Chapter 5 Disorders of Potassium

Hypokalemia and Hyperkalemia

Stephen P. DiBartola, Helio Autran De Morais

Potassium is the major intracellular cation in mammalian cells, whereas sodium is the major extracellular cation. Normally, the extracellular fluid (ECF) sodium concentration is approximately 140 mEq/L, and the ECF potassium concentration is approximately 4 mEq/L. This relationship is reversed in intracellular fluid (ICF), in which the sodium concentration is approximately 10 mEq/L and the potassium concentration is approximately 140 mEq/L. In experimental studies of dogs, control values for ICF sodium and potassium concentrations in skeletal muscle were 8.4 to 13.7 and 139 to 142 mEq/L, respectively.^20,¹⁰⁷

Total body potassium content in humans is approximately 50 to 55 mEq/kg body weight, and almost all of this potassium is readily exchangeable.^6,⁷⁰ In one study of potassium depletion in dogs, the control value for total exchangeable potassium as determined by ⁵⁰K dilution was 47.1 mEq/kg body weight (range, 39.8 to 61.1 mEq/kg).¹ In cats, total body potassium is approximately 55 mEg/kg body weight.^184a As much as 95% or more of total body potassium is located within cells, with muscle containing 60% to 75% of this potassium. Muscle potassium content in normal dogs and cats is approximately 400 mEq/kg.^20,^107,^147,¹⁹¹ As a solute, intracellular potassium is crucial for maintenance of normal cell volume. Intracellular potassium also is important for normal cell growth because it is required for the normal function of enzymes responsible for nucleic acid, glycogen, and protein synthesis.

The remaining 5% of the body’s potassium is located in the ECF. Maintaining the ECF potassium concentration within narrow limits is critical to avoid the life-threatening effects of hyperkalemia on cardiac conduction. In humans, the serum potassium concentration is inversely correlated with the total body deficit of potassium (Fig. 5-1). Likewise, in dogs with potassium depletion induced by dietary restriction, the muscle potassium content was strongly correlated (r = 0.87) with the serum potassium concentration.¹⁴⁷ During translocation of potassium between ICF and ECF, however, the serum potassium concentration can change without any change in the total body potassium content. One of the most important functions of potassium in the body is its role in generation of the normal resting cell membrane potential.

Figure 5-1 Relationship of serum potassium concentration to bodily potassium deficit. The data are derived from seven metabolic balance studies carried out on 24 human subjects depleted of potassium.

(From Sterns RH et al.: Medicine 60:339, 1981.)

The resting cell membrane potential

The normal relationship between ECF and ICF potassium concentrations is maintained by sodium, potassium-adenosinetriphosphatase (Na⁺, K⁺-ATPase) in cell membranes. This enzyme pumps sodium ions out of, and potassium ions into, the cell in a 3:2 Na/K ratio so that the intracellular concentration of potassium is much higher than its extracellular concentration. As a result, K⁺ ions diffuse out of the cell down their concentration gradient. However, the cell membrane is impermeable to most intracellular anions (e.g., proteins and organic phosphates). Therefore, a net negative charge develops within the cell as K⁺ ions diffuse out, and a net positive charge accumulates outside the cell. Consequently, a potential difference is generated across the cell membrane.

The principal extracellular cation is sodium, and it enters the cell relatively slowly down its concentration and electrical gradients, because the permeability of the cell membrane to potassium is 100-fold greater than its permeability to sodium. Diffusion of K⁺ ions from the cell continues until the ECF acquires sufficient positive charge to prevent further diffusion of K⁺ ions out of the cell. The ratio of the intracellular to extracellular concentrations of potassium ([K⁺]_I/[K⁺]_O) is the major determinant of the resting cell membrane potential as described by the Nernst equation:

The Goldman-Hodgkin-Katz equation is a modification of the Nernst equation that allows prediction of E_m based on the ionic permeability characteristics of the cell membrane to sodium and potassium and the concentrations of these ions inside and outside the cell:

where P_Na and P_K are the membrane permeabilities for sodium and potassium. The term r is included in the equation to account for the effect of the electrogenic Na⁺, K⁺-ATPase pump under steady-state conditions. This term is assigned the Na/K transport ratio of 3:2 so that r = 1.5. If the membrane permeability for potassium is assigned a value of 1.0 and the cell membrane is 100 times more permeable to potassium than sodium:

For example, using the hypothetical ECF and ICF concentrations of sodium and potassium given at the beginning of this chapter:

In one study of dogs with potassium deficiency, the predicted E_m was −86.6 mV and the measured E_m in skeletal muscle of control animals was −90.1 mV.²⁰ The resting cell membrane potential plays a vital role in the normal function of skeletal and cardiac muscle, nerves, and transporting epithelia.

