Chapter 7

Disorders of Potassium
Metabolism
The diagnosis and management of potassium disorders require
a clear appreciation of the underlying physiology of potassium
homeostasis. This chapter reviews aspects of the physiology of
potassium homeostasis judged to be most pertinent to bedside
management; a more detailed review is provided in Chapter
5, Transport of Inorganic Solutes: Sodium, Chloride, Potas-
sium, Magnesium, Calcium and Phosphate, in Brenner and
Rector’s The Kidney.

NORMAL POTASSIUM BALANCE

Despite the wide variation in dietary potassium intake (<35 to

>110 mmol/day), homeostatic mechanisms precisely maintain
serum Kþ levels between 3.5 and 5 mmol/L. In a healthy individ-
ual at steady state, the entire daily intake of potassium is excreted
(urine 90%, stool 10%). More than 98% of total body potassium
is intracellular, located mainly in muscle ("75% within skeletal
muscle). Buffering of extracellular Kþ by this large intracellular
pool plays a crucial role in the regulation of serum Kþ levels.
The rapid exchange of intracellular Kþ with extracellular Kþ is
accomplished by overlapping and synergistic regulation of a
number of cellular and renal transport pathways.

Potassium Transport Mechanisms

The intracellular accumulation of Kþ against its electrochemi-
cal gradient is an energy-consuming process, mediated pri-
marily by the Naþ/Kþ-ATPase enzyme, which exchanges
three intracellular Naþ ions for two extracellular Kþ ions.
Changes in skeletal muscle Naþ/Kþ-ATPase activity and abun-
dance are major determinants of the capacity for extrarenal Kþ
homeostasis. Hypokalemia induces a marked decrease in
Naþ/Kþ-ATPase activity and muscle Kþ content helping main-
tain the serum Kþ within the normal range. In contrast, hyper-
kalemia due to potassium loading is associated with adaptive
increases in Naþ/Kþ-ATPase activity and muscle Kþ content.
Potassium can also accumulate in cells by coupling to the gra-
dient for Naþ entry via the electroneutral Naþ-Kþ-2Cl#
136
cotransporters NKCC1 and NKCC2. The efflux of Kþ out of 137
cells is largely achieved by Kþ channels, which make up the
largest family of ion channels in the human genome, and are
increasingly implicated in the control of potassium
homeostasis. CH 7

Disorders of Potassium Metabolism

Factors Affecting Internal Distribution of
Potassium

Several hormones and physiologic conditions have acute

effects on the distribution of Kþ between the intracellular
and extracellular space (Table 7-1).

Insulin
Insulin and potassium appear to form a loop-feedback mecha-
nism, as increases in serum potassium independently stimu-
late insulin secretion. Insulin promotes a fall in serum Kþ by

Factors Affecting Kþ Distribution between

Table 7-1
Intra- and Extracellular Compartments
Acute
Factor Effect on Potassium
Insulin Enhanced cell uptake
b-Catecholamines Enhanced cell uptake
a-Catecholamines Impaired cell uptake
Acidosis Impaired cell uptake
Alkalosis Enhanced cell uptake
External potassium balance Loose correlation
Cell damage Impaired cell uptake
Hyperosmolality Enhanced cell efflux
Chronic
Factor Effect on ATP Pump Density
Thyroid Enhanced
Adrenal steroids Enhanced
Exercise (training) Enhanced
Growth Enhanced
Diabetes Impaired
Potassium deficiency Impaired
Chronic renal impairment Impaired
From Giebisch G: Renal potassium transport: Mechanisms and regulation. Am J
Physiol 274:F817–F833, 1998.
138 stimulating the uptake of Kþ by the liver, skeletal and cardiac
muscle, and fat via activation of several Kþ transport path-
ways, including activation of Naþ/Kþ-ATPase, Naþ/Hþ
exchange, and Naþ-Kþ-2Cl# and Kþ-Cl#cotransport. The
II impact of insulin on plasma potassium and glucose appears
to be mediated by independent mechanisms.
Disturbances in Control of Body Fluid Volume and Composition

Sympathetic Nervous System

Uptake of Kþ by liver and muscle is stimulated through b2-
adrenergic receptors and is largely independent of changes
in circulating insulin. In contrast to b-adrenergic stimulation,
a-adrenergic agonists impair the ability to buffer increases in
Kþ induced by intravenous loading or exercise.

Acid-Base Status
It has long been recognized that acute disturbances in acid-
base equilibrium effect changes in serum Kþ, such that alkale-
mia results in an intracellular shift of Kþ, whereas acidemia is
associated with Kþ release from cells. A change of 0.1 unit in
plasma pH results in a 0.6 mmol/L change in plasma Kþ in the
opposite direction. Despite the complexities of changes in Kþ
homeostasis associated with various acid-base disorders, a
few general observations can be made. Induction of metabolic
acidosis by infusion of mineral acids (NH4þCl# or HþCl#)
consistently increases serum Kþ, whereas organic acidosis
fails to increase serum Kþ in most cases. The use of the oral
phosphate binder sevelamer in patients with end-stage renal
disease is associated with a metabolic acidosis, as previously
described in Chapter 6. This acidosis is associated with an
increase in plasma Kþ, which is ameliorated by a rise in the
dialysate bicarbonate concentration. Metabolic alkalosis
induced by sodium bicarbonate infusion usually results in a
modest reduction in the plasma Kþ level. Respiratory alkalosis
reduces the serum Kþ concentration by a magnitude compara-
ble to that of metabolic alkalosis. Acute respiratory acidosis
increases the serum level of Kþ; however, the absolute
increase is smaller than that induced by metabolic acidosis
from inorganic acids.

RENAL POTASSIUM EXCRETION

Potassium Secretion in the Distal Nephron

A substantial fraction of filtered potassium is reabsorbed prior
to entry into the superficial distal tubules, a process mediated
by the proximal tubule and loop of Henle. Renal potassium
excretion is primarily determined by regulated secretion in
the principal cells of the connecting segment (CNT) and corti- 139
cal collecting duct (CCD). Apical Naþ entry through the amilor-
ide-sensitive epithelial Naþ channel (ENaC) results in the
generation of a lumen-negative potential difference in the
CNT and CCD, which drives passive Kþ exit through apical CH 7
Kþ channels. A critical consequence of this relationship is that

Disorders of Potassium Metabolism

Kþ secretion depends on delivery of adequate luminal Naþ to
the CNT and CCD. Basolateral exchange of Naþ and Kþ is
mediated by Naþ/Kþ-ATPase, which provides the driving force
for Naþ entry and Kþ exit at the apical membrane. There are
several subpopulations of apical Kþ channels in the CCD, most
prominently a small-conductance 30-pS renal outer medullary
potassium (ROMK) channel and a large-conductance, Ca2þ-acti-
vated, 150-pS (maxi-K or BK) channel. The higher density and
higher open probability of the ROMK channel suggest that it
probably mediates Kþ secretion under baseline conditions. In
contrast, the pharmacology of flow-dependent Kþ secretion in
the CCD is consistent with the voltage-gated, calcium-sensitive
maxi-K channels. Apical KþCl# cotransport is increasingly
thought to play a role in Kþ excretion.

Potassium Reabsorption in the Distal Nephron

The distal nephron has considerable capacity to reabsorb
potassium, particularly in the setting of dietary potassium
restriction. This is achieved primarily via activation of apical
Hþ/Kþ-ATPase pumps in the intercalated cells in the outer
medullary collecting duct (OMCD).

