Chapter 23: Fluid, Electrolyte, & Acid–Base Disorders & Therapy

Total body water (TBW) constitutes 50%–75% of the total body mass, depending on age, sex, and fat content. After an initial brisk postnatal diuresis, the TBW slowly decreases to the adult range near puberty. TBW is divided into the intracellular and extracellular spaces. Intracellular fluid (ICF) accounts for two-thirds of the TBW and extracellular fluid (ECF) for one-third. The ECF is further compartmentalized into plasma (intravascular) volume and interstitial fluid (ISF).

The principal constituents of plasma are sodium, chloride, bicarbonate, and protein (primarily albumin). The ISF is similar to plasma but lacks significant amounts of protein. Conversely, the ICF is rich in potassium, magnesium, phosphates, sulfates, and protein.

An understanding of osmotic shifts between the ECF and ICF is fundamental to understanding disorders of fluid balance. Iso-osmolality is generally maintained between fluid compartments. Because the cell membrane is water-permeable, abnormal fluid shifts occur if the concentration of solutes that cannot permeate the cell membrane in the ECF does not equal the concentration of such solutes in the ICF. Thus, NaCl, mannitol, and glucose (in the setting of hyperglycemia) remain restricted to the ECF space and contribute effective osmoles by obligating water to remain in or be drawn into the ECF compartment. In contrast, a freely permeable solute such as urea does not contribute effective osmoles because it is not restricted to the ECF and readily crosses cell membranes. Tonicity, or effective osmolality, differs from measured osmolality in that it accounts only for osmotically active impermeable solutes rather than all osmotically active solutes, including those that are permeable to cell membranes. Osmolality may be estimated by the following formula:

Although osmolality and osmolarity differ, the former being an expression of osmotic activity per weight (kg) and the latter per volume (L) of solution, for clinical purposes they are similar and occasionally used interchangeably. Oncotic pressure, or colloid osmotic pressure, represents the osmotic activity of macromolecular constituents such as albumin in the plasma and body fluids. The importance of albumin in maintaining intravascular volume status is reflected in the setting of the nephrotic syndrome, protein losing enteropathy, and other low serum albumin states wherein fluids accumulate in the interstitial compartment leading to edema.

The principal mechanisms that regulate ECF volume and tonicity are thirst, vasopressin or antidiuretic hormone (ADH), aldosterone, and atrial natriuretic peptide (ANP), the latter three exerting their influence by their effects on renal water and sodium handling.

Thirst

Water intake is commonly determined by cultural and behavioral factors. Thirst is not physiologically stimulated until plasma osmolality reaches 290 mOsm/kg, a level at which ADH release induces maximal antidiuresis. Thirst provides control over a wide range of fluid states and allows maintenance of appropriate intravascular volume, even in polyuric states (such as central or nephrogenic DI, or obstructive uropathy). An adequate thirst mechanism is central to maintaining adequate fluid balance.

Antidiuretic Hormone

In the kidney, ADH increases water reabsorption in the cortical and medullary collecting ducts, leading to formation of concentrated urine. In the absence of ADH, dilute urine is produced. Typically, ADH secretion is regulated by the tonicity of body fluids rather than intravascular fluid volume; it becomes detectable at a plasma osmolality of 280 mOsm/kg or greater. Tonicity may, however, be sacrificed to preserve ECF volume (ie, with hyponatremic dehydration), wherein ADH secretion and renal water retention are maximal, despite comparatively lower plasma osmolality.

Aldosterone

Aldosterone is released from the adrenal cortex in response to (1) decreased effective circulating volume and resultant stimulation of the renin-angiotensin-aldosterone axis or (2) in response to increasing plasma K⁺. Aldosterone enhances renal tubular reabsorption of Na⁺ in exchange for K⁺, and to a lesser degree H⁺. At a constant osmolality, retention of Na⁺ leads to expansion of ECF volume and suppression of aldosterone release.

Atrial Natriuretic Peptide

ANP, a polypeptide hormone secreted principally by the cardiac atria in response to atrial dilation, contributes to regulation of blood volume and blood pressure. ANP inhibits renin secretion and aldosterone synthesis and causes an increase in glomerular filtration rate and renal sodium excretion. ANP also guards against excessive plasma volume expansion in the face of increased ECF volume by shifting fluid from the vascular to the interstitial compartment. ANP inhibits angiotensin II and norepinephrine-induced vasoconstriction and acts in the brain to decrease the desire for salt and inhibit the release of ADH. Thus, the net effect of ANP is a decrease in blood volume and blood pressure associated with natriuresis and diuresis.