The threshold cell membrane potential

The threshold cell membrane potential is reached when sodium permeability increases to the point that sodium entry exceeds potassium exit, depolarization becomes self-perpetuating, and an action potential develops. The ability of specialized cells to develop an action potential is crucial to normal cardiac conduction, muscle contraction, and nerve impulse transmission. The excitability of a tissue is determined by the difference between the resting and threshold potentials (the smaller the difference, the greater the excitability).

Hypokalemia increases the resting potential (i.e., makes it more negative) and hyperpolarizes the cell, whereas hyperkalemia decreases the resting potential (i.e., makes it less negative) and initially makes the cell hyperexcitable (Fig. 5-2). If the resting potential decreases to less than the threshold potential, depolarization results, repolarization cannot occur, and the cell is no longer excitable. Translocation of potassium between body compartments results in a greater change in the ratio of intracellular to extracellular potassium concentrations ([K⁺]_I/[K⁺]_O) than does a change in total body potassium. In the former instance, the potassium concentrations of the two compartments change in opposite directions, whereas in the latter instance, they change in the same direction.

Figure 5-2 Effects of serum calcium and potassium on membrane potentials of excitable tissues. The concentration of potassium in extracellular fluids affects the resting potential, whereas calcium concentrations alter the threshold potential.

(From Leaf A, Cotran R. Renal pathophysiology. New York: Oxford University Press, 1976: 116.)

Membrane excitability also is affected by ionized calcium concentration and acid-base balance. Calcium affects the threshold potential rather than the resting potential. Ionized hypocalcemia increases membrane excitability by allowing self-perpetuating sodium permeability to be reached with a lesser degree of depolarization, whereas ionized hypercalcemia requires greater than normal depolarization for this threshold to be reached (see Fig. 5-2). Thus, hypercalcemia counteracts hyperkalemia by normalizing the difference between the resting and threshold potentials, whereas hypocalcemia exacerbates the effect of hyperkalemia on membrane excitability. This principle is the basis for treating hyperkalemia with calcium salts (see the Treatment of Hyperkalemia section). Membrane excitability is increased by alkalemia and decreased by acidemia. As a result of these factors, clinical signs are not necessarily correlated with serum potassium concentrations. Electrocardiographic findings and muscle strength reflect the functional consequences of abnormalities in serum potassium concentration.

Potassium balance

External potassium balance

External balance for potassium is maintained by matching output (primarily in urine) to input (from the diet). In the normal animal, potassium enters the body only through the gastrointestinal tract, and virtually all ingested potassium is absorbed in the stomach and small intestine. Transport of potassium in the small intestine is passive, whereas active transport (responsive to aldosterone) occurs in the colon. Colonic secretion of potassium may play an important role in extrarenal potassium homeostasis in some disease states (e.g., chronic renal failure) (Fig. 5-3).

Figure 5-3 Relationship between the degree of renal insufficiency and fecal potassium excretion. Data points are compiled from three studies comprising 98 balance periods in 40 human patients. Variation in dietary protein or sodium intake did not produce consistent changes in fecal potassium excretion; thus, data points from these balance periods were included without special designation.

(From Alexander EA, Perrone RD. Regulation of extrarenal potassium metabolism. In: Maxwell MH, Kleeman CR, Narins RG, editors. Clinical disorders of fluid and electrolyte metabolism, 4th ed. New York: McGraw-Hill, 1987: 105–117, with permission of the McGraw-Hill Companies.)