Control of Potassium Secretion: Aldosterone

and Dietary Kþ Intake
Aldosterone and downstream effectors of this hormone have
clinically relevant effects on Kþ excretion, and the ability to
excrete Kþ is modulated by systemic aldosterone levels. Aldo-
sterone mediates Kþ loss by inducing a marked increase in the
density of apical ENaCs in the CNT/CCD, via mechanisms
including inhibition of retrieval of ENaC from the apical mem-
brane and increasing the open probability of the channel, thus
increasing the driving force for apical Kþ excretion. Aldoste-
rone also increases the expression of Naþ/Kþ-ATPase subunits
in the CCD. Other factors favoring Kþ excretion include
enhanced dietary Kþ intake, which increases the density of
expression of ROMK channels in the CCD, along with a mod-
est increase in ENaC density, and a brisk urine flow rate in the
distal nephron segment, which activates maxi-K channels in
the CNT/CCD.
140 Regulation of Renal Renin and Adrenal
Aldosterone

Modulation of the renin-angiotensin-aldosterone stystem

þ
II (RAAS) has profound clinical effects on K homeostasis.
Renin secretion by juxtaglomerular cells within the afferent
Disturbances in Control of Body Fluid Volume and Composition

arteriole in response to various stimuli, including decreased

Cl# delivery to the macula densa, renal hypoperfusion, and
heightened renal sympathetic tone, ultimately results in aldo-
sterone release from the adrenal gland by means of angioten-
sin II (A-II). Hyperkalemia is also an independent and
synergistic stimulus for adrenal release of aldosterone,
although dietary Kþ loading is less potent than dietary NaCl
restriction in stimulating aldosterone release. Aldosterone
induces a marked increase in the density of apical Naþ chan-
nels in the CNT and CCD, increasing the stimulus for apical
Kþ excretion. The relationship between renal renin release,
RAAS, and cyclooxygenase-2 (COX-2) is particularly complex.
COX-2 is densely expressed in the macula densa, with marked
recruitment of COX-2þ cells seen with salt restriction or furo-
semide treatment. Increased expression of COX-2 appears to
be mediated via reduced intracellular chloride in macula
densa cells, whereas both aldosterone and A-II reduce its
expression. COX-2 derived prostaglandins play a dominant
role in stimulating renal renin release by salt restriction, furo-
semide treatment, renal artery occlusion, or angiotensin-con-
verting enzyme (ACE) inhibition.

Urinary Indices of Potassium Excretion

A widely used surrogate to measure distal tubule Kþ excretion
is the transtubular Kþ gradient (TTKG), which is defined by
the following equation:

½Kþ &urine ' Osmblood

TTKG ¼
½Kþ &blood ' Osmurine

Water absorption is an important determinant of the absolute

Kþ concentration in the final urine—hence, the use of a ratio
of urine to plasma osmolality. The expected values of the TTKG
are less than 3 in the presence of hypokalemia and more than 7
to 8 in the presence of hyperkalemia. The TTKG may be less use-
ful in patients ingesting diets of changing Kþ and mineralocorti-
coid intake, and is therefore ideally used in a steady-state
situation. There is, however, a linear relationship between
serum aldosterone and the TTKG, suggesting that it provides a
rough approximation of the ability to respond to aldosterone
with a kaliuresis. Moreover, determination of urinary
electrolytes provides a measurement of urinary Naþ, which can 141
determine whether significant prerenal stimuli are limiting dis-
tal Naþ delivery and thus Kþ excretion.

CH 7
CONSEQUENCES OF HYPERKALEMIA AND

Disorders of Potassium Metabolism

HYPOKALEMIA
Consequences of Hypokalemia

Excitable Tissues: Muscle and Heart

Hypokalemia is a risk factor for ventricular and atrial arrhyth-
mias, especially in those with other risk factors including preex-
isting heart disease or digoxin or antiarrhythmic medication.
A serum Kþ of less than 3.5 mmol/L in patients undergoing car-
diac surgery is an independent predictor of serious peri- and
intraoperative arrhythmias. Electrocardiographic (ECG) changes
include broad flat T waves, ST depression, and QT prolongation,
most marked when serum Kþ is below 2.7 mmol/L. Hypokale-
mia, often accompanied by hypomagnesemia, is an important
cause of the long QT syndrome (LQTS), either alone or in combi-
nation with drug toxicity or with the inherited LQTS-associated
mutations in cardiac Kþ or Naþ channels. In muscle, hypokale-
mia impairs membrane depolarization, resulting in muscle
weakness that can progress to paralysis in severe cases. Hypoka-
lemia of diverse etiology predisposes to rhabdomyolysis with
acute kidney injury.

Renal and Cardiovascular Consequences

The prominent functional changes that occur in response to
hypokalemia include NaþCl# retention, polyuria, phosphatu-
ria, hypocitraturia, and increased ammoniagenesis. Polyuria
may result from a vasopressin-resistant defect in urinary con-
centrating ability. Hypokalemic or kaliopenic nephropathy
can cause end-stage renal disease, mostly in patients with
long-standing hypokalemia due to eating disorders or laxative
abuse. Histologic changes include a relatively specific proxi-
mal tubular vacuolization, interstitial nephritis, and renal
cysts. In animal models, hypokalemia has been shown to
increase susceptibility to acute kidney injury caused by ische-
mia, gentamicin, and amphotericin.
There is substantial evidence to support an excess of cardio-
vascular morbidity in patients with hypokalemia or reduced
dietary Kþ. Short-term Kþ restriction in healthy humans or
those with essential hypertension results in NaþCl# retention
and hypertension. Correction of hypokalemia is particularly
important in hypertensive patients treated with diuretics; nor-
mokalemia improves control of blood pressure, and the cardio-
vascular benefits of diuretic agents are blunted by hypokalemia.
142 Consequences of Hyperkalemia

Excitable Tissues: Muscle and Heart

Hyperkalemia constitutes a medical emergency largely because
II of its effect on the heart. Classically, the ECG manifestations in
hyperkalemia progress as shown in Table 7-2. However, these
Disturbances in Control of Body Fluid Volume and Composition

changes are notoriously variable and not always observed.

Hyperkalemia potentiates the blocking effect of lidocaine on
the cardiac Naþ channel, which may precipitate asystole or ven-
tricular fibrillation. Hemodialysis patients and patients with
chronic kidney disease may not demonstrate ECG changes, per-
haps because of concomitant abnormalities in serum Ca2þ. In
severe cases, hyperkalemia may present with a flaccid paralysis;
complete resolution is seen with treatment of the hyperkalemia.
Interestingly, myopathy due to hyperkalemia rarely affects the
diaphragm, cranial nerves, or sensory function.

Renal Consequences
In contrast to hypokalemia, hyperkalemia does not affect prox-
imal tubular ammoniagenesis, but has a significant effect on
the ability to excrete an acid urine via reduction in urinary
excretion of ammonium (NH4þ).

Approximate Relationship between

Table 7-2 Hyperkalemic Electrocardiographic Changes
and Serum Potassium Levels
Serum Kþ Levels
(mmol/L) Electrocardiographic Abnormality
Mild hyperkalemia: Tall peaked T waves with narrow base
5.5–6.5 (best seen in precordial leads)
Moderate Peaked T waves
hyperkalemia: 6.5–8 Prolonged PR interval
Decreased amplitude of P waves
Widening of QRS complex
Severe hyperkalemia: Absence of P wave
>8 Intraventricular blocks, fascicular
blocks, bundle branch blocks
QRS axis shift
Progressive widening of the QRS
complex
Sine wave pattern (sinoventricular
rhythm)
Ventricular fibrillation
Asystole
Data from Mattu A, Brady WJ, Robinson DA: Electrocardiographic manifestations
of hyperkalemia. Am J Emerg Med 18:721–729, 2000.
 HYPOKALEMIA 143
þ
Hypokalemia (K < 3.6 mEq/L) in clinical practice is usually
mild, with Kþ levels ranging between 3 and 3.5 mmol/L, but
in up to 25% of patients, it can be moderate to severe CH 7
(<3 mmol/L). It is a common problem in patients receiving thi-

Disorders of Potassium Metabolism

azide diuretics, with an incidence of up to 48%, and is also a
common finding in the peritoneal dialysis population. Hypoka-
lemia has been reported to increase in-hospital mortality rates
up to 10-fold, likely due to significant effects on arrhythmogen-
esis, blood pressure, and cardiovascular morbidity.

Spurious Hypokalemia

Delayed sample analysis can cause spurious hypokalemia due

to cellular uptake. Rarely, patients with profound leukocytosis
due to acute leukemia present with artifactual hypokalemia
caused by time-dependent uptake of Kþ by the large white
blood cell mass. Such patients do not develop clinical or elec-
trocardiographic complications of hypokalemia.

