administration. In the latter two cases, metformin should only be reinstituted when the patient’s renal function is stable.
Propylene glycol is commonly used as a solubilizing agent in IV drug preparations (e.g., lorazepam, pentobarbital) and is predominantly metabolized to lactic acid via the hepatic enzyme alcohol dehydrogenase. The administration of large doses of propylene glycol, particularly to patients with renal or liver insufficiency, can lead to a lactic acidosis with an osmolar gap and thus serial measurement of the osmolar gap can be used to detect propylene glycol accumulation.18,19
Reports of the association between propofol and lactic acidosis were initially described in children.20 This association is now recognized in adults and has come to be known as the propofol-related infusion syndrome. In addition to lactic acidosis, cardiac failure, rhabdomyolysis, and renal failure have been observed primarily because of uncoupling of oxidative phosphorylation and impaired oxidation of free fatty acids. This syndrome is most frequently seen in patients receiving propofol at high doses (>5 mg/kg/h) for more than 2 days.
Chronic metabolic acidosis is usually not associated with severe acidemia and is relatively asymptomatic. The major manifestations are in the bones, where chronic acidemia causes bone demineralization with the development of rickets in children and osteomalacia and osteopenia in adults.26 In infants and children, chronic metabolic acidosis is associated with growth failure and short stature and can be associated with nonspecific symptoms including anorexia, nausea, weight loss, and muscle weakness.
Severe metabolic acidosis is usually associated with acute processes. The manifestations of severe acidemia (pH <7.20) involve the cardiovascular, respiratory, and CNS. Hyperventilation is often the first sign of metabolic acidosis. At a pH of 7.2, pulmonary ventilation increases approximately fourfold, and an eightfold increase has been noted at a pH of 7.24,25 Respiratory compensation can occur as Kussmaul respirations—the deep, rapid respirations seen commonly in patients with diabetic ketoacidosis. In extremely severe acidosis (pH <6.8), CNS function is disrupted to such a degree that the respiratory center is depressed.
CNS depression correlates more closely with spinal fluid pH than with blood pH. For this reason, neurologic symptoms tend to occur more frequently and to a greater degree in patients with respiratory acidosis because the CO2accumulated in the respiratory form readily crosses the blood–brain barrier to cause acidosis in the CNS.1 Because of the slow penetration of administered bicarbonate into the CNS, the CNS pH fails to normalize as rapidly as blood pH. Therefore patients continue to hyperventilate because of sustained CNS acidity, and severe respiratory alkalosis can occur. Sustained lowering of the PaCO2 within 12 to 36 hours is to be anticipated during the correction of any metabolic acidosis.1
Systemic acidosis can cause peripheral arteriolar dilatation, characterized by flushing, a rapid heart rate, and wide pulse pressure. Initially, cardiac output can be increased, but as acidosis becomes more severe, myocardial contractility becomes impaired, and cardiac output decreases. The effects of vagal stimulation are also enhanced at pH levels lower than 7.1, probably as a consequence of inhibition of acetylcholinesterase. This increases the danger of vagally mediated bradycardia and heart block during acidosis.
GI symptoms of metabolic acidosis include loss of appetite, nausea, and vomiting. Severe acidosis (pH <7.1) interferes with carbohydrate metabolism and insulin utilization, and results in hyperglycemia. Metabolic acidosis alters potassium homeostasis and contributes to the development of hyperkalemia. The magnitude of the effect on serum potassium depends on the type of acidosis: Acidosis caused by mineral acids (e.g., hydrochloric acid) is associated with a greater change in potassium levels than acidosis caused by organic acids (e.g., lactic acidosis), in which the increase in potassium attributable to the acidosis per se is minimal.
The patient’s primary means to compensate for metabolic acidosis is to increase carbon dioxide excretion by increasing the respiratory rate. This results in a decrease in PaCO2. This ventilatory compensation results from stimulation of the respiratory center by changes in cerebral bicarbonate concentration and pH.1 For every 1-mEq/L (1 mmol/L) decrease in bicarbonate concentration below the average of 24, the PaCO2 decreases by approximately 1 to 1.5 mm Hg (0.13 to 0.20 kPa) from the normal value of 40 (5.3 kPa) (Table 37-7).
TABLE 37-7 Guidelines for Initial Interpretation of Acid–Base Disorders
CLINICAL PRESENTATION Metabolic Acidosis
• The patient is usually relatively asymptomatic if the acidosis is acute and mild. In those with severe acidemia (pH <7.15–7.20), the cardiovascular, respiratory and central nervous systems can be affected
• The patient may complain of loss of appetite, nausea, and vomiting
• Cardiac: Flushing, a rapid heart rate, wide pulse pressure, and an increase in cardiac output can be seen initially. This can be followed by a reduction in cardiac output, blood pressure, and liver and kidney blood flow
• Cerebral: Obtundation or coma
• Metabolic: Insulin resistance; increased protein degradation; increased metabolic demands
• GI: Nausea, vomiting, loss of appetite
• Respiratory: Dyspnea, hyperventilation with deep, rapid respirations is seen in those with severe acidosis
• Chronic acidemia causes bone demineralization with the development of rickets in children and osteomalacia and osteopenia in adults
• Serum CO2 is low. Hyperglycemia and hyperkalemia are common. Patients with a pH of <7.2 are deemed to have a severe acidosis
The anticipated PaCO2 associated with a given bicarbonate concentration for patients with uncomplicated metabolic acidosis can be calculated as25:
For example, 95% of patients with a plasma bicarbonate of 16 mEq/L (16 mmol/L) should have an arterial PCO2 of 30 to 34 mm Hg (4.0 to 4.5 kPa). An observed arterial PCO2 within this range is consistent with physiologic respiratory compensation for a metabolic acidosis and suggests that there is no respiratory disturbance. In contrast, if the PCO2 is less than 30 mm Hg (4.0 kPa), a superimposed respiratory alkalosis can be present, whereas if the PCO2is greater than 34 mm Hg (4.5 kPa), a superimposed respiratory acidosis is likely present.
Chronic Metabolic Acidosis
Asymptomatic patients with mild to moderate degrees of acidemia (plasma bicarbonate of 12 to 20 mEq/L ([12 to 20 mmol/L]; pH 7.2–7.4) do not require emergent therapy. They can usually be managed with gradual correction of the acidemia, over a period of days to weeks, using oral sodium bicarbonate or other alkali preparations (Table 37-8). In all forms of chronic metabolic acidosis, primary therapy should be directed at treating the underlying disease state. GI pathology should be treated to reduce ongoing bicarbonate losses, and factors that exacerbate RTA should be treated. If acidemia persists, alkali therapy should be instituted with the goal of normalization of blood pH. The loading dose (LD) of alkali to initially correct the acidemia can be calculated as follows:
where VD is the volume of distribution of bicarbonate.7
For a 60-kg patient with a serum bicarbonate of 15 mEq/L (15 mmol/L), the LD is calculated thus:
The calculated LD of alkali should be administered over several days to avoid volume overload from the accompanying sodium load. For this scenario, a regimen of 60 to 70 mEq (60 to 70 mmol) three times a day for 3 to 5 days should result in an increase in HCO3– levels toward normal. In addition to the calculated LD, supplemental alkali must also be provided to replace ongoing losses, which can be approximated to be 2 mEq/kg (2 mmol/kg) per day or 40 mEq (40 mmol) three times a day. In patients with associated volume depletion, bicarbonate replacement can be provided simultaneous with volume resuscitation by substituting bicarbonate for chloride in IV crystalloid solutions.
TABLE 37-8 Therapeutic Alternatives for Oral Alkali Replacement
In patients with chronic metabolic acidosis because of GI bicarbonate losses, maintenance therapy should provide sufficient alkali to replace ongoing bicarbonate losses. The magnitude of this replacement is variable and can be substantial (>10 mEq/kg [>10 mmol/kg] per day). In addition, associated losses of other electrolytes, such as potassium and magnesium, may need to be replaced (see Chap. 36).