Potassium derived from the diet and endogenous cellular breakdown is removed from the body primarily by the kidneys and, to a much lesser extent, by the gastrointestinal tract. During zero balance, 90% to 95% of ingested potassium is excreted in urine, and the remaining 5% to 10% is excreted via the gastrointestinal tract. This pattern of output has been observed during control balance studies in normal dogs.^12,^139,^152,^169,¹⁸² In a study of renal handling of potassium in dogs, 90% to 98% of potassium intake was eliminated from the body by the kidneys.²⁴

Adaptation occurs during chronic potassium loading so that the animal is protected from hyperkalemia that could occur as a result of an acute potassium load. This effect results from enhanced renal and colonic excretion of potassium, as well as from enhanced uptake of potassium by the liver and muscle, mediated by the effects of insulin and catecholamines. Potassium deprivation is associated with decreased aldosterone secretion, suppression of potassium secretion in the distal nephron, and increased reabsorption of potassium in the inner medullary collecting ducts. Skeletal muscle potassium concentration decreases, but brain and heart potassium concentrations are minimally affected during potassium depletion.^20,^107,¹⁷⁸ The colon adapts to potassium deprivation by decreasing its secretion of potassium.

Internal potassium balance

Internal balance for potassium is maintained by translocation of potassium between ECF and ICF. One half to two thirds of an acute potassium load appears in the urine within the first 4 to 6 hours, and effective translocation of potassium from ECF to ICF is crucial in preventing life-threatening hyperkalemia until the kidneys have sufficient time to excrete the remainder of the potassium load. Endogenous insulin secretion and stimulation of β₂-adrenergic receptors by epinephrine promote cellular uptake of potassium in the liver and muscle by increasing the activity of cell membrane Na⁺, K⁺-ATPase. The main effects of these hormones are to facilitate distribution of an acute potassium load and not to mediate minor adjustments in serum potassium concentration. The ECF concentration of potassium itself plays an important role in translocation because potassium movement into cells is facilitated by the change in chemical concentration gradient resulting from addition of potassium to ECF. The fraction of an acute potassium load taken up by the body is increased during chronic potassium depletion and decreased when total body potassium is excessive. In summary, any change in serum potassium concentration must arise from a change in intake, distribution, or excretion (Fig. 5-4).

Figure 5-4 Components of potassium homeostasis. ECF, Extracellular fluid; ICF, intracellular fluid.

(Drawing by Tim Vojt.)

Effect of acid-base balance on potassium distribution

The effect of acute pH changes on translocation of potassium between ICF and ECF is complex. In general, acidosis is associated with movement of potassium ions from ICF to ECF, and alkalosis is associated with movement of potassium ions from ECF to ICF. Early animal studies and observations in a small number of human patients led to the prediction that acute metabolic acidosis would be associated with a 0.6-mEq/L increment in serum potassium concentration for each 0.1-U decrement in pH. This rule of thumb has circulated widely among clinicians.^31,^181,¹⁸⁸

However, a critical review of experimental studies in animals and humans demonstrated that changes in serum potassium concentration during acute acid-base disturbances were quite variable.⁴ The change in serum potassium concentration was greatest during acute mineral acidosis. In dogs, the increase in serum potassium concentration after administration of a mineral acid (e.g., HCl or NH₄Cl) was very variable, ranging from a 0.17- to 1.67-mEq/L increment in serum potassium concentration per 0.1-U decrement in pH (mean, 0.75 mEq/L). The increment in serum potassium concentration during acute respiratory acidosis in dogs was much lower, averaging only 0.14 mEq/L per 0.1-U decrement in pH. The decrement in serum potassium concentration during metabolic alkalosis in dogs averaged 0.18 mEq/L per 0.1-U increment in pH, whereas it averaged 0.27 mEq/L per 0.1-U increment in pH during respiratory alkalosis. In another study, respiratory alkalosis induced by hyperventilation in anesthetized dogs caused a somewhat greater decrement in serum potassium concentration (0.4 mEq/L) for each 0.1-U increment in pH.¹³⁶ An increase in serum potassium concentration did not occur in acute metabolic acidosis caused by organic acids (e.g., lactic acid and ketoacids).* Acute infusion of β-hydroxybutyric acid in normal dogs caused an increase in insulin in portal venous blood and hypokalemia, presumably as a result of potassium uptake by cells. Conversely, acute infusion of HCl led to increased portal vein glucagon concentration and hyperkalemia, possibly caused by potassium release from cells.⁵ In summary, only mineral acidosis is expected to cause any clinically relevant change in serum potassium concentration during acute acid-base disturbances.