The pH of arterial blood is maintained between 7.38 and 7.42 to ensure that pH-sensitive enzyme systems function normally. Acid–base balance is maintained by interaction of the lungs, kidneys, and systemic buffering systems. Over 50% of the blood’s buffering capacity is provided by the carbonic acid–bicarbonate system, roughly 30% by hemoglobin, and the remainder by phosphates and ammonium. The carbonic acid–bicarbonate system, depicted chemically as

CO₂ + H₂O ↔ H₂CO₃ ↔ H⁺ + HCO₃⁻

interacts via the lungs and kidneys, and in conjunction with the nonbicarbonate systems, to stabilize systemic pH. The concentration of dissolved CO₂ in blood is established by the respiratory system and that of HCO₃⁻ by the kidneys. Disturbances in acid–base balance are initially stabilized by chemical buffering, compensated for by pulmonary or renal regulation of CO₂, and ultimately corrected when the primary cause of the acid–base disturbance is eliminated.

Renal regulation of acid–base balance is accomplished by the reabsorption of filtered HCO₃⁻, primarily in the proximal tubule, and the excretion of H⁺ or HCO₃⁻ in the distal nephron to match the net input of acid or base. When urine is alkalinized, HCO₃⁻ enters the kidney and is ultimately lost in the urine. However, urinary alkalinization will not occur if there is a deficiency of urine Na⁺ or K⁺, as electroneutrality must be maintained. In contrast, the urine may be acidified if an absolute or relative decrease occurs in systemic HCO₃⁻. In this setting, proximal tubular HCO₃⁻ reabsorption and distal tubular H⁺ excretion are maximal. A “paradoxical aciduria” with low urinary pH may be seen in the setting of hypokalemic metabolic alkalosis and systemic K⁺ depletion wherein H⁺ is exchanged and excreted in preference to K⁺ in response to mineralocorticoid. Some of the processes involved in acid–base regulation are shown in Figure 23–1.

Therapy of fluid and electrolyte disorders should be phased to (1) expand the ECF volume and restore tissue perfusion, (2) replenish fluid and electrolyte deficits while correcting attendant acid–base abnormalities, (3) meet the patient’s nutritional needs, and (4) replace ongoing losses.

The cornerstone of therapy involves a detailed understanding of “maintenance” fluid and electrolyte requirements. “Maintenance” requirements call for provision of enough water, glucose, and electrolytes to prevent deterioration of body stores for a euvolemic patient under normal conditions. During short-term parenteral therapy, sufficient glucose is provided to prevent ketosis and limit protein catabolism, although this usually provides little more than 20% of the patient’s true caloric needs. Prior to the administration of “maintenance” fluids, it is important to consider the patient’s volume status and to determine whether intravenous fluids are truly needed.

Various models have been devised to facilitate calculation of “maintenance” requirements based on body surface area, weight, and caloric expenditure. A system based on caloric expenditure is most helpful, because 1 mL of water is needed for each kilocalorie expended. The system presented in Table 23–1 is based on caloric needs and is applicable to children weighing more than 3 kg.

As depicted in Table 23–1, a child weighing 30 kg would need 1700 kcal or 1700 mL of water daily. If the child received parenteral fluids for 2 days, the fluid would usually contain 5% glucose, which would provide 340 kcal/day, or 20% of the maintenance caloric needs. Maintenance fluid requirements take into account normal insensible water losses (Table 23–2) and water lost in sweat, urine, and stool, and assume the patient to be afebrile, at their true dry weight, and relatively inactive. Thus, if excessive losses occur, standard “maintenance fluids” will be inadequate. In contrast, if losses are reduced for any reason, standard “maintenance fluid” administration would be excessive. Maintenance requirements are greater for low-birth-weight and preterm infants. Table 23–3 lists other factors that commonly alter fluid and caloric needs.