Redistribution and Hypokalemia

Administered insulin is a frequent cause of iatrogenic hypoka-

lemia. Alterations in the activity of the endogenous sympa-
thetic nervous system can cause hypokalemia in several
settings, including alcohol withdrawal, acute myocardial
infarction, and head injury. Redistributive hypokalemia after
severe head injury has been reported to be particularly pro-
found, with marked rebound hyperkalemia following repletion.
b2-Agonists are powerful activators of cellular Kþ uptake and
hypokalemia may complicate the therapy of asthma and the
use of tocolytics during labor. Downstream activation of cyclic
AMP (cAMP) by xanthines such as theophylline and caffeine
may also induce hypokalemia; a synergistic effect may be seen
with b2-agonists. Occult sources of sympathomimetics, such as
pseudoephedrine and ephedrine in cough syrup or dieting
agents, are a frequently overlooked cause of hypokalemia.

Hypokalemic Periodic Paralysis

Hypokalemic periodic paralysis disorders (HOKP type I and

II) are genetic disorders characterized by reversible attacks of
paralysis with hypokalemia, and are typically precipitated
by rest after exercise and meals rich in carbohydrate.
Although the induction of endogenous insulin by carbohy-
drate meals is thought to reduce serum Kþ levels, insulin
144 can precipitate paralysis in HOKP in the absence of significant
hypokalemia. Paralysis is associated with multiple other
causes of hypokalemia, both acquired and genetic, including
Fanconi syndrome, Gitelman syndrome, and hypokalemic dis-
II tal renal tubular acidosis (RTA).
Thyrotoxic periodic paralysis is classically seen in the
Disturbances in Control of Body Fluid Volume and Composition

Asian population, and typically presents with weakness of

the extremities and limb girdles. Attacks may be precipitated
by rest and carbohydrate-rich meals and almost never occur
during vigorous activity. Clinical signs and symptoms of
hyperthyroidism are not always present. Hypokalemia is pro-
found, ranging between 1.1 and 3.4 mmol/L, and is frequently
accompanied by hypophosphatemia and hypomagnesemia.
Treatment involves high-dose propranolol (3 mg/kg), which
rapidly reverses the metabolic disturbances. Of note, aggres-
sive Kþ replacement can result in hyperkalemia.

Nonrenal Potassium Loss

The loss of Kþ from skin is typically low, with the exception

of extremes in physical exertion. Direct gastric loss of Kþ
due to vomiting or nasogastric suctioning is typically mini-
mal, but the ensuing hypochloremic alkalosis results in per-
sistent kaliuresis due to secondary hyperaldosteronism and
bicarbonaturia. Intestinal loss of Kþ due to diarrhea is an
important cause of hypokalemia and may be associated with
a non–anion gap metabolic acidosis with a negative urinary
anion gap (consistent with an intact ability to increase NH4þ
excretion).

Renal Potassium Loss

Drugs
Diuretics are an important cause of hypokalemia because of
their ability to increase distal flow rate and distal delivery of
Naþ. Thiazides generally cause more hypokalemia than loop
diuretics, despite their lower natriuretic efficacy. Other drugs
associated with hypokalemia due to kaliuresis include high
doses of penicillin-related antibiotics, which are thought to
increase obligatory Kþ excretion by acting as non-reabsorbable
anions in the distal nephron. Kþ and magnesium wasting can
result from tubule toxins, including gentamicin (which can
masquerade as Bartter syndrome), cisplatin, amphotericin,
foscarnet, and ifosfamide. Aggressive replacement of magne-
sium is obligatory in the treatment of combined hypokalemia
and hypomagnesemia, because successful Kþ replacement
depends on treatment of the hypomagnesemia.
Hyperaldosteronism 145
Hyperaldosteronism may be primary or secondary; increased
levels of circulating renin in secondary hyperaldosteronism
lead to increased angiotensin II (A-II) and thus aldosterone,
and may be associated with hypokalemia. Causes of secondary CH 7
hyperaldosteronism include renal artery stenosis, and para-

Disorders of Potassium Metabolism

neoplastic or renin-secreting renal tumors. Primary hyperal-
dosteronism may be genetic or acquired. The two major
forms of isolated primary hyperaldosteronism are denoted
familial hyperaldosteronism type I (FH-I, also known as gluco-
corticoid-remediable hyperaldosteronism [GRA]), and familial
hyperaldosteronism type II (FH-II). Patients with FH-II are
clinically indistinguishable from sporadic forms of primary
hyperaldosteronism due to bilateral adrenal hyperplasia.
Patients with FH-I/GRA are generally hypertensive and typi-
cally present at an early age; the severity of hypertension is
variable, such that some affected individuals are normoten-
sive. FH-I is caused by a fusion of the ACTH-responsive 11b-
hydroxylase promoter to the coding region of aldosterone
synthase. Aldosterone levels are modestly elevated and regu-
lated solely by adrenocorticotropic hormone (ACTH). The
diagnosis can be confirmed by a dexamethasone suppression
test, with suppression of aldosterone to less than 4 ng/dL con-
sidered diagnostic. The majority of patients with FH-I are nor-
mokalemic, and are able to appropriately increase renal Kþ
excretion in response to Kþ loading or fludrocortisone, but fail
to increase serum aldosterone in response to hyperkalemia.
Acquired causes of primary hyperaldosteronism include
aldosterone-producing adenomas (APAs), primary or unilat-
eral adrenal hyperplasia (PAH), idiopathic hyperaldosteron-
ism (IHA) due to bilateral adrenal hyperplasia, and adrenal
carcinoma. APA and IHA account for close to 60% and 40%,
respectively, of diagnosed hyperaldosteronism. As surgery
can be curative in cases of APA, adequate differentiation of
these tumors from IHA is critical; this may require adrenal
imaging and adrenal venous sampling.
Increasing use of the plasma aldosterone concentration (PAC)/
plasma renin activity (PRA) has led to reports of a much higher
incidence of primary hypertension than was previously appre-
ciated, with incidence rates of up to 72% reported. The true
incidence remains difficult to evaluate for a number of reasons,
including historical selection bias due to screening for hyperal-
dosteronism only when hypokalemia or hypertension were
present. However, hypokalemia is not a universal feature of pri-
mary hyperaldosteronism, as compensatory mechanisms can
return the serum Kþ to the normal range. Hypokalemia may also
occur with systemic increases in glucocorticoids. In classic Cush-
ing syndrome the incidence of hypokalemia is 10%, compared to
57% to 100% in patients with ectopic ACTH secretion.
146 Syndromes of Apparent Mineralocorticoid Excess
In the classic form of apparent mineralocorticoid excess (AME),
recessive loss-of-function mutations in the gene encoding 11b-
hydroxysteroid dehydrogenase-2 (11bHSD-2) cause a defect in
II the peripheral conversion of cortisol to the inactive glucocorti-
coid cortisone. Because the mineralocorticoid receptor has
Disturbances in Control of Body Fluid Volume and Composition

equivalent affinity for aldosterone and cortisol, in patients with

AME, the unregulated mineralocorticoid effect of glucocorti-
coids results in hypertension, hypokalemia, and metabolic
alkalosis, with suppressed PRA and aldosterone.
Pharmacologic inhibition of 11bHSD-2 by glycyrrhetinic or
glycyrrhizinic acid, contained in licorice, and carbenoxolone
is also associated with hypokalemia and AME. A rare, mecha-
nistically distinct form of AME has also been reported, due to
a gain-of-function mutation in the mineralocorticoid receptor
(MR), such that the MR is constitutively activated in the
absence of ligand, and demonstrates significant affinity for
progesterone.

Liddle Syndrome, Bartter Syndrome, and Gitelman

Syndrome
These important genetic disorders resulting in hypokalemia
are discussed in Chapter 20, Inherited Disorders of the
Kidney.