Many factors probably contribute to the variable changes observed in serum potassium concentration during acute acid-base disturbances, including blood pH and HCO₃⁻ concentration, nature of the acid anion (mineral versus organic), osmolality, hormonal activity (e.g., catecholamines, insulin, glucagon, and aldosterone), and the metabolic and excretory roles of the liver and kidneys.⁴ Hyperosmolality and lack of insulin are more likely to be responsible for hyperkalemia observed in patients with diabetic ketoacidosis than is the acidosis itself.

Hyperkalemia associated with acute metabolic acidosis induced by mineral acids is transient. In a study of acute and chronic metabolic acidosis induced in dogs by administration of HCl or NH₄Cl, hyperkalemia was observed after acute infusion of HCl, but hypokalemia developed after 3 to 5 days of NH₄Cl administration.¹²² The observed hypokalemia was associated with inappropriately high urinary excretion of potassium and increased plasma aldosterone concentration.¹²² Similar findings have been reported in rats with chronic metabolic acidosis induced by NH₄Cl. Despite a total body deficit of potassium, rats with chronic metabolic acidosis did not conserve potassium appropriately.¹⁷⁰ This effect may be caused by a decreased filtered load of HCO₃⁻, increased distal delivery of sodium, and increased distal tubular flow. Thus, metabolic acidosis of at least 2 to 3 days’ duration is associated with increased urinary potassium excretion and mild hypokalemia rather than hyperkalemia.⁷⁹

Renal handling of potassium

The kidneys are the primary regulators of potassium balance. Potassium is filtered at the glomerulus, and approximately 70% of the filtered load is reabsorbed isosmotically with water and sodium in the proximal tubule. An additional 10% to 20% of filtered potassium is reabsorbed in the ascending limb of Henle’s loop. Finally, 10% to 20% of the filtered load is delivered to the distal nephron, where final adjustments in potassium reabsorption and secretion are made. Potassium experiences either net reabsorption or secretion in the connecting tubule, cortical collecting duct, and first portion of the outer medullary collecting duct, depending on the body’s needs. Net movement of potassium in these segments of the nephron determines urinary excretion of potassium. Potassium once again experiences reabsorption in the last portion of the outer medullary collecting duct and inner medullary collecting duct regardless of the body’s needs.

Mechanisms of renal tubular transport of potassium

The transepithelial electrical potential difference is lumen negative in the early proximal tubule, but no active transport mechanism for potassium has been discovered in this segment of the nephron. In the proximal tubule, potassium is reabsorbed along with water by solvent drag via the paracellular route. Apparently, water reabsorption increases the luminal concentration of potassium enough to overcome the unfavorable transepithelial potential difference. The transepithelial electrical potential difference becomes lumen positive in the late proximal tubule, and this facilitates reabsorption of potassium by the paracellular route. Transcellular transport of potassium in the proximal tubular cells occurs by means of potassium channels in both luminal and basolateral membranes and by a K⁺-Cl⁻ cotransporter in basolateral membranes (Fig. 5-5).

Figure 5-5 Renal tubular transport mechanisms for potassium in the proximal tubule.

TEPD, Transepithelial potential difference. (Drawing by Tim Vojt.)

In the thick ascending limb of the Henle loop, the transepithelial electrical potential difference is strongly lumen positive, and most potassium reabsorption occurs by the paracellular route. Potassium channels in the luminal membranes allow potassium to exit the cell down its concentration gradient and facilitate the electrochemical gradient for potassium reabsorption via the paracellular route. Transcellular reabsorption of potassium is facilitated by the luminal Na⁺-K⁺-2Cl⁻ cotransporter and by potassium channels and a K⁺-Cl⁻ cotransporter in the basolateral membranes (Fig. 5-6).

Figure 5-6 Renal tubular transport mechanisms for potassium in the thick ascending limb of Henle’s loop. TEPD, Transepithelial potential difference.

(Drawing by Tim Vojt.)