Electrolyte losses occur primarily through the urinary tract and to a lesser degree via the skin and stool. Maintenance sodium and potassium electrolyte needs have historically been approximated to be in the 3 mEq Na/100 kcal and 2 mEq K/100 kcal range, leading to the common use of hypotonic intravenous fluids with 77 mEq/L of sodium (½ normal saline) and 20 mEq/L of potassium. Over the past 10 years many authors have drawn attention to the serious problem of hospital-acquired hyponatremia in children with the use of hypotonic IV solutions; notably, hyponatremia is the most common electrolyte abnormality in children and affects approximately 25% of hospitalized pediatric patients. Furthermore, the astute clinician will bear in mind the dynamic nature of clinical context in treating patients and that the choice of IV solution (Table 23–4) and rate of infusion must be made on an individual basis and must be reassessed often. A child with profound water loss stools and hypernatremia who is placed on hypotonic IV fluids and whose diarrhea ceases, but is continued on hypotonic solution without close monitoring of serum electrolytes is at risk for the devastating clinical consequences related to sequelae of rapid changes in sodium balance. In recent years, there also has been a trend for total parenteral nutrition solution sodium and other electrolytes to be calculated and ordered on a milliequivalents-per-kilogram basis rather than the more classic milliequivalents-per-liter basis (eg, 0.2 or 0.45 normal). If the administered fluid volume is decreased in this setting, as the child is weaned from supplemental IV fluids to enteral intake, the sodium and other electrolytes will need to be reduced accordingly to avoid changes in IV fluid tonicity that can result in hypernatremia or other electrolyte derangements.

Patient weight, total output, urinary output, and total fluid input should be monitored daily. Electronic medical records may calculate the net total volume in a 24 hour period but this should be interpreted cautiously as it does not account for the patient’s insensible losses; obtaining daily weights is a corner-stone of appropriate monitoring of fluid and electrolyte balance. If fluid or electrolyte balance is abnormal, serial determination of serum electrolyte concentrations, blood urea nitrogen, and creatinine are necessary; for example, in patients with significant burns, anuria, oliguria, or persistent abnormal stool or urine losses. Serial labs should also be monitored in patients receiving IV fluids or parental nutrition.

DEHYDRATION

Depletion of body fluids is one of the most commonly encountered problems in clinical pediatrics. The clinical evaluation of a child with dehydration, or volume contraction, should focus on the composition and volume of fluid intake and losses (vomiting, diarrhea, urine, “insensible losses”). An updated weight is central to calculating the magnitude of volume depletion. Important clinical features in estimating the degree of dehydration include the capillary refill time, postural blood pressure, and heart rate changes; dryness of the lips and mucous membranes; lack of tears; lack of external jugular venous filling when supine; a sunken fontanelle in an infant; oliguria; and altered mental status (Table 23–5). Children generally respond to a decrease in circulating volume with a compensatory increase in pulse rate and may maintain their blood pressure in the face of severe dehydration. A low or falling blood pressure is, therefore, a late sign of shock in children, and when present should prompt emergent treatment. Salient laboratory parameters include a high urine-specific gravity (in the absence of an underlying renal concentrating defect as seen in diabetes insipidus or chronic obstructive or reflux nephropathy), a relatively greater elevation in blood urea nitrogen than in serum creatinine, a low urinary [Na⁺] excretion (< 20 mEq/L) or fractional excretion of sodium less than 0.1%, and an elevated hematocrit or serum albumin level secondary to hemoconcentration.

Emergent intravenous therapy is indicated when there is evidence of compromised perfusion (inadequate capillary refill, tachycardia, poor color, oliguria, or hypotension). The initial goal is to rapidly expand the plasma volume and to prevent circulatory collapse. A 20-mL/kg bolus of isotonic fluid should be given intravenously as rapidly as possible. Either colloid (5% albumin) or crystalloid (normal saline or Ringer lactate) may be used. Colloid is particularly useful in hypernatremic patients in shock, in malnourished infants, and in neonates. If no intravenous site is available, fluid may be administered intra-osseously through the marrow space. If there is no response to the first fluid bolus, a second bolus may be given. When adequate tissue perfusion is demonstrated by improved capillary refill, decreased pulse rate and improved mental status, deficit replacement may be instituted. If adequate perfusion is not restored after 40 mL/kg of isotonic fluids, other pathologic processes must be considered such as sepsis, occult hemorrhage, or cardiogenic shock. Isotonic dehydration may be treated by providing half of the remaining fluid deficit over 8 hours and the second half over the ensuing 16 hours in the form of 5% dextrose with 0.45% saline containing 20 mEq/L of KCl. In the presence of metabolic acidosis, the addition of sodium or potassium acetate may be considered. Serum potassium and calcium concentrations should be closely monitored, as these can decrease significantly with resolution of acidosis. Maintenance fluids and replacement of ongoing losses should also be provided. Typical electrolyte compositions of various body fluids are depicted in Table 23–6, although it may be necessary to measure the specific constituents of a patient’s fluid losses to guide therapy. If the patient is unable to eat for a prolonged period, nutritional needs must be met through hyperalimentation or enteral tube feedings.