Magnesium Deficiency
Magnesium deficiency results in refractory hypokalemia, par-
ticularly if the serum Mg2þ concentration is less than 0.5 mg/dL
(0.22 mmol/L). Mg2þ deficiency often accompanies hypoka-
lemia, in part because the causative renal tubule disorders
(e.g., aminoglycoside nephrotoxicity) may cause both
kaliuresis and magnesium wasting. Magnesium depletion
inhibits muscle Naþ/Kþ-ATPase activity triggering Kþ efflux
from muscle and cardiac myocytes promoting a secondary
kaliuresis and depletion of intracellular myocardial Kþ
stores, even though serum Kþ levels may remain normal.
This phenomenon is particularly important in patients with
cardiac disease on diuretics and digoxin. In such patients,
hypokalemia and arrhythmias respond to correction of
magnesium deficiency and potassium supplementation.

The Clinical Approach to Hypokalemia

The initial priority in the evaluation of hypokalemia is an
assessment for signs and symptoms suggestive of an impend-
ing emergency that necessitates immediate management. The
underlying cause can usually be established from the history,
physical examination, basic laboratory tests, or all three.
However, persistent hypokalemia despite appropriate initial 147
management requires a more rigorous investigation; in most
cases, a systematic approach reveals the underlying etiology
(Fig. 7-1).
The most common causes of chronic, diagnosis-resistant CH 7
hypokalemia are Gitelman syndrome, surreptitious vomiting,

Disorders of Potassium Metabolism

and diuretic abuse. An associated acidosis suggests the diagno-
sis of hypokalemic distal or proximal tubular acidosis. Patients
with hypokalemia secondary to eating disorders may have a
constellation of associated symptoms and signs, including den-
tal erosion and depression. Hypokalemic patients with bulimia
have an associated metabolic alkalosis, with an obligatory natri-
uresis accompanying the loss of bicarbonate; the urinary Cl#
level is typically less than 10 mmol/L. Although laxative abuse
is perhaps a less common cause of chronic hypokalemia, an
associated metabolic acidosis with a negative urinary anion
gap should raise clinical suspicion of this underlying diagno-
sis. Urinary excretion of Naþ, Kþ, and Cl# is high in patients
who abuse diuretics. Marked variability in urinary electrolytes
is an important clue for diuretic abuse, which can be verified
with urinary drug screens. Nephrocalcinosis is very common
in furosemide abuse due to the increase in urinary calcium
excretion. The differentiation of Gitelman syndrome from Bart-
ter syndrome is discussed in Chapter 20.

Treatment of Hypokalemia*

The goals of therapy in hypokalemia are as follows:

• Prevention of life-threatening conditions
• Replacement of the Kþ deficit
• Diagnosis and correction of the underlying cause
• Avoidance of rebound hyperkalemia
The urgency of therapy depends on the severity of hypokale-
mia, associated conditions, and the rate of decline in the serum
Kþ. A rapid drop to less than 2.5 mmol/L poses a high risk of
cardiac arrhythmias and necessitates urgent replacement.
Although replacement is usually limited to patients with a true
deficit, it should be considered in patients with hypokalemia
due to redistribution (e.g., hypokalemic periodic paralysis) when
serious complications such as muscle weakness, rhabdomyoly-
sis, and cardiac arrhythmias are present or imminent. The risk
of arrhythmias is highest in older patients, patients with organic
heart disease, and those on antiarrhythmic drugs, where even
mild-modest hypokalemia may prove pro-arrhythmogenic.

*All recommended doses in this section are for adults only.

148

Move to Yes Hypokalemia Yes No further

Emergency? Pseudohypokalemia?
therapy (serum K+ < 3.5 mmol/L) workup
No No

Treat accordingly Yes Clear evidence History, physical examination Clear evidence of Yes Treat
and re-evaluate of low intake and basic laboratory tests transcellular shift accordingly

No No
• Insulin excess
Urine K+ • β2-Adrenergic agonists
• FHPP
• Hyperthyroidism
• Barium intoxication
<15 mmol/day >15 mmol/day • Theophylline
• Chloroquine

Extrarenal loss/ Renal loss

remote renal loss

Acid-base status TTKG

 Metabolic acidosis Normal Metabolic alkalosis >4 <3
• GI K+ loss • Profuse • Remote diuretic use
sweating • Remote vomitting or
stomach drainage ↑ Distal K+ ↑ Tubular flow
• Profuse sweating secretion • Osmotic diuresis

BP and/or volume

Low or normal High

Non-reabsorbable
anions other than
HCO3– Variable Acid-base status Aldosterone
• Hippurate
• Penicillins

Metabolic acidosis Metabolic alkalosis High Low

• Proximal RTA
• Distal RTA
• DKA Urine Cl– (mmol/L) Renin Cortisol
• Amphotericin B
• Acetazolamide

Figure 7-1. The diagnostic approach to hypokalemia.

Continued
149
150

>20 <10 High Low

• Vomiting • Ras • PA
Urine Ca/Cr • Chloride- • RST • GRA
(molar ratio) losing • Malignant
diarrhea HTN

>0.20 <0.15 High Normal

• Loop diuretic • Thiazide diuretic • Cushing • Liddle syndrome

• Bartter syndrome • Gitelman syndrome syndrome • Licorice
• AME

Figure 7-1, cont’d. AME, apparent mineralocorticoid excess; BP, blood pressure; CCD, cortical collecting duct; DKA,
diabetic ketoacidosis; FHPP, familial hypokalemic periodic paralysis; GI, gastrointestinal; GRA, glucocorticoid remediable
aldosteronism; HTN, hypertension; PA, primary aldosteronism; RAS, renal artery stenosis; RST, renin-secreting tumor;
RTA, renal tubular acidosis; TTKG, transtubular potassium gradient. (From Mount DB, Zandi-Nejad K: Disorders of
potassium balance. In Brenner & Rector’s The Kidney, 8th ed. Philadelphia, WB Saunders, 2005, pp 547–588.)
 The goal is to raise the serum Kþ to a safe range rapidly and 151
then replace the remaining deficit at a slower rate over days to
weeks. In the absence of abnormal Kþ redistribution, the serum
Kþ drops by approximately 0.27 mmol/L for every 100-mmol
reduction in total body stores. The treatment of asymptomatic CH 7
patients with borderline or low normal serum Kþ is controver-

Disorders of Potassium Metabolism

sial, but supplementation is recommended in patients with
serum Kþ levels lower than 3 mmol/L. In high-risk patients
(i.e., those with cardiac or severe hepatic disease), serum Kþ
concentrations should be maintained above 4 mmol/L. Patients
with severe hepatic disease may poorly tolerate even mild-
moderate hypokalemia due to the associated increase in ammo-
niagenesis. In patients with mild-to-moderate hypertension,
potassium supplementation should be considered when the
serum Kþ concentration falls below 3.5 mmol/L.
Potassium chloride is the most widely administered salt, as, in
most patients, metabolic alkalosis typically accompanies the
chloride depletion induced by losses through upper gastrointes-
tinal secretion or diuretic use, and significantly contributes to
renal Kþ wasting. In these cases, replacing chloride along with
Kþ is essential in treating the alkalosis and preventing further
Kþ loss. Slow-release forms are more palatable and better toler-
ated; however, they have been associated with gastrointestinal
ulceration and bleeding. Potassium phosphate is indicated when
phosphate deficit accompanies Kþ depletion (e.g., diabetic
ketoacidosis). Potassium bicarbonate or its precursors should
be considered in patients with hypokalemia and metabolic aci-
dosis (e.g., diarrheal conditions, renal tubular acidosis).
Rapid correction of hypokalemia through oral supplementa-
tion is possible; 125 to 165 mmol of Kþ as a single oral dose
can increase the serum Kþ concentration by approximately 2.5
to 3.5 mmol/L in 60 to 120 minutes. If the patient is experien-
cing life-threatening signs and symptoms of hypokalemia, how-
ever, intravenous infusion of Kþ should be administered
acutely. Concentrations of KCl of up to 400 mmol/L (40 mmol
in 100 mL of normal saline) have been used in this setting
under continuous ECG monitoring. These solutions are best
given through a large central vein; the femoral vein is preferable
as it avoids acute local increases in the concentration of Kþ
with deleterious effects on cardiac conduction. To avoid
venous pain, irritation, and sclerosis, concentrations of more
than 60 mmol/L should not be given through a peripheral vein.
The maximum rate of administration should be 10 to 20
mmol/hour, although rates of 40 to 100 mmol/hour (for a short
period) have been used in patients with life-threatening condi-
tions. Higher rates of infusion (>80 mmol/hour) are associated
with rapid increase in serum Kþ levels with accompanying
ECG changes. Infusion rates greater than 10 mmol/hour
require continuous ECG monitoring. It is essential to remember
152 that volume expansion in patients with moderate-to-severe
hypokalemia and Cl#-responsive metabolic alkalosis should
be performed cautiously, and with repeated Kþ measurements,
as the bicarbonaturia associated with volume expansion may
þ
II aggravate renal K wasting and hypokalemia. In patients with
combined severe hypokalemia and hypophosphatemia (e.g.,
Disturbances in Control of Body Fluid Volume and Composition

diabetic ketoacidosis), intravenous Kþ phosphate can be used.