Proximal (type II) RTA is a bicarbonate-wasting disorder that requires the administration of large maintenance doses of alkali (10 to 15 mEq/kg [10 to 15 mmol/kg] per day). As alkali replacement raises the serum bicarbonate concentration toward normal, the proximal tubule’s capacity to reabsorb bicarbonate is overwhelmed, and renal bicarbonate wasting increases. In children, aggressive therapy of proximal RTA is necessary to avoid growth retardation and osteopenia. Because this is generally a mild, nonprogressive acidosis in adults, the benefit of alkali therapy is frequently outweighed by the risks of increased potassium wasting. In patients with classic distal (type I) RTA, maintenance therapy usually requires only enough alkali to buffer the amount of acid generated from dietary intake and metabolism. This usually approximates 1 to 3 mEq/kg per day (1 to 3 mmol/kg per day).
After initial potassium deficits are replaced, ongoing potassium supplementation may not be required, as renal potassium losses decrease following initiation of appropriate alkali therapy. The use of potassium alkali salts can, however, be desirable in patients with associated nephrolithiasis, because sodium salts can increase urinary calcium excretion.
The metabolic acidosis associated with hyperkalemic distal (type IV) RTA with hyporeninemic-hypoaldosteronemia that is often seen in patients with diabetes mellitus can be corrected by the treatment of hyperkalemia alone (see Chap. 36). The use of supplemental alkali (1 to 2 mEq/kg [1 to 2 mmol/kg] per day) to increase sodium intake and stimulate distal tubular potassium secretion can be beneficial. A minority of patients require the administration of pharmacologic amounts of fludrocortisone.7 Type IV RTA resulting from a generalized distal tubular disorder often responds to low doses of alkali (1.5 to 2.0 mEq/kg [1.5 to 2.0 mmol/kg] per day).27 Corrections of the acidosis along with modest dietary potassium restriction (to 1 mEq/kg [1 mmol/kg] per day) will often result in the maintenance of serum potassium levels of 5 mEq/L (5 mmol/L) or less.
Acute Severe Metabolic Acidosis
The management of patients with life-threatening acute metabolic acidosis (plasma bicarbonate of 8 mEq/L [8 mmol/L] and pH <7.20) is dependent on the underlying cause and the patient’s cardiovascular status. In some cases, patients will require emergent hemodialysis therapy (see Chap. 30). Patients with hyperchloremic acidosis (e.g., diarrhea-induced) are unable to regenerate bicarbonate, and the generation of new bicarbonate by the kidneys can require several days before one can observe a meaningful change in their status.11 Thus IV alkali therapy is often required for these patients.
Although conventional wisdom recommends the use of alkali replacement in patients with severe acidemia because of the deleterious effects of acidemia on circulatory function,6,7,10,11 studies have not demonstrated that its administration improves patient outcomes.28–31 Alkali therapies may either improve or worsen clinically relevant endpoints such as [H+], PaCO2, lactate concentrations, and cardiac output. The specific patient populations most likely to benefit or be harmed from alkalinizing therapy are presented in Table 37-9.
TABLE 37-9 Patient Populations Likely to Benefit or Suffer from Alkalinizing Therapy
There are several therapeutic alternatives available for the acute correction of severe metabolic acidosis. Sodium acetate, sodium citrate, and sodium lactate are unreliable sources of alkali because their alkalinizing effect is dependent on their oxidative conversion to bicarbonate. This process is often impaired in critically ill patients, especially those with liver disease or circulatory failure. Although sodium bicarbonate is the most widely used IV alkalotic agent,7 several studies suggest that it is frequently ineffective and can actually be deleterious, especially in patients with lactic acidosis.28–31 Two of the three remaining alternatives (Carbicarb and dichloroacetate [DCA]) are investigational and not available in most clinical settings. Tromethamine, or THAM, is a carbon dioxide-consuming, commercially available solution that buffers respiratory as well as metabolic acids.
The role of alkali therapy in patients with severe lactic acidosis is controversial. Treatment should be directed at the underlying causes as serial bicarbonate administration is often not effective and in some settings can be deleterious.
While sodium bicarbonate administration provides fluid and electrolyte replacement and increases arterial pH, neither animal nor clinical studies demonstrate an improvement in cardiac function, organ perfusion, or intracellular pH.28–31 In addition, sodium bicarbonate administration can actually have paradoxical adverse effects on intracellular pH. When bicarbonate is given by IV infusion, the carbon dioxide generated diffuses more readily than bicarbonate across cell membranes and into cerebrospinal fluid. Therefore, the intracellular pH can actually be decreased by administration of bicarbonate.4
Excessive sodium bicarbonate administration can result in (a) a shift of the oxyhemoglobin saturation curve to the left, thereby impairing oxygen release from hemoglobin to tissues; (b) sodium and water overload, with subsequent pulmonary congestion and hypernatremia; (c) paradoxical tissue acidosis as a result of the production of CO2 that freely diffuses into myocardial and cerebral cells32; and (d) decreased ionized calcium with a resultant decrease in myocardial contractility. If there is an endogenous source of bicarbonate, such as can occur in the case of ketoacidosis or lactic acidosis, a bicarbonate “overshoot” can develop because the ketoacids (acetoacetic acid and β-hydroxybutyric acid) or lactic acid are converted in the liver to bicarbonate once the underlying cause of acidosis is corrected.10,11,33Alkalosis can also result if too much sodium bicarbonate is administered too quickly.
If IV sodium bicarbonate is used, one must be mindful that the goals are to increase, not normalize, pH (to approximately 7.20) and plasma bicarbonate (to 8 to 10 mEq/L [8 to 10 mmol/L]). There is no calculative method that will assure attainment of these goals with a given dose of sodium bicarbonate because of the multiplicity of competing processes that can affect acid–base status (e.g., vomiting, potential increases in endogenous acid production, and renal failure) and the marked variability in the volume of distribution of bicarbonate (50% of body weight in patients with mild acidosis to approximately 100% in those with severe acidosis).10,32–33 Kraut and Madias7 recommend that the dose of sodium bicarbonate be calculated using a distribution volume of 50% of body weight for all patients to avoid overtreatment. The total dose calculated as described previously in the RTA section should be administered as an infusion over one-half to several hours. Follow-up monitoring of ABGs, beginning no sooner than 30 minutes after the end of the infusion, should be used to guide further therapeutic decisions.
Although it has been recommended that sodium bicarbonate be administered to raise the arterial pH to approximately 7.20, in an effort to prevent complications such as ventricular tachyarrhythmia, there are no controlled clinical trials demonstrating that sodium bicarbonate administration is significantly better than general supportive care in reducing morbidity and mortality in these patients.30,31,34
Bicarbonate therapy is generally not necessary for patients with cardiac arrest, even if the initial arrest was unmonitored. The American Heart Association’s Advanced Cardiac Life Support (ACLS) provider manual states that sodium bicarbonate is not useful or effective during resuscitation in hypoxic patients with lactic acidosis.34 Additionally, sodium bicarbonate is considered to be not useful or effective in those who are undergoing prolonged resuscitation with effective ventilation.34 Furthermore, if sodium bicarbonate is used, it should be used only after defibrillation, cardiac compression, support of ventilation including intubation, and drug therapies such as epinephrine and antiarrhythmic agents have been employed.34 The initial dose of sodium bicarbonate in this situation is (1 mEq/kg [1 mmol/kg]) administered by rapid, direct IV injection.35 Subsequent doses of sodium bicarbonate should be based on measurements of arterial blood pH and PaCO2 given the propensity for it to cause alkalemia.36
THAM, available as a 0.3 N solution, is a highly alkaline, sodium-free organic amine that acts as a proton acceptor to prevent or correct acidosis.7,28 THAM combines with hydrogen ions from carbonic acid to form bicarbonate and a cationic buffer. THAM also acts as an osmotic diuretic to increase urine flow, urine pH, and the excretion of fixed acids, CO2, and electrolytes. At pH 7.4, 30% of THAM is not ionized and therefore can penetrate into cells and neutralize acidic anions of the intracellular fluid. Intracellular pH increases have been noted within 1 hour after the infusion of THAM. There is, however, no clinical or physiologic evidence that this action is beneficial, or that THAM is more efficacious than sodium bicarbonate.28,37
When THAM is used, it must be administered slowly, with careful monitoring to avoid alkalosis. The usual empiric dosage range for THAM is 1 to 5 mmol/kg administered IV over 1 hour, but doses up to 1.25 mmol/kg can be given over 5 to 15 minutes in acute situations. The dose of THAM can be individualized using the following equation35:
Dose of THAM (in mL) = 1.1 × BW (in kg) × base deficit
where base deficit = normal [HCO3-] – current [HCO3-].