The mechanisms of renal potassium handling in the distal convoluted tubule are shown in Figure 5-7. The thiazide-sensitive Na⁺-Cl⁻ cotransporter and the K⁺-Cl⁻ cotransporter in the luminal membranes of these tubular cells result in secretion of potassium and reabsorption of sodium while chloride is recycled across the luminal membrane. The basolateral Na⁺, K⁺-ATPase maintains a low intracellular concentration of sodium and a high intracellular concentration of potassium that facilitate sodium reabsorption and potassium secretion across the luminal membranes.

Figure 5-7 Renal tubular transport mechanisms for potassium in the distal convoluted tubule (early distal tubule).

TEPD, Transepithelial potential difference. (Drawing by Tim Vojt.)

Principal cells are found in the connecting tubule and collecting duct and are responsible for potassium secretion. The basolateral membranes of principal cells are rich in Na⁺, K⁺-ATPase, which maintains a high intracellular potassium concentration. The luminal membranes of the principal cells contain an electrogenic sodium channel (ENaC). This sodium channel is directly blocked by the diuretics amiloride and triamterene, whereas spironolactone antagonizes the effect of aldosterone on the channel. Sodium movement through this channel renders the tubular lumen negative, and the resultant increase in lumen electronegativity facilitates secretion of K⁺ ions through luminal K⁺ channels (Fig. 5-8).

Figure 5-8 Renal tubular transport mechanisms for potassium in the principal cells of the late distal tubule and collecting duct.

TEPD, Transepithelial potential difference. (Drawing by Tim Vojt.)

There are two types of intercalated cells in the distal nephron. Type A or α intercalated cells contain H⁺-ATPase and H⁺, K⁺-ATPase in their luminal membranes and Cl⁻– HCO₃⁻ countertransporters and Cl⁻ and K⁺ channels in their basolateral membranes. They also contain carbonic anhydrase. This arrangement allows the intercalated cell to secrete H⁺ ions and reabsorb K⁺ and HCO₃⁻ ions. Potassium is actively transported across the luminal membranes of type α intercalated cells by H⁺, K⁺-ATPase and then diffuses down its concentration gradient through potassium channels in the basolateral membranes (Fig. 5-9). Type A and α intercalated cells are found in the connecting tubule, cortical collecting duct, and outer medullary collecting duct. Type B and β intercalated cells are found only in the cortical collecting ducts and secrete HCO₃⁻ ions. They are able to do so because their polarity is reversed as compared with type α intercalated cells (i.e., the H⁺-ATPase is in the basolateral membrane, and the Cl⁻– HCO₃⁻ countertransporter is in the luminal membrane).

Figure 5-9 Renal tubular transport mechanisms for potassium in the α intercalated cells of the late distal tubule and collecting duct.

CA, Carbonic anhydrase; TEPD, Transepithelial potential difference. (Drawing by Tim Vojt.)

Potassium is reabsorbed from the last portion of the outer medullary collecting duct and throughout the inner medullary collecting duct. In these segments of the nephron, potassium is reabsorbed by the paracellular route despite a lumen-negative transepithelial potential difference, because reabsorption of water increases the chemical concentration gradient sufficiently to overcome the unfavorable electrical gradient.

Determinants of urinary potassium excretion

Three main factors affect potassium secretion in the distal nephron: the magnitude of the chemical concentration gradient for potassium between the tubular cells and tubular lumen, the tubular flow rate, and the transmembrane potential difference across the luminal membranes of the tubular cells. Gastrointestinal absorption of a potassium load increases the ECF concentration of potassium. This results in an increase in the number of K⁺ ions available for uptake at the basolateral membranes of the distal tubular cells by Na⁺, K⁺-ATPase, and the resulting increase in intracellular potassium concentration increases the chemical concentration gradient for diffusion of K⁺ ions out of the tubular cells across their luminal membranes.

Aldosterone is the most important hormone affecting urinary potassium excretion. Its secretion by the zona glomerulosa of the adrenal gland is stimulated directly by hyperkalemia and angiotensin II (produced in response to volume depletion), whereas adrenocorticotropic hormone (ACTH), hyponatremia, and decreased extracellular pH play permissive roles in promoting aldosterone secretion. Aldosterone release is inhibited by dopamine and atrial natriuretic factor, both of which are released in response to volume expansion.