However, this solution should be infused at a rate of less than
50 mmol over 8 hours to prevent the risk of hypocalcemia and
metastatic calcification.
In addition to potassium supplementation, strategies to limit
Kþ losses should be considered; these strategies include mini-
mizing the dose of non–Kþ-sparing diuretics, restricting Naþ
intake, and using a combination of non–Kþ-sparing and
Kþ-sparing medications (e.g., angiotensin-converting enzyme
inhibitors, angiotensin receptor blockers, Kþ-sparing diuretics,
b-blockers). The use of a Kþ-sparing diuretic is of particular
importance in hypokalemia resulting from primary hyperaldos-
teronism and related disorders, such as Liddle syndrome and
AME, as Kþ supplementation alone may be ineffective in this
setting. It is important to measure serum magnesium and cor-
rect hypomagnesemia in all patients with hypokalemia (see pre-
vious discussion). The risk of overcorrection or rebound
hyperkalemia in redistribution hypokalemia is high and may
result in fatal arrhythmias. When increased sympathetic tone
or increased sympathetic response is thought to play a domi-
nant role (thyrotoxic periodic paralysis, theophylline overdose,
and acute head injury), the use of nonspecific b-adrenergic
blockade with propranolol generally avoids this complication.

HYPERKALEMIA
Hyperkalemia is usually defined as a potassium level of
5.5 mmol/L or greater. In most hospitalized patients, the cause
of hyperkalemia is multifactorial, with reduced renal function,
medications, older age (60 years or older), and hyperglycemia
being the most common contributing factors. In patients with
end-stage renal disease (ESRD), the prevalence of hyperkalemia
is reportedly 5% to 10%, and is the reason for emergency hemo-
dialysis in 24% of ESRD patients already maintained on
hemodialysis.

Pseudohyperkalemia
Factitious or pseudohyperkalemia is an artifactual increase in
serum Kþ due to the release of Kþ during or after venipunc-
ture. Fist clenching or tourniquet use may increase Kþ efflux
from local muscle. Severe thrombocytosis, leukocytosis, or 153
erythrocytosis may increase the measured Kþ concentration
due to release from these cellular elements. In addition, acute
anxiety may provoke a respiratory alkalosis with redistribu-
tive hyperkalemia. Cooling of blood prior to separation of cells CH 7
from plasma or serum is also a well-recognized cause of arti-

Disorders of Potassium Metabolism

factual hyperkalemia. Hereditary forms of pseudohyperkale-
mia have also been described, due to an increase in the
passive Kþ permeability of erythrocytes.

Excess Potassium Intake and Tissue Necrosis

Hyperkalemia is uncommon in the absence of a defect in distal
tubule secretion usually combined with a reduction in glomerular
filtration rate (GFR). However, increased intake of even small
amounts of Kþ may provoke hyperkalemia in susceptible patients
such as those with renal impairment and hyporeninemic hypoal-
dosteronism. Occult sources of Kþ must also be considered,
including salt substitutes and alternative medicines. Iatrogenic
causes include simple overreplacement with KCl or administra-
tion of a potassium-containing medication, such as potassium
penicillin, to a susceptible patient. Red blood cell transfusion is
a well-described cause of hyperkalemia, particularly in children
or in massive transfusions; risk factors include the rate and vol-
ume of the transfusion, the use of central venous infusion, the
use of irradiated blood, and the age of the blood infused. Hyperka-
lemia often complicates rhabdomyolysis due to the enormous
store of Kþ in muscle and this is further compounded by the fall
in GFR triggered by pigment-mediated acute tubular necrosis.
Those patients with hypokalemia-associated rhabdomyolysis,
where the hypokalemia is due to redistribution, are at particular
risk of subsequent hyperkalemia. Massive releases of Kþ and other
intracellular contents may also occur due to acute tumor lysis.

Redistribution and Hyperkalemia

Several different mechanisms can induce efflux of intracellular
Kþ, resulting in hyperkalemia. Increases in serum Kþ due to
hypertonic mannitol or hypertonic saline are generally attribu-
ted to a “solvent drag” effect as water moves in response to the
osmotic gradient. Similarly, diabetics are prone to severe
hyperkalemia in response to intravenous hypertonic glucose
in the absence of co-administered insulin. Succinylcholine
depolarizes muscle cells, resulting in a rapid but usually tran-
sient hyperkalemia. The use of this agent is contraindicated
in patients who have sustained thermal trauma, neuromuscular
injury, mucositis, or prolonged immobilization in an intensive
154 care unit setting, because Kþ efflux through acetylcholine
receptors is enhanced in these patients and can result in signifi-
cant hyperkalemia. Digoxin inhibits Naþ/Kþ-ATPase and
impairs the uptake of Kþ by skeletal muscle, such that digoxin
II overdose can result in hyperkalemia. b-Blockers cause hyperka-
lemia in part by inhibiting cellular uptake but also through
Disturbances in Control of Body Fluid Volume and Composition

hyporeninemic hypoaldosteronism induced by inhibiting renal

renin release and adrenal aldosterone release.

Reduced Renal Potassium Excretion

Hypoaldosteronism
Acquired hyperreninemic hypoaldosteronism has been
described in critical illness, type II diabetes, amyloidosis,
and metastasis of carcinoma to the adrenal gland. Aldosterone
synthesis is selectively reduced by heparin, with both unfrac-
tionated and low-molecular-weight heparin associated with
hyperkalemia. Heparin also reduces the adrenal aldosterone
response to A-II and hyperkalemia, resulting in hyperrenine-
mic hyperaldosteronism.
Most primary adrenal insufficiency is due to autoimmunity,
either in Addison disease or in the context of a polyglandular
endocrinopathy. The incidence of hyperkalemia in patients with
Addison disease is likely 50% to 60%, reflecting the importance
of aldosterone-independent modulation of Kþ excretion by the
CCD. Adrenal insufficiency also occurs in renal amyloidosis,
the antiphospholipid syndrome (adrenal infarction), and HIV
disease. Although the adrenal involvement in HIV disease is usu-
ally subclinical, adrenal insufficiency may be precipitated by
stress, drugs such as ketoconazole that inhibit steroidogenesis,
or the acute withdrawal of steroid agents such as megestrol.
Inherited causes of hypoaldosteronism include X-linked
adrenal hypoplasia congenita (AHC) which presents with pri-
mary adrenal failure and hyperkalemia either shortly after
birth or much later in childhood; lipoid congenital adrenal
hyperplasia (lipoid CAH), a severe autosomal recessive syn-
drome characterized by impaired synthesis of mineralocorti-
coids, glucocorticoids, and gonadal steroids, where patients
present in early infancy with adrenal crisis; congenital
adrenal hyperplasia (CAH), the classic, salt-wasting form of
CAH due to 21-hydroxlase deficiency. Isolated deficits in
aldosterone synthesis with hyperreninemia may also be
caused by loss-of-function mutations in aldosterone synthase.

Hyporeninemic Hypoaldosteronism
Hyporeninemic hypoaldosteronism is a very common predis-
posing factor for hyperkalemia. It is associated with diabetes
mellitus, older age, and patients with renal insufficiency. It
has also been described in systemic lupus erythematosus 155
(SLE), multiple myeloma, and acute glomerulonephritis. Clas-
sically, patients have suppressed plasma renin activity (PRA)
and aldosterone levels, which cannot be activated by furose-
mide administration or sodium restriction. Approximately CH 7
50% have an associated acidosis, with reduced renal excretion

Disorders of Potassium Metabolism

of NH4þ, a positive urinary anion gap, and urine pH less than
5.5 (see Chapter 6, Disorders of Acid-Base Balance).