The need for additional THAM is determined by serial measurements of the serum bicarbonate concentration and calculation of the base deficit. Large doses can cause respiratory depression as a result of an increase in blood pH and a decrease in PaCO2 concentration.35 THAM solution is highly alkaline and can cause severe inflammation, vascular spasm, or tissue damage (necrosis, sloughing, pain, chemical phlebitis, or thrombosis) if infiltration occurs. Hyperkalemia, hypoglycemia, hypocalcemia, and impaired coagulation have also been reported.28,35 This agent should only be used with extreme caution in patients with severe liver or kidney failure.
Carbicarb is an equimolar mixture of sodium carbonate (Na2CO3) and sodium bicarbonate (NaHCO3).38,39 Given that the carbonate ion is a stronger base than bicarbonate, Carbicarb preferentially buffers hydrogen ions resulting in the formation of bicarbonate rather than CO2. Thus Carbicarb limits, but does not eliminate, the generation of CO2. Unlike bicarbonate, which can produce a paradoxical intracellular acidosis and thereby impair cardiac function, Carbicarb appears to correct intracellular acidosis if present.40,41 Despite these effects, there are no consistent data on the effects of Carbicarb on hemodynamic endpoints and this agent is not available for use in humans.28
DCA, another investigational agent, facilitates aerobic lactate metabolism by stimulating the activity of lactate dehydrogenase, thus reversing hyperlactatemia and elevating blood pH.42,43 DCA, when compared to conventional management in controlled studies, however, has not been shown to improve hemodynamic parameters or clinical outcomes.42–44 DCA can cause mild drowsiness and peripheral neuropathy that can be ameliorated or prevented with thiamine supplementation.45 The future role of DCA in the management of metabolic acidosis, particularly lactic acidosis, remains to be clarified.28
Metabolic alkalosis is a simple acid–base disorder that presents as alkalemia (increased arterial pH) with an increase in plasma bicarbonate. It is an extremely common entity in hospitalized patients with acid–base disturbances. Under normal circumstances, the kidney is readily able to excrete an alkali load. Thus evaluation of patients with metabolic alkalosis must consider two separate issues: (a) the initial process that generates the metabolic alkalosis; and (b) alterations in renal function that maintain the alkalemic state.46,47
The generation of metabolic alkalosis can also result from excessive losses of hydrogen ions from the kidneys or stomach or from a gain secondary to the ingestion or administration of bicarbonate-rich fluids. Gastric juice, rich in chloride and hydrogen ions, is secreted at a rate of less than 50 mL/h in the basal state, but can increase up to fivefold with stimulation.8 In the gastric parietal cells, the hydrogen ion and bicarbonate are generated from CO2and water.46 The hydrogen ion is secreted into gastric fluid, and the bicarbonate is retained in the ECF. Normally, an amount of bicarbonate equal to the bicarbonate generated in the stomach is eliminated in the alkaline pancreatic and small-bowel secretions, maintaining hydrogen ion balance. With vomiting and nasogastric suctioning, the hydrogen ion is lost externally and metabolic alkalosis results. Diarrhea, as seen with secretory villous adenomas and other secretory diarrheas, often results in excessive GI losses of chloride-rich, bicarbonate-poor fluid, and thus leads to the generation of metabolic alkalosis.
Diuretic agents acting on the thick ascending limb of the loop of Henle (e.g., furosemide, bumetanide, and torsemide) and distal convoluted tubule (e.g., thiazides) have most commonly been associated with the generation of metabolic alkalosis.47,48 These agents promote the excretion of sodium and potassium almost exclusively in association with chloride, without a proportionate increase in bicarbonate excretion. Collecting duct hydrogen ion secretion is stimulated directly by the increased luminal flow rate and sodium delivery, and indirectly by intravascular volume contraction, which results in secondary hyperaldosteronism. Renal ammoniagenesis can also be stimulated by concomitant hypokalemia, further augmenting net acid excretion.
Increased renal acid excretion can also be the result of excess mineralocorticoid activity. Elevated mineralocorticoid levels directly stimulate collecting duct hydrogen ion secretion and indirectly increase ammoniagenesis by causing hypokalemia.46 Increased mineralocorticoid activity can result from Cushing’s syndrome, primary hyperaldosteronism, or hyperaldosteronism secondary to increased renin activity (e.g., malignant hypertension). In Bartter’s and Gitelman’s syndromes, defects in sodium transport in the loop of Henle (Bartter) or distal convoluted tubule (Gitelman) lead to hypokalemia, secondary hyperaldosteronism, and metabolic alkalosis.46 In Liddle’s syndrome, enhanced sodium reabsorption by the cortical collecting duct epithelial sodium channel results in a syndrome of pseudohyperaldosteronism.46 Administration of high doses of penicillins (e.g., ticarcillin) can produce metabolic alkalosis because they act as nonreabsorbable anions. High concentrations of poorly reabsorbable anions in the distal renal tubule increase luminal flow rate and luminal electronegativity, which enhances the secretion of potassium and hydrogen ions and results in hypokalemia and metabolic alkalosis.
Metabolic alkalosis can also be generated by the gain of exogenous alkali. This can be seen as a result of bicarbonate administration or from the infusion of organic anions that are metabolized to bicarbonate, such as acetate, lactate, and citrate. The milk-alkali syndrome was historically a common cause of metabolic alkalosis in patients with peptic ulcer disease secondary to the ingestion of large quantities of milk products and antacids. With the advent of alternative therapies for dyspeptic syndromes that are far more effective than milk, this syndrome is now rarely seen.
Metabolic alkalosis is predominantly maintained because of an abnormality in renal function. Normally, the kidneys are capable of excreting all of the excess bicarbonate presented to them, even during periods of increased bicarbonate loads.4 As the serum bicarbonate concentration increases, the filtered bicarbonate load exceeds the maximal rate for bicarbonate reabsorption, and the excess bicarbonate is excreted in the urine. Under normal circumstances, the excess bicarbonate is rapidly excreted, and metabolic alkalosis does not occur or is corrected in a matter of hours.46
Several mechanisms can impair renal bicarbonate excretion and contribute to the maintenance phase of metabolic alkalosis.46 In general, these mechanisms can be divided into volume-mediated processes (sodium chloride-responsive) and volume-independent processes (sodium chloride-resistant) that are predominantly associated with excess mineralocorticoid activity and hypokalemia (Table 37-10). Intravascular volume depletion maintains metabolic alkalosis through a number of mechanisms. Decreases in the glomerular filtration rate reduce the filtered load of bicarbonate at any given serum concentration, thereby decreasing the kidney’s ability to excrete a bicarbonate load. Although this can play a role in patients with chronic kidney disease, it is also an important factor in patients in whom intravascular volume contraction accompanies metabolic alkalosis. Decreased effective arterial blood volume also enhances proximal and distal tubular sodium reabsorption. Sodium reabsorption must be coupled with reabsorption of an anion, such as chloride or bicarbonate, or exchange with a cation, such as potassium or hydrogen, to maintain charge neutrality. In the proximal tubule, increased sodium reabsorption stimulates bicarbonate reabsorption. In the distal nephron, enhanced sodium reabsorption, particularly in the setting of hypokalemia, stimulates hydrogen ion secretion.