Aldosterone increases reabsorption of Na⁺ and secretion of K⁺ and H⁺ ions in the distal nephron. Its primary effect is to increase the number of open Na⁺ channels in the luminal membranes of the principal cells. Sodium reabsorption via these luminal Na⁺ channels is electrogenic (i.e., it generates electronegativity in the tubular lumen). This electronegativity can be dissipated either by K⁺ or H⁺ ion secretion or by Cl⁻ reabsorption in the distal nephron. Aldosterone increases the activity and number of Na⁺, K⁺-ATPase pumps in the basolateral membranes of the principal cells, and this effect may occur as a result of increased entry of Na⁺ ions across the luminal membranes. Increased Na⁺, K⁺-ATPase activity in turn increases the intracellular K⁺ concentration and facilitates K⁺ secretion across the luminal membranes. Aldosterone also increases the number of open K⁺ channels in the luminal membrane, thus facilitating K⁺ exit into tubular fluid.

Aldosterone can influence H⁺ secretion in two ways. It directly promotes H⁺ ion secretion in H⁺-secreting type α intercalated cells by stimulation of the H⁺-ATPase present in their luminal membranes. Aldosterone also promotes H⁺ secretion in the distal tubule by stimulating electrogenic Na⁺ reabsorption in principal cells and increasing lumen electronegativity, which favors enhanced H⁺ secretion.

An increase in distal tubular flow enhances potassium secretion by rapidly moving secreted K⁺ ions downstream and providing new tubular fluid from upstream in the nephron. This allows maintenance of a high chemical concentration gradient for potassium secretion and provides a “sink” for movement of K⁺ ions into tubular fluid. A decrease in distal tubular flow has the opposite effect and promotes dissipation of the chemical gradient for diffusion of K⁺ ions from principal cells into tubular fluid.

Lumen electronegativity is generated by sodium reabsorption through Na⁺ channels in the luminal membranes of principal cells. Normally, some of this electronegativity is dissipated by passive Cl⁻ reabsorption. If a large concentration of a relatively nonresorbable anion (e.g., SO₄²⁻, HCO₃⁻, penicillin) is present in distal tubular fluid, less dissipation of the electronegativity occurs, and K⁺ secretion is enhanced. This factor contributes to the pathophysiology of metabolic alkalosis. In this setting, there is less Cl⁻and more HCO₃⁻ in the distal tubular fluid, and HCO₃⁻ is relatively nonresorbable in the cortical collecting duct. This is one reason metabolic alkalosis promotes urinary K⁺ excretion. Amiloride is a diuretic that impairs luminal Na⁺ entry into principal cells by decreasing the number of open Na⁺ channels. This in turn reduces lumen electronegativity and impairs K⁺ secretion. Thus, the magnitude of distal tubular lumen electronegativity has an important effect on urinary K⁺ excretion.

The Na⁺ and Cl⁻ concentrations of distal tubular fluid usually have little effect on K⁺ secretion. When the luminal Na⁺ concentration is very low (<25 to 35 mEq/L), however, diffusion of Na⁺ ions into distal tubular cells may be impaired sufficiently to produce an increase in the tubular cell transmembrane potential (making the cell interior more negative) and impeding diffusion of K⁺ ions from the cell into the tubular lumen.^85,^204,²⁰⁵ Extremely low luminal Cl⁻ concentrations (<10 mEq/L) may increase net potassium secretion, possibly because some fraction of K⁺ reabsorption or secretion may be accomplished by K⁺-Cl⁻ cotransport.¹⁹⁸ Such a mechanism may also play a role in the pathophysiology of enhanced urinary K⁺ excretion during metabolic alkalosis. Antidiuretic hormone (ADH) helps minimize disruption of potassium balance during water deprivation by increasing the number of open luminal K⁺ channels in principal cells and facilitating potassium excretion at a time when distal tubular flow is reduced.^36,^69,¹⁸³ Conversely, potassium excretion is not necessarily increased despite increased distal tubular flow during water diuresis because ADH is suppressed. Major factors affecting renal excretion of potassium are summarized in Figure 5-10.

Figure 5-10 Factors affecting urinary excretion of potassium.