Acquired Tubule Defects and Potassium Excretion

Unlike hyporeninemic hypoaldosteronism, hyperkalemic distal
renal tubular acidosis is associated with a normal or increased
aldosterone or PRA, or both. Urine pH in these patients is greater
than 5.5, and they are unable to increase acid or Kþ excretion in
response to furosemide or fludorcortisone. Classic causes
include SLE, sickle cell anemia, and amyloidosis.

Hereditary Tubule Defects and Potassium Excretion

These disorders include pseudohypoaldosteronism and Gordon
syndrome. See Chapter 20, Inherited Disorders of the Kidney.

Medication-Related Hyperkalemia

Nonsteroidal Anti-inflammatory Drugs

Hyperkalemia is a well-recognized complication of NSAIDs,
including COX-2 inhibitors, and may be seen particularly in
the setting of hypovolemia. NSAIDs cause hyperkalemia by a
variety of mechanisms, including the following:
• Decreased glomerular filtration rate and increased sodium
retention, causing decreased distal delivery of Naþ and
reduced distal flow rate
• Indirect inhibition of the flow-activated apical maxi-K
channel in the CNT and CCD (prostaglandin-dependent)
• Hyporeninemic hypoaldosteronism
• Blunting of the adrenal response to hyperkalemia (prosta-
glandin-dependent)

Cyclosporin and Tacrolimus

Cyclosporin and tacrolimus cause hyperkalemia by a variety
of mechanisms including hyporeninemic hypoaldosteronism,
partly via an inhibitory effect on COX-2 expression in the
macula densa, but also via inhibition of the apical secretory
Kþ channels in the distal nephron and basolateral Naþ/Kþ
-ATPase. Cyclosporin also causes distributive hyperkalemia,
particularly when used in combination with b-blockers. A
recent preliminary report suggests that the ability of cyclo-
sporin to activate KATP channels, which are widely distributed
156 in vascular smooth muscle, may have an underappreciated
propensity to cause hyperkalemia.

Amiloride-Sensitive Epithelial Sodium Ion Channel

II Inhibition
Inhibition of apical ENaC activity in the distal nephron by
Disturbances in Control of Body Fluid Volume and Composition

amiloride, and other Kþ-sparing diuretics, predictably results

in hyperkalemia. Amiloride is structurally similar to the anti-
biotics trimethoprim and pentamidine, which also inhibit
ENaC and can cause significant hyperkalemia even with stan-
dard doses. Risk factors for hyperkalemia due to normal-dose
trimethoprim include renal insufficiency and hyporeninemic
hypoaldosteronism.

Angiotensin-Converting Enzyme Inhibitors,

Mineralocorticoid and Angiotensin Antagonists
Hyperkalemia is a predictable and common effect of ACE inhi-
bition and antagonism of the mineralocorticoid receptor. The
adrenal release of aldosterone due to an increased Kþ concen-
tration depends on an intact adrenal-renal angiotensin system;
systemic ACE inhibitors and angiotensin receptor blockers
(ARBs) therefore abrogate this response. The increasing ratio-
nale to combine spironolactone with ACE inhibitors and ARBs
is likely to magnify the potential for serious hyperkalemia,
and adherence to measures which minimize the likelihood
of associated hyperkalemia will become increasingly impor-
tant. The patients at risk for development of hyperkalemia in
response to drugs targeting the RAAS, either alone or in com-
bination, are those in whom the renal ability to excrete a
potassium load is markedly diminished, owing to one or a
combination of the following:
• Decreased delivery of Naþ to the CCD (congestive heart fail-
ure, volume depletion, etc.)
• Decreased circulating aldosterone (hyporeninemic hypoal-
dosteronism, drugs such as ketoconazole or heparin)
• Inhibition of amiloride-sensitive Naþ channels in the CNT
and CCD, by co-administration of TMP/SMX, pentamidine,
or amiloride
• Chronic tubulointerstitial disease, with associated distal
nephron dysfunction
• Increased potassium intake (salt substitutes, diet, etc.)
Attention to, and correction of, the acid-base balance is also
important. In the preceding group, it is essential to closely and
regularly monitor the serum Kþ following the initiation of treat-
ment with ACE inhibitors, ARBs, and mineralocorticoid recep-
tors, and following each dose increment. It is clearly prudent to
initiate therapy with a single agent, usually at a low dose, with
the addition of sequential agents only when the patient has been 157
demonstrated to tolerate such medications.

The Clinical Approach to Hyperkalemia CH 7

Disorders of Potassium Metabolism

As for hypokalemia, the priority in the initial management of
hyperkalemia is to assess the requirement for emergency treat-
ment. Thereafter, a comprehensive investigation into the
underlying etiology should be undertaken, with particular
focus on medication history, diet (including supplements),
risk factors for kidney disease, and volume status (Fig. 7-2).
Initial laboratory tests should include a full electrolyte screen,
blood urea nitrogen (BUN), creatinine, serum osmolality, com-
plete blood count, and urinary pH, osmolality, creatinine, and
electrolytes, with calculation of the TTKG.

Treatment of Hyperkalemia

Most cases of hyperkalemia result from Kþ accumulation due to

a combination of renal insufficiency and impaired distal Kþ
excretion; Kþ elimination is therefore the aim of treatment. How-
ever, occasionally, hyperkalemia may be due to Kþ redistribu-
tion, and the clinical approach is addressed at achieving
intracellular translocation of Kþ rather than Kþ loss, as this may
result in rebound hypokalemia (e.g., diabetic ketoacidosis).
The management of hyperkalemia is based on a combination
of factors, including the severity of the hyperkalemia, the pres-
ence of ECG evidence of cardiac toxicity, and clinical signs of
hyperkalemia, which are also influenced by other electrolyte
abnormalities, acid-base status, and chronicity. Severe hyper-
kalemia (serum Kþ ( 8 mmol/L or serum Kþ > 6.5 mmol/L with
ECG changes other than peaked T waves) is a medical emer-
gency and requires urgent intervention (Table 7-3). Given the
limitations of ECG changes as a predictor of cardiac toxicity,
patients with moderate hyperkalemia (Kþ ( 6.5 mmol/L) in
the absence of ECG changes or symptomatic patients (muscle
weakness) should also receive prompt medical intervention.
Urgent management of severe hyperkalemia includes a 12-lead
electrocardiogram, admission to the hospital, continuous ECG
monitoring, and immediate treatment. The treatment of hyper-
kalemia is generally divided into three categories:
1. Antagonism of the cardiac effects of hyperkalemia
2. Rapid reduction in Kþ by redistribution into cells
3. Removal of Kþ from the body
The necessary measures to treat the underlying conditions
causing hyperkalemia should be undertaken to minimize the
158

Emergency Yes K+ ≥ 6 or Hypokalemia Yes No further

Pseudohyperkalemia?
therapy ECG changes (serum K+ ≥ 5.5 mmol/L) action
No No

Treat accordingly Yes Evidence of increased History, physical examination Evidence of Yes Treat accordingly
and re-evaluate potassium load and basic laboratory tests transcellular shift and re-evaluate
No No

Decreased urinary K+ excretion (<40 mmol/day) • Hypertonicity (e.g., mannitol)

• Hyperglycemia
• Succinylcholine
Decreased distal Urine Na+
Urine electrolytes • ε-Aminocaproic acid
Na+ delivery < 25 mmol/L
• Digoxin
• β-Blockers
TTKG • Metabolic acidosis (nonorganic)
• Arginine infusion
• Hyperkalemic periodic paralysis
• ↓ Insulin
>7 <6 • Exercise

Reduced tubular flow Reduced distal K+ secretion

(GFR > 20 mL/min)
 Advanced kidney failure Reduced Administer
(GFR ≤ 20 mL/min) ECV 9α-fludrocortisone

TTKG < 7 TTKG ≥ 7

(tubular resistance)