TABLE 37-10 Causes of Metabolic Alkalosis Differentiated on the Basis of Their Responsiveness to Sodium Chloride
Mineralocorticoid excess also plays a significant role in the maintenance of metabolic alkalosis. In patients with volume-responsive metabolic alkalosis, intravascular volume depletion stimulates aldosterone secretion. As discussed earlier, excess mineralocorticoid activity can also underlie the generation of metabolic alkalosis. In either situation, the increased mineralocorticoid effect stimulates collecting duct hydrogen ion secretion. Metabolic alkalosis can also be maintained by persistent hypokalemia, enhancing proximal tubular bicarbonate reabsorption, stimulating ammoniagenesis, and increasing distal tubular hydrogen ion secretion.46
There are no unique signs or symptoms associated with mild to moderate metabolic alkalosis, but patients may complain of symptoms related to the underlying cause of the disorder (e.g., muscle weakness with hypokalemia or postural dizziness with volume depletion). They may have a history of vomiting, gastric drainage, or diuretic use, all of which contribute to the development of metabolic alkalosis. Severe alkalemia (blood pH >7.60) has been associated with cardiac arrhythmias, particularly in patients with heart disease, hyperventilation, and hypoxemia.46 Neuromuscular irritability can be present, with signs of tetany or hyperactive reflexes, possibly caused by the decreased ionized calcium concentration that occurs secondary to the increase in pH. This decrease in ionized calcium may be caused by a conformational change in the albumin molecules to which the calcium is bound, resulting in increased binding, or by decreased competition from hydrogen ions for binding sites on the albumin molecule. Mental confusion, muscle cramping, and paresthesia can also occur. Lastly, patients will be more difficult to liberate from mechanical ventilation.
The respiratory response to metabolic alkalosis is hypoventilation, which results in an increased PaCO2. Respiratory compensation is initiated within hours when the central and peripheral chemoreceptors sense an increase in pH. The PaCO2 increases 6 to 7 mm Hg (0.8 to 0.9 kPa) for each 10 mEq/L (10 mmol/L) increase in bicarbonate, up to a PaCO2 of approximately 50 to 60 mm Hg (6.7 to 8.0 kPa) (see Table 37-7) before hypoxia sensors react to prevent further hypoventilation. If the PaCO2 is normal or less than normal, one should consider the presence of a superimposed respiratory alkalosis, which can be secondary to fever, gram-negative sepsis, or pain.
Because the body tolerates alkalemia far less well than acidemia, treatment of metabolic alkalosis is nearly always required and should be aimed at correcting the factor(s) responsible for the maintenance of the alkalosis.46 For example, vomiting should be treated with antiemetics, gastric losses of hydrogen ions during nasogastric suction can be modulated by giving histamine blockers such as ranitidine or proton pump inhibitors such as omeprazole, and reducing or discontinuing diuretic therapy.46,49 Metabolic alkalosis will persist until the renal mechanism responsible for maintaining the disorder is corrected, despite the fact that the original cause of the elevated plasma bicarbonate may have resolved. For example, hypovolemia should be treated with sodium chloride (i.e., diuretic abuse or nasogastric suction) to allow excretion of bicarbonate by the kidney. However, patients with severely compromised cardiovascular function may not be able to tolerate this therapeutic approach. In situations such as this and/or the presence of life-threatening alkalosis, some have advocated reduction in pH by control of ventilation.4 Although controlled hypoventilation, sometimes using inspired CO2 with supplemental oxygen to prevent hypoxia can be lifesaving,4 this approach is not universally accepted.33 Therapy for metabolic alkalosis can be conceptualized on the basis of the sodium chloride responsiveness of the disorders as shown in Figure 37-4.
FIGURE 37-4 Treatment algorithm for patients with primary metabolic alkalosis. (BID, twice daily; CHF, chronic heart failure; K, potassium [serum potassium in mEq/L is numerically equivalent to mmol/L]; PO, orally; QD, every day.)
SODIUM CHLORIDE-RESPONSIVE DISORDERS
Sodium chloride-responsive disorders usually result from volume depletion and chloride loss, which can accompany severe vomiting, prolonged nasogastric suction, and diuretic therapy. Initially, therapy is directed at expanding intravascular volume and replenishing chloride stores. Sodium and potassium chloride-containing solutions should be administered to patients who can tolerate the volume load.46Patients with metabolic alkalosis who are volume overloaded or intolerant to volume administration because of congestive heart failure can benefit from the carbonic anhydrase inhibitor acetazolamide. This agent inhibits the action of carbonic anhydrase, thereby inhibiting renal bicarbonate reabsorption. Unfortunately, it also increases the renal losses of potassium and phosphate. Administration of acetazolamide (250 to 375 mg once or twice daily) can promote a sufficient bicarbonate diuresis and return the pH toward normal.50 However, because the clinical effectiveness of the drug declines as the HCO3–concentration decreases, only rarely will this approach fully correct the alkalosis.46
Acidifying agents including hydrochloric acid, ammonium chloride, and arginine monohydrochloride can be used to treat severe (pH >7.6) symptomatic metabolic alkalosis.51 In general, this management is reserved for patients who are unresponsive to conventional fluid and electrolyte management or who are unable to tolerate the requisite volume load because of decompensated congestive heart failure or advanced renal failure.46 Alternatively, hemodialysis using a low-bicarbonate dialysate can be used for the rapid correction of metabolic alkalosis.
Hydrochloric acid is usually infused IV via a large central vein as a 0.1 to 0.25 N HCl solution in either 5% dextrose or normal saline, although sterile water has also been used. Extemporaneously prepared solutions can be made by adding 100 to 250 mEq (100 to 250 mmol) of HCl through a 0.22-mm filter into a glass container of saline or dextrose. Hydrochloric acid can also be added to parenteral nutrient solutions and administered via a central line without serious degradation of proteins.52 The rate of infusion should be 100 to 125 mL/h (10 to 25 mEq/h [10 to 25 mmol/h]), with frequent monitoring of ABGs. To prevent overcorrection, the infusion should be stopped when the arterial pH decreases to 7.50.46
The dose of hydrochloric acid can be based on an estimate of the total body chloride deficit35:
where the estimated chloride space is 0.2 times the body weight, and the average serum chloride is 103 mEq/L (103 mmol/L). Alternatively, the dose can be calculated based on the estimated base deficit46:
At present, there are no comparative data that address the relative accuracy of these two formulas for determining the dose of hydrochloric acid.
The dose of hydrochloric acid is usually infused IV over 12 to 24 hours.35 A severe transient respiratory acidosis can occur if the hydrochloric acid is infused too quickly because of a slower reduction of the elevated bicarbonate concentration in the cerebrospinal fluid than in the ECF. Improvement is usually seen within 24 hours of initiating therapy. ABGs and serum electrolytes should be drawn every 4 to 8 hours to evaluate and adjust therapy.
Ammonium chloride has a limited role in the treatment of metabolic alkalosis. The liver converts ammonium chloride (NH4Cl) to urea and free hydrochloric acid35:
The dose of ammonium chloride can be calculated on the basis of the chloride deficit using the same method as for HCl and assuming that 20 g ammonium chloride will provide 374 mEq (374 mmol) of H+. However, only one half of the calculated dose of ammonium chloride should be administered so as to avoid ammonia toxicity. Ammonium chloride is available as a 26.75% solution containing 100 mEq (100 mmol) of H+ in 20 mL, which should be further diluted prior to administration. A dilute solution can be prepared by adding 20 mL of ammonium chloride to 500 mL of normal saline and infusing the solution at a rate of no more than 1 mEq/min (1 mmol/min). Improvement in metabolic status is usually seen within 24 hours. CNS toxicity, marked by confusion, irritability, seizures, and coma, has been associated with more rapid rates of administration. Ammonium chloride must be administered cautiously to patients with renal or hepatic impairment. In patients with hepatic dysfunction, impaired conversion of ammonia to urea can result in increased ammonia levels and worsened encephalopathy. In patients with renal failure, the increased urea synthesis can exacerbate uremic symptoms.35
Arginine monohydrochloride at a dose of 10 g/h given IV has been used to treat metabolic alkalosis, although it was never FDA approved for this purpose.35 Like ammonium chloride, arginine must undergo metabolism by the liver to produce hydrogen ions, with a conversion of 100 g to 475 mEq (475 mmol) of H+. Unlike ammonium chloride, arginine combines with ammonia in the body to synthesize urea; thus it can be used in patients with relative hepatic insufficiency. Patients with renal insufficiency should not receive arginine monohydrochloride because it can significantly elevate blood urea nitrogen and is associated with severe hyperkalemia.35,46 The increase in potassium is caused by arginine-induced shifts of potassium from the intracellular to the extracellular space.