(Drawing by Tim Vojt.)

Factors influencing renal potassium excretion

Normal serum concentrations

Ion-selective potentiometry and flame photometry are methods used by clinical laboratories to measure sodium and potassium concentrations in body fluids. Electrolytes in plasma are excluded from the fraction of plasma (normally about 7%) that is occupied by solids (e.g., lipids and proteins) and are confined to the aqueous phase of plasma (about 93% of total plasma volume). Flame photometry and indirect potentiometry are affected by the exclusion of electrolytes from the fraction of plasma that is occupied by solids, whereas direct potentiometry is not.¹⁹⁴ The resulting error is usually small, but for serum sodium concentration it may be clinically relevant in patients with hyperlipemia (see Chapter 3). Potassium is present in ECF at a much lower concentration than sodium, and the effect of hyperlipemia on the measured serum potassium concentration is much less apparent.

Normal values for serum potassium concentration in dogs and cats vary slightly among laboratories but are expected to be 3.5 to 5.5 mEq/L, with an average value of approximately 4.5 mEq/L. Serum potassium concentrations exceed plasma concentrations because potassium is released from platelets during the clotting process. There is a positive correlation between platelet count and serum potassium concentration in dogs.^49,¹⁵⁵ The difference between serum and plasma potassium concentrations is most pronounced in animals with thrombocytosis.^50,^124,¹⁵⁵ In one study, serum potassium concentration was greater than plasma potassium concentration by a mean of 0.63 mEq/L in dogs with normal platelet counts and by a mean of 1.55 mEq/L in dogs with thrombocytosis.¹⁵⁵

The potassium content of erythrocytes varies in mammalian species, and hemolysis can result in hyperkalemia in species that have high red cell potassium concentrations (Table 5-1). Normal adult canine and feline red cells usually contain potassium in concentrations similar to those of plasma, and hemolysis is not associated with hyperkalemia.^42,^48,^65,^93,¹⁵³ In one study, storage of canine red cells in citrate-phosphate-dextrose-adenine for 40 days resulted in an increase in plasma potassium concentration from 5 to almost 9 mEq/L despite the fact that the original intracellular potassium concentration in the red cells was only 3.8 mEq/L.¹⁵³ Regardless of the underlying mechanism, this magnitude of increase in plasma potassium concentration would be unlikely to result in detectable hyperkalemia in a recipient dog transfused with blood stored in this manner.

Table 5-1 Sodium and Potassium Concentrations of Mammalian Erythrocytes

Species	Sodium (mEq/L)	Potassium (mEq/L)
Human	10-21	104-155
Dog LK*	93-150	4-11
Dog HK*¹²¹	54	124
Cat	104-142	6-8
Horse	4-16	80-140
Cow LK*	72-102	7-37
Cow HK*	15	70
Sheep LK*	74-121	8-39
Sheep HK*	10-43	60-88
Swine	11-19	100-124

HK, High potassium; LK, low potassium.

* Sheep, cattle, and dogs demonstrate polymorphism with respect to their intracellular cation concentrations, depending on the level of Na⁺, K⁺ -ATPase activity in the mature red cell membranes.

The potassium concentrations of red cells from neonatal dogs are higher than those of red cells from adult dogs.^42,^132,¹⁴⁵ Red cell concentrations of potassium decrease during the first weeks of life and reach normal adult concentrations by approximately 8 to 13 weeks of age. In one study, mean red cell potassium concentrations in puppies were 19.0 mEq/L at 1 day of age, 15.1 mEq/L at 5 weeks of age, and 8.7 mEq/L at 13 weeks of age.⁴² Reticulocytes from adult dogs also contain higher potassium concentrations than do mature red cells.¹²⁰ In adult Akitas, red cell potassium concentrations may exceed 70 mEq/L, and hemolysis results in a progressive increase in plasma potassium concentration (up to 24 mEq/L) during storage of blood.^48,¹⁵⁸