Low aldosterone
Drugs Other causes
• Amiloride • Tubulointerstitial diseases
• Spironolactone • Urinary tract obstruction Renin
• Triamterene • PHA type I
• Trimethoprim • PHA type II
• Pentamidine • Sickle cell disease High Low
• Eplerenone • Renal transplant
• Calcineurin inhibitors • SLE • Primary adrenal insufficiency • Diabetes mellitus
• Isolated aldosterone deficiency • Acute GN
• Heparin/LMW heparin • Tubulointerstitial diseases
• ACE-I/ARB • PHA type II
• Ketoconazole • NSAIDs
• β-Blockers

Figure 7-2. The diagnostic approach to hyperkalemia. ACE-I, angiotensin-converting enzyme inhibitor; acute GN, acute
glomerulonephritis; ARB, angiotensin II receptor blocker; CCD, cortical collecting duct; ECG, electrocardiogram; ECV,
effective circulatory volume; GFR, glomerular filtration rate; LMW heparin, low-molecular-weight heparin; PHA,
pseudohypoaldosteronism; SLE, systemic lupus erythematosus; TTKG, transtubular potassium gradient. (From Mount DB,
Zandi-Nejad K: Disorders of potassium balance. In Brenner & Rector’s The Kidney, 8th ed. Philadelphia, WB Saunders,
2005, pp 547–588.)
159
160
Table 7-3 Acute Management of Severe Hyperkalemia
Stabilize myocardium with calcium salts
Calcium gluconate 10% 10 mL as IV bolus
II Shift potassium into cells
Intravenous glucose (50 mL of 50%) with 5–10 IU of insulin
Disturbances in Control of Body Fluid Volume and Composition

and
10–20 mg nebulized albuterol over 10 min (2–4 mL of 5 mg/mL
albuterol solution)
or
Subcutaneous terbutaline injection (7 mg/kg)
Potassium removal with dialysis (in patients with ESRD, or when
conservative measures are unsuccessful in the non-ESRD
population)
ESRD, end-stage renal disease.
From Putcha N, Allon M: Management of hyperkalemia in dialysis patients. Semin
Dial 20:431–439, 2007.

factors that are contributing to hyperkalemia and to prevent

future episodes. Dietary restriction (usually 60 mEq/day) with
emphasis on the Kþ content of total parenteral nutrition solu-
tions and enteral feeding products (typically 25–50 mmol/L)
and adjustment of medications and intravenous fluids are nec-
essary; hidden sources of Kþ, such as intravenous antibiotics,
should not be overlooked.

Antagonism of Cardiac Effects

Intravenous calcium raises the action potential threshold and
reduces excitability without changing the resting membrane
potential, and reverses the depolarization blockade that occurs
with hyperkalemia. Calcium is available as calcium chloride
or calcium gluconate (10-mL ampules of 10% solutions) for
intravenous infusion. Calcium gluconate is less irritating to
the veins and can be used through a peripheral intravenous
line; calcium chloride can cause tissue necrosis if it extrava-
sates and it requires a central line. The recommended dose is
10 mL of 10% calcium gluconate (3–4 mL of calcium chloride),
infused intravenously over 2 to 3 minutes and under continu-
ous ECG monitoring. The effect of the infusion starts in 1 to 3
minutes and lasts 30 to 60 minutes. The dose should be
repeated if there is no change in ECG findings or if they recur
after initial improvement. Of note, calcium should be used with
extreme caution in patients taking digoxin, because hypercalce-
mia potentiates the cardiotoxic effects of digoxin. In such cases,
10 mL of 10% calcium gluconate should be added to 100 mL of
5% dextrose in water and infused over 20 to 30 minutes to avoid
hypercalcemia. To prevent the precipitation of calcium carbon-
ate, calcium should not be administered in solutions containing
bicarbonate. Intravenous calcium should be administered to all
patients with ECG changes; some authors advocate its use 161
in patients with serum potassium greater than 6 to 6.5 mmol/L,
even in the absence of ECG changes.

Redistribution of Potassium into Cells CH 7

Insulin and Glucose. Insulin lowers the serum Kþ concentra-
tion by shifting Kþ into cells. The effect is reliable, dose

Disorders of Potassium Metabolism

dependent, and effective, even in patients with chronic kid-
ney disease and ESRD. The recommended dose is 5 to 10 units
of regular insulin administered intravenously, followed by
50 mL of 50% dextrose. The effect of insulin on the Kþ level
begins in 10 to 20 minutes, peaks at 30 to 60 minutes, and
lasts for 4 to 6 hours. The expected fall in serum Kþ concentra-
tion should be 0.5 to 1.2 mmol/L and the dose can be repeated
as necessary. Hypoglycemia may occur in up to 75% of
patients. Therefore, a continuous infusion of 10% dextrose at
50 to 75 mL/hour and close monitoring of the blood glucose
levels is recommended following administration of the bolus
insulin/dextrose dose. Administration of glucose without sup-
plemental insulin is not recommended because the endoge-
nous insulin response to such a glucose load may be
variable, and glucose, in the absence of insulin, may in fact
increase serum Kþ, by increasing plasma osmolality. In hyper-
glycemic patients with glucose levels of 200 to 250 mg/dL,
insulin should be administered without glucose, but should
be followed by close monitoring of the serum glucose.
b2-Adrenergic Agonists. b2-Agonists are important, but
underutilized, agents for the acute management of hyperkale-
mia. They exert their effect by activating Naþ/Kþ-ATPase and
the NKCC1 Naþ-Kþ-2Cl# cotransporter, shifting Kþ into hepato-
cytes and skeletal myocytes. Albuterol (Salbutamol), a selective
b2-agonist, is the most widely used agent; both intravenous and
nebulized forms are effective. The recommended dose for intra-
venous administration is 0.5 mg of albuterol in 100 mL of 5%
dextrose, given over 10 to 15 minutes. The Kþ-lowering effect
starts in few minutes, is maximal at about 30 to 40 minutes,
and lasts for 2 to 6 hours. It reduces serum Kþ levels by approx-
imately 0.9 to 1.4 mmol/L. The recommended dose for inhaled
albuterol is 10 to 20 mg of nebulized albuterol in 4 mL of normal
saline, inhaled over 10 minutes. Its effects start at 30 minutes,
peaks at 90 minutes, and last for 2 to 6 hours. Inhaled albuterol
reduces serum Kþ levels by approximately 0.5 to 1 mmol/L.
Administration of inhaled or parenteral albuterol has an addi-
tive effect when given with glucose/insulin in reducing serum
Kþ levels, by approximately 1.2 to 1.5 mmol/L in total. Treat-
ment with albuterol may result in increased heart rate, an effect
seen more prominently with intravenous administration, and it
should be used with caution in patients with ischemic heart
disease. Importantly, up to 40% of patients with ESRD are not
162 responsive to Kþ-lowering effect of albuterol; it should not
therefore be used as a single agent in this population. In an
attempt to reduce pharmacokinetic variability, a weight-based
regimen of subcutaneous terbutaline (7 mg/kg) was recently
II given a trial in a small group of ESRD patients, and effected a
mean reduction in serum Kþ levels of 1.3 mmol/L in 30 to 90
Disturbances in Control of Body Fluid Volume and Composition

minutes, suggesting promise as an alternate approach, charac-

terized by simplicity of administration and speed of onset.
Sodium Bicarbonate. The role of bicarbonate in the acute
treatment of hyperkalemia is increasingly being challenged.
The effect of combining bicarbonate with either insulin-glu-
cose or with albuterol has been studied, with no convincing
benefit demonstrated in terms of lowering of the serum Kþ.
Potential adverse effects include a reduction in ionized cal-
cium levels and volume overload. Indeed some studies utiliz-
ing hypertonic bicarbonate have shown an increase in the
serum Kþ level, thought to be due to increased plasma osmo-
lality, promoting Kþ shift out of cells. As a result the routine
administration of bicarbonate, especially as a single agent,
has no role in the acute treatment of hyperkalemia. One excep-
tion may be the management of severely acidemic patients.