SODIUM CHLORIDE-RESISTANT DISORDERS
Management of these disorders usually consists of treatment of the underlying cause of the mineralocorticoid excess. For patients taking a corticosteroid, a dosage reduction or a switch to a corticosteroid with less mineralocorticoid activity (e.g., methylprednisolone) should be considered. Patients with an endogenous source of excess mineralocorticoid activity can require surgery or the administration of spironolactone, amiloride, or triamterene.46,48,53,54
Spironolactone is a competitive antagonist of the mineralocorticoid receptor. Amiloride and triamterene are potassium-sparing diuretics that inhibit the epithelial sodium channel in the distal convoluted tubule and collecting duct. All three agents inhibit aldosterone-stimulated sodium reabsorption in the collecting duct. In addition, spironolactone directly inhibits aldosterone stimulation of the hydrogen ion secretory pump. Thus, most patients with mineralocorticoid excess, including Bartter’s and Gitelman’s syndromes, respond to therapy with these agents.46,53–56 Liddle’s syndrome, which is a form of pseudohypoaldosteronism caused by overactivity of the epithelial sodium channel, is not responsive to spironolactone but can be treated with either amiloride or triamterene. Although experience is limited, some patients with Bartter’s and Gitelman’s syndromes may respond to NSAIDs or ACE inhibitors.53–56 Finally, aggressive potassium repletion can correct the alkalosis in those who have not responded to the approaches outlined above.
RESPIRATORY ACID–BASE DISORDERS
As with the metabolic acid–base disturbances, there are two cardinal respiratory acid–base disturbances: respiratory acidosis and respiratory alkalosis. These disorders are generated by a primary alteration in carbon dioxide excretion, which changes the concentration of carbon dioxide, and therefore the carbonic acid concentration in body fluids. A primary reduction in PaCO2 causes an increase in pH (respiratory alkalosis), and a primary increase in PaCO2 causes a decrease in pH (respiratory acidosis). Unlike the metabolic disturbances, for which respiratory compensation is rapid, metabolic compensation for the respiratory disturbances is slow. Hence, these disturbances can be further divided into acute disorders, with a duration of minutes to hours, and where metabolic compensation has yet to occur, and chronic disorders that have been present long enough for metabolic compensation to be complete.
Respiratory alkalosis is characterized by a primary decrease in PaCO2 that leads to an elevation in pH. The PaCO2 decreases when the excretion of CO2 by the lungs exceeds the metabolic production of CO2. It is the most frequently encountered acid–base disorder, occurring physiologically in normal pregnancy and in persons living at high altitudes.56,57 Respiratory alkalosis also occurs frequently among hospitalized patients (Table 37-11).
TABLE 37-11 Causes of Respiratory Alkalosis
A decrease in PaCO2 occurs when ventilatory excretion exceeds metabolic production. Because endogenous production of CO2 is relatively constant, negative CO2 balance is primarily caused by an increase in ventilatory excretion of CO2 (hyperventilation). The metabolic production of CO2, however, can be increased during periods of stress or with excess carbohydrate administration (e.g., parenteral nutrition). Hyperventilation can develop from an increase in neurochemical stimulation via either central or peripheral mechanisms, or be the result of voluntary or mechanical (iatrogenic) hyperventilation.
A decrease in PaCO2 can occur in patients with cardiogenic, hypovolemic, or septic shock because oxygen delivery to the carotid and aortic chemoreceptors is reduced. This relative deficit in PaO2 stimulates an increase in ventilation. The hyperventilation in sepsis is also mediated via a central mechanism. Hyperventilation-induced respiratory alkalosis with an elevation in cardiac index and hypotension without peripheral vasoconstriction can therefore be an early sign of sepsis.
Although most patients are asymptomatic, respiratory alkalosis can cause adverse neuromuscular, cardiovascular, and GI effects.56 During periods of decreased PaCO2, there is a decrease in cerebral blood flow, which can be responsible for symptoms of light-headedness, confusion, decreased intellectual functioning, syncope, and seizures. Nausea and vomiting can occur, probably as a result of cerebral hypoxia. In severe respiratory alkalosis, cardiac arrhythmias can occur because of sensitization of the myocardium to the arrhythmogenic effects of circulating catecholamines.2 Acute respiratory alkalosis has no effect on blood pressure or cardiac output in awake individuals. Anesthetized patients, however, can experience a decrease in both cardiac output and blood pressure, possibly owing to the lack of a tachycardic response.56
The concentration of serum electrolytes can also be altered secondary to the development of respiratory alkalosis. The serum chloride concentration is usually slightly increased, and serum potassium concentration can be slightly decreased. Clinically significant hypokalemia can be a consequence of extreme respiratory alkalosis, although the effect is usually very small or negligible.2,56 Serum phosphorus concentration can decrease by as much as 1.5 to 2.0 mg/dL (0.48 to 0.65 mmol/L) because of the shift of inorganic phosphate into cells. Reductions in the blood ionized calcium concentration can be partially responsible for symptoms such as muscle cramps and tetany. Approximately 50% of calcium is bound to albumin, and an increase in pH results in an increase in binding.56
The initial response of the body to acute respiratory alkalosis is chemical buffering: hydrogen ions are released from the body’s buffers—intracellular proteins, phosphates, and hemoglobin—and titrate down the serum bicarbonate concentration. This process occurs within minutes. Acutely, the bicarbonate concentration can be decreased by a maximum of 3 mEq/L (3 mmol/L) for each 10-mm Hg (1.3 kPa) decrease in PaCO224 (see Table 37-7). When only physicochemical buffering has occurred, the disturbance is referred to as acute respiratory alkalosis.
Metabolic compensation occurs when respiratory alkalosis persists for more than 6 to 12 hours. In response to the alkalemia, proximal tubular bicarbonate reabsorption is inhibited, and the serum bicarbonate concentration decreases. Renal compensation is usually complete within 1 to 2 days. The renal bicarbonaturia, as well as decreased NH4+ and titratable acid excretion, are direct effects of the reduced PaCO2and pH on renal reabsorption of chloride and bicarbonate.2 The acuity of the respiratory alkalosis can be assessed on the basis of the degree of renal compensation (see Table 37-7). In fully compensated respiratory alkalosis, the bicarbonate concentration decreases by 4 mEq/L (4 mmol/L) below 24 for each 10-mm Hg (1.3 kPa) drop in PaCO2. For example, a sustained decrease in PaCO2 of 20 mm Hg (2.7 kPa) will lower serum bicarbonate from 24 to 14 mEq/L (24 to 14 mmol/L) with a resultant pH of 7.46. Bicarbonate concentrations differing from those anticipated using the preceding guidelines suggest a mixed acid–base disorder.
CLINICAL PRESENTATION Respiratory Alkalosis
• The patient is usually asymptomatic if the condition is chronic and mild
• The patient may complain of light-headedness, confusion, muscle cramps and tetany, and decreased intellectual functioning
• Nausea and vomiting can occur, probably as a result of cerebral hypoxia
• In severe respiratory alkalosis pH >7.60
• Syncope and seizures
• Cardiac arrhythmias
• Serum chloride concentration is usually slightly increased. Serum ionized calcium, potassium, and phosphorus concentration can be decreased
Because most patients with respiratory alkalosis, especially chronic cases, have few or no symptoms and pH alterations are usually mild (pH not exceeding 7.50), treatment is often not required.56,57 The first consideration in the treatment of acute respiratory alkalosis with pH >7.50 is the identification and correction of the underlying cause. Relief of pain, correction of hypovolemia with IV fluids, treatment of fever or infection, treatment of salicylate overdose, and other direct measures can prove effective. A rebreathing device, such as a paper bag, can be useful in controlling hyperventilation in patients with the anxiety/hyperventilation syndrome.57 Oxygen therapy should be initiated in patients with severe hypoxemia. Patients with life-threatening alkalosis (pH >7.60), particularly if it is a mixed respiratory and metabolic condition, tend to have complications, such as arrhythmias or seizures, which can require mechanical ventilation with sedation and/or paralysis to control hyperventilation.