Dogs may be divided genetically into two groups based on the presence or absence of Na⁺, K⁺-ATPase activity in the membranes of their mature red cells.^99,¹²¹ Dogs with red cell membrane Na⁺, K⁺-ATPase activity maintain high intracellular potassium concentrations, whereas those without red cell Na⁺, K⁺-ATPase activity maintain red cell potassium concentrations similar to those of plasma. Reticulocytes from low-potassium (LK) dogs possess Na⁺, K⁺-ATPase, but it is rapidly and completely degraded by a proteolytic process during cell maturation.^100,¹²⁰ Reticulocytes from high-potassium (HK) dogs have twice as much Na⁺, K⁺-ATPase activity as reticulocytes from LK dogs, but in the HK dogs, degradation of the enzyme ceases early in maturation, and sufficient activity remains in the mature red cell to account for the observed high intracellular concentration of potassium.¹²⁰ The HK phenotype is inherited as an autosomal recessive trait and occurs with an incidence of 26% to 38% in the Shiba and Akita breeds in Japan and 42% in the Jindo breed in Korea.⁷⁶ The HK phenotype also may be seen in the Chinese shar-pei breed and can cause pseudohyperkalemia.¹⁶ Some dogs with the HK phenotype also accumulate large amounts of reduced glutathione in their erythrocytes (so-called HK/HG phenotype), which predisposes them to oxidative injury and hemolytic anemia associated with onion ingestion.²⁰⁹

Red cells of English springer spaniel dogs with phosphofructokinase deficiency had potassium concentrations of 19.2 to 28 mEq/L as compared with 5.1 to 7.7 mEq/L in control dogs, and hemolytic crises in affected dogs were associated with hyperkalemia.⁸³ The higher potassium concentration in the red cells of affected dogs was attributed in part to the large number of circulating reticulocytes (7% to 26%). The same mutation has been reported to cause phosphofructokinase deficiency and hyperkalemia in whippets.⁸⁰

Hemolysis in Akitas (and presumably in other HK dogs) and thrombocytosis cause what has been called pseudohyperkalemia because these effects occur in vitro. Pseudohyperkalemia also has been reported in a dog with acute lymphoblastic leukemia before chemotherapy.⁹⁵ Leakage of potassium from the leukemic cells in vitro was thought to be responsible for pseudohyperkalemia in this case. Use of plasma from small blood samples collected in an excessive volume of tripotassium ethylenediaminetetraacetic acid also may result in measured hyperkalemia.

Hypokalemia

Clinical and laboratory features

Many dogs and cats with hypokalemia have no clinical signs. Muscular weakness, polyuria, polydipsia, and impaired urinary concentrating capacity are the clinical signs most likely to be recognized in dogs and cats with symptomatic hypokalemia. The pathophysiology of these clinical signs is discussed here.

The clinician should verify the abnormal serum potassium concentration with the laboratory, but measurement of potassium by flame photometry and ion-selective potentiometry is reliable, and errors are uncommon. The clinical history often provides information about the likely source of potassium loss (e.g., chronic vomiting and diuretic administration) or the possibility of translocation (e.g., insulin administration and alkalosis).

Determination of the fractional excretion of potassium (FE_K) may help differentiate between renal and nonrenal sources of potassium loss. Fractional potassium excretion can be calculated and expressed as a percentage using:

where U_K is the urine concentration of K (mEq/L), S_K is the serum concentration of K (mEq/L), U_Cr is the urine concentration of creatinine (mg/dL), and S_Cr is the serum concentration of creatinine (mg/dL).

The FE_K should be less than 4% for nonrenal sources of loss, and in the presence of hypokalemia, values above 4% may indicate inappropriate renal loss.⁶⁰ In one study, however, FE_K values for normal cats were 10.6 ± 2.1%.³ In another study of normal cats receiving a potassium-deficient diet, FE_K values decreased from 10% to 12% to 3% to 6%.⁶¹ Thus, FE_K values up to 6% should probably be considered normal in potassium-depleted animals with normal renal function. However, the clinical utility of FE_K calculations is limited by the fact that FE_K does not correlate well with 24-hour urinary excretion of potassium.^3,⁷⁴ The occurrence of hypokalemia in patients with metabolic alkalosis suggests vomiting of stomach contents or diuretic administration as likely causes of potassium loss. In patients with hypokalemia and metabolic acidosis, diarrhea caused by small intestinal disease, chronic renal failure, and distal renal tubular acidosis are more likely causes of potassium loss (Fig. 5-11).