Removal of Potassium
Diuretics. Diuretics have a relatively modest effect on uri-
nary Kþ excretion in patients with chronic kidney disease.
However, they are useful in correcting chronic hyperkalemia
in patients with the syndrome of hyporeninemic hypoaldos-
teronism and selective renal Kþ secretory problems (e.g.,
post-transplantation or trimethoprim administration). In
patients with impaired renal function, the following are
recommended:
• Oral diuretics with the highest bioavailability and the least
renal metabolism (e.g., torsemide, bumetanide) to minimize
the chance of accumulation and toxicity
• Intravenous agents (short-term) with the least hepatic
metabolism (e.g., furosemide rather than bumetanide)
• Combination loop-thiazide therapy, although this may acti-
vate tubuloglomerular feedback and decrease GFR
• Use of the maximal effective “ceiling” dose
Mineralocorticoids. Fludrocortisone may be useful in treat-
ing chronic hyperkalemia in patients with hypoaldosteronism
with or without hyporeninism, those with SLE, kidney trans-
plant recipients on cyclosporine, and as a preventive agent
in ESRD patients on hemodialysis with interdialytic hyperka-
lemia, although the available data are limited. Mineralocorti-
coids are thought to lower potassium by two mechanisms:
(1) augmentation of colonic potassium excretion, and (2) stim-
ulation of Naþ/Kþ-ATPase on the cell membrane to enhance
extrarenal potassium excretion. The recommended dose is 0.1 163
to 0.3 mg/day. In patients with ESRD on hemodialysis, this
regimen reduces serum Kþ by up to 0.5 to 0.7 mmol/L. Close
monitoring of blood pressure and weight after initiation of
these medications is prudent, especially in patients without CH 7
ESRD.

Disorders of Potassium Metabolism

Cation Exchange Resins. Sodium polystyrene sulfonate
(SPS, Kayexalate) exchanges Naþ for Kþ in the gastrointesti-
nal tract. To prevent constipation and to facilitate the pas-
sage of the resin through the gastrointestinal tract, sorbitol
may be added. The current recommended dose is 15 to 30 g
of powder in water or, preferably, 70% sorbitol, one to four
times daily. It can take from 4 to 24 hours for the full effect
to occur. Therefore, this approach should be used only in
conjunction with other measures in the treatment of acute
hyperkalemia. Each gram of resin binds 0.5 to 1.2 mEq of
Kþ in exchange for 2 to 3 mEq of Naþ. It is worth noting that
a recent, albeit small, study failed to demonstrate any potas-
sium-lowering effect when examining the use of single-dose
resins in ESRD patients on hemodialysis. Kayexalate can also
be administered rectally as a retention enema in patients
unable to take or tolerate the oral form. The recommended
dose is 30 to 50 g of resin every 6 hours administered as a
warm (i.e., body temperature) emulsion in 100 mL 20% dex-
trose in water after an initial cleansing enema (i.e., body tem-
perature tap water) and through a rubber tube secured at
about 20 cm from the rectum and well into the sigmoid
colon. The emulsion should be introduced by gravity,
flushed with an additional 50 to 100 mL of non–sodium-
containing fluid, retained for at least 30 to 60 minutes, and
followed by a cleansing enema (250 to 1000 mL of body tem-
perature tap water). The use of sorbitol in the enema prepara-
tion is not recommended due to multiple reported cases of
colonic necrosis, ischemic colitis, and perforation secondary
to SPS-sorbitol. This complication can also occur with oral
administration of SPS in sorbitol, although the incidence
tends to be much lower. Other potential complications,
although rare, include reduction of serum calcium, volume
overload, interference with lithium absorption, and iatro-
genic hypokalemia. Given the slow onset of action and lim-
ited potassium-lowering effect of resins, their use in the
acute treatment of hyperkalemia is of limited benefit.
Dialysis. Dialysis is the definitive treatment for severe
hyperkalemia in ESRD patients, and may also be required fol-
lowing failure of conservative measures to control severe
hyperkalemia and in settings of massive Kþ release (e.g.,
rhabdomyolysis). Continuous hemodiafiltration is increas-
ingly used for the management of critically ill and hemody-
namically unstable patients; peritoneal dialysis, although
164 less effective in an acute setting, can remove significant
amounts of Kþ over a 24-hour period using multiple
exchanges. However, hemodialysis is the preferred mode
when rapid correction of an acute hyperkalemic episode is
II desired in patients. An average 3- to 5-hour hemodialysis
session removes approximately 40 to 120 mmol of Kþ, with
Disturbances in Control of Body Fluid Volume and Composition

the greatest decline in serum Kþ levels (1.2–1.5 mmol/L)

occurring during the first hour. The amount of potassium
removed depends primarily on the type and surface area of
the dialyzer used, blood flow rate, duration, and serum to
dialysate Kþ gradient. Other factors include the dialysate
concentration of bicarbonate and glucose. Dialysates with a
lower Kþ concentration are more effective in reducing serum
Kþ levels. However, a rapid decline in serum Kþ is associated
with an increased incidence of cardiac arrhythmias, and may
also be associated with rebound hypertension. Therefore,
dialysates with a very low Kþ concentration (0 or 1 mmol/L)
should be used cautiously, particularly in high-risk patients
(digoxin use, history of arrhythmia, coronary artery disease,
left ventricular hypertrophy, hypertension, older age). Con-
tinuous cardiac monitoring is recommended for all patients
dialyzed against a 0- or 1-mmol/L Kþ bath. A graded reduc-
tion in the dialysate Kþ concentration over the course of the
dialysis session may be used, particularly in high-risk
patients. Glucose-free dialysates are more efficient in remov-
ing Kþ, thought to be due to alterations in endogenous insu-
lin levels and associated Kþ shifts. The bicarbonate
concentration of dialysate has also been shown to influence
Kþ removal: the use of high concentration (39 mmol/L) bicar-
bonate dialysate is associated with a more rapid reduction in
serum Kþ, although the total amount of Kþ removal is greater
with low concentration (27 mmol/L) dialysate. A rebound
increase in serum Kþ can occur after hemodialysis. This phe-
nomenon can be especially marked in cases of massive
release from devitalized tissues (e.g., tumor lysis, rhabdomy-
olysis). Other factors attenuating Kþ removal and thus
increasing the risk and magnitude of post-dialysis rebound
include pretreatment with b2-agonists, insulin and glucose,
eating early during the dialysis treatment, a high predialysis
serum Kþ, and higher dialysate Naþ concentrations.

Other Therapeutic Options for Prevention of

Hyperkalemia in ESRD
Strategies to prevent hyperkalemia in the ESRD population are
critically important (Table 7-4) and include dietary restriction
of potassium, and use of b2-agonists and glucose with or with-
out insulin in the prolonged fasting situation (e.g., presurgery).
 165
Table 7-4 Strategies to Prevent Hyperkalemia in the
ESRD Population
Restriction of dietary potassium
Administration of nebulized albuterol and/or 10% dextrose at CH 7
50 mL/hour (add 10 IU insulin per liter if patient is diabetic)

Disorders of Potassium Metabolism

during prolonged fasting
Avoidance of medications that may increase serum potassium,
including Kþ-sparing diuretics, ACE inhibitors, heparin, NSAIDs,
and nonselective b-blockers
Potential therapies requiring further study
Mineralocorticoids, such as fludrocortisone
Bisacodyl
Fosinopril
ACE, angiotensin-converting enzyme; ESRD, end-stage renal disease.
From Putcha N, Allon M: Management of hyperkalemia in dialysis patients. Semin
Dial 20:431–439, 2007.

Use of the ACE inhibitor fosinopril in a group of hemodial-

ysis patients was shown to result in a significant reduction in
predialysis serum potassium levels. Fosinopril is excreted by
both renal and biliary pathways, and it is thought that fosino-
pril may increase potassium secretion via the biliary system,
thereby decreasing interdialytic hyperkalemia.
Daily administration of the laxative bisacodyl has also
undergone trial as a modality for the reduction of interdialytic
potassium. Daily use of 5 to 10 mg of bisacodyl in hemodialy-
sis patients was shown to reduce serum potassium levels, and
although it increased stool frequency, it did not cause overt
diarrhea. Bisacodyl is thought to stimulate cAMP, thereby
activating potassium secretion in the gut.