Respiratory alkalosis in patients receiving mechanical ventilation is usually iatrogenic. It can often be corrected by decreasing either the set respiratory rate or tidal volume, although other measures can also be employed. The use of a capnograph and spirometer in the breathing circuit enables a more precise adjustment of the ventilator settings. Another method of treating respiratory alkalosis is to increase the amount of dead space in the ventilator circuit by placing a known length of tubing between the artificial airway and the “Y” piece of the ventilator. This results in “rebreathing” of expired gas, and therefore an increase in the inspired carbon dioxide concentration, which should increase the carbon dioxide tension of the patient, correcting the respiratory alkalosis. In patients breathing more rapidly than the ventilator settings, sedation with or without paralysis can be employed.
Respiratory acidosis occurs when the lungs fail to excrete carbon dioxide resulting in a lower pH. This can be the result of conditions that centrally inhibit the respiratory center, diseases that interfere with pulmonary perfusion or neuromuscular function, and intrinsic airway or parenchymal pulmonary disease (Table 37-12). Acute respiratory acidosis with hypoxemia, hypercarbia, and acidosis is life-threatening. Those disorders that produce an increase in PaCO2 and hypoxemia to a degree compatible with life (e.g., chronic obstructive pulmonary disease), with or without oxygen therapy, can result in chronic respiratory acidosis (Table 37-13). These patients can function normally without noticeable neurologic defects with PaCO2 concentrations in the range of 90 to 100 mm Hg (12 to 13.3 kPa) (normal, 40 mm Hg [5.3 kPa]), provided that adequate oxygenation is maintained.56
TABLE 37-12 Causes of Acute Respiratory Acidosis
TABLE 37-13 Causes of Chronic Respiratory Acidosis
Respiratory acidosis can produce neurologic symptoms, including altered mental status, abnormal behavior, seizures, stupor, and coma. Hypercapnia can mimic stroke or CNS tumors by producing headache, papilledema, focal paresis, and abnormal reflexes. These CNS symptoms are attributable to the vasodilator effects of CO2 in the brain that result in an increase in cerebral blood flow.2 The CNS response to hypercapnia is extremely variable between patients and is most influenced by the acuity of presentation. Given that chronic hypercapnia blunts the usual respiratory stimulus of an elevated PaCO2, hypoxemia rather than hypercapnia provides the primary ventilatory stimulus in patients with severe chronic respiratory acidosis.56
CLINICAL PRESENTATION Respiratory Acidosis
• The patient is usually symptomatic
• The patient may complain of confusion or difficulty thinking and headache
• In severe respiratory acidosis:
• Cardiac: Increased cardiac output if moderate that decreases if severe. Refractory hypotension can be present in some patients
• CNS: Abnormal behavior, seizures, stupor, and coma. Papilledema, focal paresis, and abnormal reflexes can also be present
• Serum potassium concentration can be modestly increased. Hypercapnia can be moderate (PaCO2 of 50 to 55 mm Hg [6.7–7.3 kPa]) to severe (PaCO2 of >80 mm Hg [>10.6 kPa]). Hypoxia (PaO2 is <70 mm Hg [<9.3 kPa]) is often present
The degree to which cardiac contractility and heart rate are altered depends on the severity of the acidosis and the rapidity with which it develops. Modest acute hypercapnia (PaCO2 of 50 to 55 mm Hg [6.7 to 7.3 kPa]) stimulates a stress-like response, with elevated catecholamines and corticosteroid hormone levels, and can result in increased cardiac output and pulmonary artery pressure.56 As the severity increases, cardiac output declines and vascular resistance decreases leading to refractory hypotension in some patients.2
In respiratory acidosis, the serum potassium concentration increases modestly secondary to cellular shifts. The increases are less than those seen with inorganic metabolic acidosis and are difficult to predict for individual patients.
The body responds to acute respiratory acidosis with chemical buffering. The increase in PaCO2 results in increased carbonic acid levels. The carbonic acid dissociates, releasing hydrogen ions, which are buffered by nonbicarbonate buffers (i.e., proteins, phosphate, and hemoglobin) and bicarbonate. Thus, on the basis of physicochemical factors, increases in PaCO2 raise the serum bicarbonate concentration. In general, in acute respiratory acidosis, the bicarbonate concentration increases by 1 mEq/L (1 mmol/L) above 24 for each 10-mm Hg (1.3 kPa) increase in PaCO2 above 40 (5.3 kPa) (see Table 37-7).
Metabolic compensation occurs when respiratory acidosis is prolonged beyond 12 to 24 hours. In response to hypercapnia and acidemia, proximal tubular bicarbonate reabsorption, ammoniagenesis, and distal tubular hydrogen secretion are enhanced, resulting in an increase in the serum bicarbonate concentration that raises the pH toward normal. Renal compensation for chronic hypercapnia generally results in the plasma bicarbonate concentration increasing by 4 mEq/L (4 mmol/L) above 24 for each 10-mm Hg (1.3 kPa) increase in PaCO2 above 40 (5.3 kPa) (see Table 37-7). The new steady state in acid–base values is generally achieved within 5 days of the onset of hypercapnia in dogs; the time interval necessary for compensation in humans has not been established.
The treatment of respiratory acidosis is dependent on the chronicity of the patient’s condition. Respiratory decompensation in patients with chronic elevations in PaCO2 is frequently seen in those with acute infections and those recently started on narcotic analgesics or oxygen therapy.56 Aggressive treatment of these conditions can offer considerable benefit and should be initiated. Furthermore, tranquilizers and sedatives should be avoided and supplemental oxygen, if used, should be minimized.
Acute Respiratory Acidosis
When carbon dioxide excretion is severely impaired (PaCO2 >80 mm Hg [>10.6 kPa]) and/or life-threatening, hypoxia is present (PaO2 <40 mm Hg [<5.3 kPa]); the immediate therapeutic goal is to provide adequate oxygenation. Under these circumstances, hypoxia, not acidemia, is the principal threat to life. A patent airway needs to be established, which can necessitate intubation. Excessive secretions must be cleared from the airway and oxygen administered to restore adequate oxygenation. Mechanical ventilation is usually required.
The underlying cause of the acidosis should be treated aggressively (e.g., bronchodilators for treatment of severe bronchospasm; narcotic or benzodiazepine antagonists to reverse the deleterious effects of these agents on the respiratory center). Bicarbonate administration is rarely necessary in the treatment of respiratory acidosis. Furthermore, rapid correction of acidosis with bicarbonate can eliminate the patient’s respiratory drive or precipitate metabolic alkalosis. Cautious use of alkali (bicarbonate or THAM) can restore the responsiveness of bronchial muscles to β-adrenergic agonists and thus can be beneficial for those patients with severe bronchospasm.56 ABGs should be monitored closely to ensure that the respiratory acidosis is resolving without creating a metabolic alkalosis as the result of compensatory elevation in HCO3- and decrease in PaCO2. ABGs should be obtained every 2 to 4 hours during the acute phase and less frequently (every 12 to 24 hours) as the acidosis improves.
Acute Respiratory Acidosis in a Compensated Chronic Respiratory Acidotic Patient
Patients with a history of chronic respiratory acidosis (e.g., those with chronic obstructive pulmonary disease) can experience an acute worsening of their respiratory acidosis. This can result in severe life-threatening hypoxemia. As with acute respiratory acidosis, the goals of therapy are maintenance of a patent airway and adequate oxygenation. Individuals with chronic respiratory acidosis are routinely able to tolerate a low PaO2 and an elevated PaCO2 because of compensation (increased number of red blood cells, hemoglobin content, and 2,3-diphosphoglycerate). The drive to breathe in these patients is dependent on hypoxemia rather than hypercarbia. Administration of oxygen to a patient with chronic respiratory acidosis can eliminate this drive to breathe and result in the syndrome of carbon dioxide narcosis. In this case, if the PaO2 is 50 mm Hg (6.7 kPa), no oxygen treatment is necessary. If the PaO2 is <50 mm Hg (<6.7 kPa), oxygen therapy should be initiated carefully using a controlled flow of oxygen.2
ABGs should be checked periodically to ensure adequate oxygenation. If the PaCO2 increases during oxygen therapy, it can be a sign of impending carbon dioxide narcosis and oxygen therapy may need to be discontinued. The underlying cause of the acute exacerbation should be aggressively managed. Pulmonary infections should be treated with the appropriate antibiotics and bronchodilators administered as necessary. Excess secretions should be cleared from the airway to allow proper gas exchange. This can involve increasing oral fluid intake to decrease the viscosity of secretions, deep breathing, and postural drainage, suction, or bronchoscopy.