Oral rehydration may be provided to children with mild to moderate dehydration and commercially available solutions provide 45–75 mEq/L of Na⁺, 20–25 mEq/L of K⁺, 30–34 mEq/L of citrate or bicarbonate, and 2%–2.5% of glucose (Table 23–7). Clear liquid beverages found in the home, such as broth, soda, juice, and tea are inappropriate for the treatment of dehydration. Frequent small aliquots (5–15 mL) should be given to provide approximately 50 mL/kg over 4 hours for mild dehydration and up to 100 mL/kg over 6 hours for moderate dehydration. Oral rehydration is contraindicated in children with altered levels of consciousness or respiratory distress who cannot drink freely; in children suspected of having an acute surgical abdomen; in infants with greater than 10% volume depletion; in children with hemodynamic instability; and in the setting of severe hyponatremia ([Na⁺] < 120 mEq/L) or hypernatremia ([Na⁺] > 160 mEq/L). Failure of oral rehydration due to persistent vomiting or inability to keep up with losses mandates intravenous therapy. Successful oral rehydration requires explicit instructions to caregivers and close clinical follow-up of the child.

The type of dehydration is often characterized by the serum [Na⁺]. If relatively more solute is lost than water, the [Na⁺] falls, and hyponatremic dehydration ([Na⁺] < 130 mEq/L) ensues. This is important clinically because hypotonicity of the plasma contributes to further volume loss from the ECF into the intracellular space. Thus, tissue perfusion is more significantly impaired for a given degree of hyponatremic dehydration than for a comparable degree of isotonic or hypertonic dehydration. It is important to note, however, that significant solute losses also occur in hypernatremic dehydration. Furthermore, because plasma volume is somewhat protected in hypernatremic dehydration, it poses the risk of the clinician underestimating the severity of dehydration. Typical fluid and electrolyte losses associated with each form of dehydration are shown in Table 23–8.

HYPONATREMIA

Hyponatremia may be factitious in the presence of high plasma lipids or proteins, which decrease the percentage of plasma volume that is water. Hyponatremia in the absence of hypotonicity also occurs when an osmotically active solute, such as glucose or mannitol, is added to the ECF. Water drawn from the ICF dilutes the serum [Na⁺] despite isotonicity or hypertonicity.

Patients with hyponatremic dehydration generally demonstrate typical signs and symptoms of dehydration (see Table 23–5), as the vascular space is compromised as water leaves the ECF to maintain osmotic neutrality. The treatment of hyponatremic dehydration is fairly straightforward. The magnitude of the sodium deficit may be calculated by the following formula:

Half of the deficit is replenished in the first 8 hours of therapy, and the remainder is given over the following 16 hours. Maintenance and replacement fluids should also be provided. The deficit plus maintenance calculations often approximate 5% dextrose with 0.45% or higher saline. The rise in serum [Na⁺] should not exceed 0.5 mEq/L/h or 6–8 mEq/L/24 h unless the patient demonstrates central nervous system (CNS) symptoms that warrant more rapid initial correction. The dangers of too rapid correction of hyponatremia include cerebral dehydration and injury due to fluid shifts from the ICF compartment, a condition known as osmotic demyelination syndrome.

Hypovolemic hyponatremia also occurs in cerebral salt-wasting (CSW) associated with CNS insults, a condition characterized by high urine output and elevated urinary [Na⁺] (> 40 mEq/L) due to an increase in ANP. CWS is a diagnosis of exclusion represented by an inappropriate natriuresis in a patient with a contracted effective circulatory blood volume in the absence of other causes for Na⁺ excretion. This must be distinguished from the syndrome of inappropriate secretion of ADH (SIADH), which may also manifest in CNS conditions and pulmonary disorders (Table 23–9). In contrast to CSW, SIADH is characterized by euvolemia or mild volume expansion and relatively low urine output due to ADH-induced water retention. Urinary [Na⁺] is high in both conditions, though generally not as high as in SIADH. It is important to distinguish between these two conditions, because the treatment of cerebral salt wasting involves replacement of urinary salt and water losses, whereas the treatment of SIADH involves water restriction. It is also important to remember that in SIADH patients are not necessarily oliguric, and that their urine does not need to be maximally concentrated but merely inappropriately concentrated for their degree of serum tonicity.

In cases of severe hyponatremia (serum [Na⁺] < 120 mEq/L) with CNS symptoms, intravenous 3% NaCl may be given to raise the [Na⁺] by 5 mEq/L to alleviate CNS manifestations and sequelae. In general, 1 mL/kg of 3% NaCl will raise the serum [Na⁺] by about 1 mEq/L. If 3% NaCl is administered, estimated Na⁺ and fluid deficits should be adjusted accordingly. Further correction should proceed slowly, as outlined earlier.