MIXED ACID–BASE DISORDERS
The diagnosis of a mixed disorder depends on an understanding of the appropriate quantitative response of the compensatory mechanisms for each of the simple acid–base disturbances. To diagnose mixed disorders, one must know how each of the four simple disorders alters pH, PaCO2, and [HCO3-] (see Table 37-7). If a given set of blood gases does not decrease within the range of expected responses for a simple acid–base disturbance, a mixed disorder should be suspected. In addition to laboratory information, a thorough history and physical examination of the patient will often lead to the diagnosis, even before the laboratory data are available. Examples of common mixed disturbances follow.
Mixed Respiratory Acidosis and Metabolic Acidosis
In mixed respiratory and metabolic acidosis, there is a failure of compensation. The respiratory disorder prevents the compensatory decrease in PaCO2 expected in the defense against metabolic acidosis. The metabolic disorder prevents the buffering and renal mechanisms from raising the bicarbonate concentration as expected in the defense against respiratory acidosis. In the absence of these compensatory mechanisms, the pH decreases markedly.
Mixed respiratory and metabolic acidosis may develop in patients with cardiorespiratory arrest, in those with chronic lung disease who are in shock, and in metabolic acidosis patients who develop respiratory failure. When treating this mixed disorder, clinicians need to respond to both the respiratory and metabolic acidosis. Improved oxygen delivery must be initiated to improve hypercarbia and hypoxia. Mechanical ventilation may be needed to reduce PaCO2. During the initial stage of therapy, appropriate amounts of alkali should be given to reverse the metabolic acidosis (see Treatment, Metabolic Acidosis above).
Mixed Respiratory Alkalosis and Metabolic Alkalosis
The combination of respiratory and metabolic alkalosis is the most common mixed acid–base disorder. This mixed disorder occurs frequently in critically ill surgical patients with respiratory alkalosis caused by mechanical ventilation, hypoxia, sepsis, hypotension, neurologic damage, pain, or drugs, and with metabolic alkalosis caused by vomiting or nasogastric suctioning and massive blood transfusions. It can also occur in patients with hepatic cirrhosis who hyperventilate, receive diuretics, or vomit, as well as in patients with chronic respiratory acidosis and an elevated plasma bicarbonate concentration who are placed on mechanical ventilation and undergo a rapid decrease in PaCO2.
The renal excretion of bicarbonate that usually occurs as compensation for the respiratory alkalosis is prevented by the complicating metabolic alkalosis. Likewise, the retention of PaCO2 expected to compensate for metabolic alkalosis is prevented by the primary respiratory alkalosis. The failure of compensation that occurs with mixed respiratory and metabolic alkalosis can result in a severe alkalemia.
Administration of sodium chloride and potassium chloride solutions will help correct the metabolic component of this disorder and adjustment of the ventilator and/or treatment of an underlying disorder that is causing hyperventilation can correct or ameliorate the respiratory component of this mixed disorder.
Mixed Metabolic Acidosis and Respiratory Alkalosis
This mixed disorder is often seen in patients with advanced liver disease, salicylate intoxication, and pulmonary-renal syndromes. The respiratory alkalosis will decrease the PaCO2 beyond the appropriate range for the respiratory compensation usually seen with metabolic acidosis. The plasma bicarbonate concentration also decreases below the level expected in compensation for a simple respiratory alkalosis. In a sense, the defense of pH for either disorder alone is enhanced; thus the pH can be normal or close to normal, with a low PaCO2 and a low [HCO3-]. Treatment of this disorder should be directed at the underlying cause. Because of the enhanced compensation, the pH is usually closer to normal than in either of the two simple disorders.
Mixed Metabolic Alkalosis and Respiratory Acidosis
This mixed disorder often occurs in patients with chronic obstructive pulmonary disease and chronic respiratory acidosis who are treated with salt restriction, diuretics, and possibly glucocorticoids. When diuretics are initiated, the plasma bicarbonate may increase because of increased renal bicarbonate generation and reabsorption, providing mechanisms for both generating and maintaining metabolic alkalosis. The elevated pH diminishes respiratory drive and may therefore worsen the respiratory acidosis.
Although the pH may not deviate significantly from normal, treatment may need to be initiated to maintain PaO2 and PaCO2 at acceptable levels. Because it is often difficult to correctly identify this mixed disorder, it is helpful to observe the patient’s response to discontinuation of diuretics and administration of sodium and potassium chloride.2 If the patient has a simple metabolic alkalosis, the PaCO2 will normalize, but it will only minimally affect the PaCO2if it is a mixed disorder. Treatment should be aimed at decreasing the plasma bicarbonate with sodium and potassium chloride therapy, thereby allowing the renal excretion of retained bicarbonate from the diuretic-induced metabolic alkalosis. This therapy should be used cautiously to avoid exacerbating any underlying congestive heart failure.
CLINICAL BOTTOM LINE
Because acid–base disorders are such a common and widespread problem, pharmacists can play a key role in identifying, preventing, and properly treating acid–base abnormalities. Acid–base disorders do not occur only in the intensive care unit setting. Patients in ambulatory and extended care settings have many chronic conditions and drug therapies that commonly affect acid–base balance. Thus pharmacists in all practice settings should use their knowledge to identify patients at high risk for developing drug-related problems that affect acid–base balance and to undertake appropriate prevention and treatment measures to improve the quality of life of the patients they care for.
1. Stewart PA. Goals, definitions and basic principles. In: Kellum JA, Ebers Paul WG, eds. Stewart’s Textbook of Acid–Base, 2nd ed. Amsterdam: Elsevier, 2007:36–62.
2. Constable P. Clinical acid–base chemistry. In: Ronco C, ed. Critical Care Nephrology, 2nd ed. Philadelphia, PA: Elsevier, 2009:581–614.
3. Eckardt KU. Disorders of acid–base balance. In: Bremmer BM, ed. Bremmer and Rector’s The Kidney, 9th ed. New York: Saunders, 2012:619–679.
4. Kamel S, Razeen-Davids M, Lin SH, et al. Interpretation of electrolytes and acid–base parameters in blood and urine. In: Bremmer BM, ed. Bremmer and Rector’s The Kidney, 9th ed. New York: Saunders, 2012:906–928.
5. Morgan TJ. Unmeasured ions and the strong ion gap. In: Kellum JA, Ebers PWG, eds. Stewart’s Textbook of Acid–Base, 2nd ed. Amsterdam: Elsevier, 2007:324–335.
6. Boniatti MM, Cardozo PR, Castilho RK, et al. Acid–base disorders evaluation in critically ill patients: We can improve our diagnostic ability. Intensive Care Med 2009;35(8): 1377–1382.
7. Kraut JA, Madias NE. Metabolic acidosis: Pathophysiology, diagnosis and management. Nat Rev Nephrol 2010;6:274–285.
8. Ayers P, Dixon C. Simple acid–base tutorial. J Parentr Enteral Nutr 2012;36:18–23.
9. Reddy P. Clinical approach to renal tubular acidosis in adult patients. Int J Clin Pract 2011;65:350–360.
10. Dzierba AL, Abraham P. A practical approach to understanding acid–base abnormalities in critical illness. J Pharm Pract 2011;24:17–26.
11. Morris CG, Low J. Metabolic acidosis in the critically ill: Part 1. Classification and pathophysiology. Anaesthesia 2008;63(3):294–301.
12. Adrogue A, Gennari FJ, Galla JH, et al. Assessing acid–base disorders. Kidney Int 2009;76:1239–1247.
13. Liamis G, Milionis HJ, Elisaf M. Pharmacologically-induced metabolic acidosis: A review. Drug Saf 2010;33:371–391.
14. Levine M, Brooks DE, Truitt CA, et al. Toxicology in the ICU: Part 1: General overview and approach to treatment. Chest 2011;140:795–806.
15. Diamini J, Ledwaba L, Mokwena N, et al. Lactic acidosis and symptomatic hyperlactataemia in a randomized trial of first-line therapy in HIV-infected adults in South Africa. Antivir Ther 2011;16:605–609.
16. Salpeter SR, Greyber E, Pasternak GA, et al. Risk of fatal and nonfatal lactic acidosis with metformin use in type 2 diabetes mellitus. Cochrane Database Syst Rev 2010;20:CD002967.
17. Lalau JD. Lactic acidosis induced by metformin: Incidence, management and prevention. Drug Saf 2010;33:727–740.
18. Horinek EL, Kiser TH, Fish DN, et al. Propylene glycol accumulation in critically ill patients receiving intravenous lorazepam infusions. Ann Pharmacother 2009;43:1964–1971.
19. Pillai U, Hothi JC, Bhat ZY. Severe propylene glycol toxicity secondary to use of anti-epileptics. Am J Ther 2012 Oct 19 (epub ahead of press).
20. Deidrich DA, Brown DR. Analytic reviews: Propofol infusion syndrome in the ICU. J Intensive Care Med 2011;26:59–72.
21. Zietse R, Zoutendijk R, Hoorn EJ. Fluid, electrolyte and acid–base disorders associated with antibiotic therapy. Nat Rev Nephrol 2009;5(4):193–202.
22. Kato K, Sugiura S, Yano K, et al. The latent risk of acidosis in commercially available total parenteral nutrition (TPN) products: A randomized clinical trial in postoperative patients. J Clin Biochem Nutr 2009;45(1):68–73.
23. Mirza NS, Alfirevic A, Jorgensen A, et al. Metabolic acidosis with topiramate and zonisamide: An assessment of its severity and predictors. Pharmacogenet Genomics 2011;21:297–302.
24. Adrogues HJ. Mixed acid–base disturbances. J Nephrol 2006;19(Suppl 9):S97–S103.
25. Albert MS, Dell RB, Winters RW. Quantitative displacement of acid–base equilibrium in metabolic acidosis. Ann Intern Med 1964;66:312–322.
26. Chauhan V, Kelepouris E, Chauban N, et al. Current concepts and management strategies in chronic kidney disease—Mineral and bone disorders. South Med J 2012;1045:479–485.
27. Chadra V, Alon US. Hereditary renal tubular disorders. Semin Nephrol 2009;29(4):399–411.
28. Thomas KW, Schmidt GA. Alkanizing therapy in the management of acid–base disorder. In: Ronco C, ed. Critical Care Nephrology, 2nd ed. Philadelphia, PA: Elsevier, 2009:685–713.
29. Rachoin JS, Weisberg LS, McFadden CB. Treatment of lactic acidosis: Appropriate confusion. J Hosp Med 2010;5:E1–E7.
30. Cooper DJ, Walley KR, Wiggs BR, Russell JA. Bicarbonate does not improve hemodynamics in critically ill patients who have lactic acidosis: A prospective controlled clinical study. Ann Intern Med 1990;112:492–498.
31. Boyd JH, Walley KR. Is there a role for sodium bicarbonate in treating lactic acidosis from shock? Curr Opion Crit Care 2008;14:379–383.
32. Adrogue HJ, Rashad MN, Gorin AB, et al. Assessing acid–base status in circulatory failure: Differences between arterial and central venous blood. N Engl J Med 1989;320:1312–1316.
33. Kellum JA. Disorders of acid–base balance. Crit Care Med 2007;35(11):2630–2636.
34. Neumar RW, Otto CW, Link MS, et al. Part 8: Adult advanced cardiovascular life support: 2010 American Heart Association Guidelines for Cardiopulmonary Resuscitation and Emergency Cardiovascular Care. Circulation 2010;122:S729–S767.
35. McEvoy GK, Litvak K, Welsh OH, et al. American Hospital Formulary Service, Drug Information. Bethesda, MD: American Society of Health-System Pharmacists, 2012.
36. Geraci MJ, Klipa D, Heckman MG, et al. Prevalence of sodium bicarbonate-induced alkalemia in cardiopulmonary arrest patients. Ann Pharmacother 2009;43:1245–1250.
37. Hoste EA, Colpaert K, Vanholder RC, et al. Sodium bicarbonate versus THAM in ICU patients with mild metabolic acidosis. J Nephrol 2005;18(3):303–307.
38. Leung JM, Landow L, Franks M, et al. Safety and efficacy of intravenous Carbicarb in patients undergoing surgery: Comparison with sodium bicarbonate in the treatment of mild metabolic acidosis. Crit Care Med 1994;22:1540–1549.
39. Shapiro JI. Functional and metabolic responses of the isolated heart during acidosis: Effects of sodium bicarbonate and Carbicarb. Am J Physiol 1990;258:H1835–H1839.
40. Shapiro JI. Pathogenesis of cardiac dysfunction during metabolic acidosis: Therapeutic implications. Kidney Int 1997;51:47–51.
41. Bersin RM, Arieff AI. Improved hemodynamic function during hypoxia with Carbicarb, a new agent for the management of acidosis. Circulation 1988;77:227–233.
42. Stacpoole PW, Wright EC, Baumgartner TG, et al. A controlled clinical trial of dichloroacetate for treatment of lactic acidosis. N Engl J Med 1992;327:1564–1569.
43. Stacpoole PW, Nagaraja NV, Hutson AD. Efficacy of dichloroacetate as a lactate-lowering drug. J Clin Pharmacol 2003;43:683–691.
44. Vary TC, Siegel JH, Zechnich A, et al. Pharmacologic reversal of abnormal glucose regulation, BCAA utilization, and muscle catabolism in sepsis by dichloroacetate. J Trauma 1988;28:1301–1311.
45. Shangraw RE, Lohan-Mannion D, Hayes A, et al. Dichloroacetate stabilizes the intraoperative acid–base balance during liver transplantation. Liver Transpl 2008; 14(7):989–998.
46. Khanna A, Kurtzman NA. Metabolic alkalosis. J Nephrol 2006;19 (Suppl 9):S86–S96.
47. Gennari FJ. Pathophysiology of metabolic alkalosis: A new classification based on the centrality of stimulated collecting duct ion transport. Am J Kidney Dis 2011;58:626–636.
48. Mikhalidis G, Mikhailidis DP, Elisaf M. Acid–base and electrolyte abnormalities observed in patients receiving cardiovascular drugs. J Cardiovasc Pharmacol Ther 2003; 8:267–279.
49. Barton CH, Vaziri ND, Ness RL, et al. Cimetidine in the management of metabolic alkalosis induced by nasogastric drainage. Arch Surg 1979;1:70–74.
50. Mazur JE, Devlin JW, Peters MJ, et al. Single versus multiple doses of acetazolamide for metabolic alkalosis in critically ill medical patients: A randomized, double-blind trial. Crit Care Med 1999;27(7):1257–1261.
51. Rowlands BJ, Tindall SF, Elliot DJ. The use of dilute hydrochloric acid and cimetidine to reverse severe metabolic alkalosis. Postgrad Med J 1978;54:118–123.
52. Mirtallo JM, Rogers KR, Johnson JA, et al. Stability of amino acids and the availability of acid in total parenteral nutrition solutions containing hydrochloric acid. Am J Hosp Pharm 1981;38:1729–1731.
53. Graziani G, Fedeli C, Moroni L, Cosmai L, Badalamenti S, Ponticelli C. Gitelman syndrome: Pathophysiological and clinical aspects. QJM 2010;103:741–748.
54. Hene RJ, Koomans HA, Dorhout Mees EJ, et al. Correction of hypokalemia in Bartter’s syndrome by enalapril. Am J Kidney Dis 1987;9:200–205.
55. Vinci JM, Gill JR Jr, Bowden RE, et al. The Kallikrein-Kinin system in Bartter’s syndrome and its response to prostaglandin synthetase inhibition. J Clin Invest 1987;61:1671–1682.
56. Effros RM, Swenson ER. Acid–base balance. In: Mason RJ, Broaddus CV, Martin TR, et al. Murray & Nadel’s Textbook of Respiratory Medicine, 5th ed. Philadelphia, PA: Saunders/Elsevier, 2010.
57. Foster GT, Vaziri ND, Sassoon CS. Respiratory alkalosis. Respir Care 2001;46:384–391.