Hypervolemic hyponatremia may occur in edematous disorders such as nephrotic syndrome, congestive heart failure, and cirrhosis, wherein water is retained in excess of salt. Treatment involves restriction of Na⁺ and water and correction of the underlying disorder. Hypervolemic hyponatremia due to water intoxication is characterized by a maximally dilute urine (specific gravity < 1.003) and is also treated with water restriction.

HYPERNATREMIA

Although diarrhea is commonly associated with hyponatremic or isonatremic dehydration, hypernatremia may develop in the presence of persistent fever or decreased fluid intake or in response to improperly mixed rehydration solutions. Extreme care is required to treat hypernatremic dehydration appropriately. If the serum [Na⁺] falls precipitously, the osmolality of the ECF drops more rapidly than that of the CNS and water shifts from the ECF compartment into the CNS to maintain osmotic neutrality. If hypertonicity is corrected too rapidly (a drop in [Na⁺] of > 0.5 mEq/L/h), cerebral edema, seizures, and CNS injury may occur. Thus, following the initial restoration of adequate tissue perfusion using isotonic fluids, a gradual decrease in serum [Na⁺] is desired (6–8 mEq/L/day). This is commonly achieved using 5% dextrose with 0.2% saline to replace the calculated fluid deficit over 48 hours or longer depending on the severity and chronicity of the fluid losses. Maintenance and replacement fluids should also be provided. If the serum [Na⁺] is not correcting appropriately, the free water deficit may be estimated as 4 mL/kg of free water for each milliequivalent of serum [Na⁺] above 145 mEq/L and provided as 5% dextrose. If metabolic acidosis is also present, it must be corrected slowly to avoid CNS irritability. Potassium is provided as indicated. Electrolyte concentrations should be assessed every 2 hours in order to control the decline in serum [Na⁺]. Elevations of blood glucose and blood urea nitrogen may worsen the hyperosmolar state in hypernatremic dehydration and should also be monitored closely. Hyperglycemia is often associated with hypernatremic dehydration and may necessitate lower intravenous glucose concentrations (eg, 2.5%).

Patients with diabetes insipidus, whether nephrogenic or central in origin, are prone to develop profound hypernatremic dehydration as a result of unremitting urinary-free water losses (urine-specific gravity < 1.010), particularly during superimposed gastrointestinal illnesses associated with vomiting or diarrhea. Treatment involves restoration of fluid and electrolyte deficits as described earlier as well as replacement of excessive water losses. Formal water deprivation testing to distinguish responsiveness to ADH should only be done during daylight hours after restoration of normal fluid volume status. The evaluation and treatment of nephrogenic and central diabetes insipidus are discussed in detail in Chapters 24 and 34, respectively.

Hypervolemic hypernatremia (salt poisoning), associated with excess total body salt and water, may occur as a consequence of providing improperly mixed formula, excessive NaCl or NaHCO₃ administration, or as a feature of primary hyperaldosteronism. Treatment includes the use of diuretics, and potentially, concomitant water replacement or even dialysis.

POTASSIUM DISORDERS

The predominantly intracellular distribution of potassium is maintained by the actions of Na⁺-K⁺-ATPase in the cell membranes. Potassium is shifted into the ECF and plasma by acidemia and into the ICF in the setting of alkalosis, hypochloremia, or in conjunction with insulin-induced cellular glucose uptake. The ratio of intracellular to extracellular K⁺ is the major determinant of the cellular resting membrane potential and contributes to the action potential in neural and muscular tissue. Abnormalities of K⁺ balance are potentially life threatening. In the kidney, K⁺ is filtered at the glomerulus, reabsorbed in the proximal tubule, and excreted in the distal tubule. Distal tubular K⁺ excretion is regulated primarily by the mineralocorticoid aldosterone. Renal K⁺ excretion is primarily dependent on the urinary flow rate, and continues for significant periods even after the intake of K⁺ is decreased. Thus, by the time urinary [K⁺] decreases, the systemic K⁺ pool has been depleted significantly. In general, the greater the urine flow the greater the urinary K⁺ excretion.

The causes of hypokalemia are primarily renal in origin. Gastrointestinal losses through nasogastric suction or vomiting reduce total body K⁺ to some degree. However, the resultant volume depletion results in an increase in plasma aldosterone, promoting renal excretion of K⁺ in exchange for Na⁺ reclamation to preserve circulatory volume. Diuretics (especially thiazides and loop diuretics), mineralocorticoids, and intrinsic renal tubular diseases (eg, Bartter syndrome) enhance the renal excretion of K⁺. Systemic K⁺ depletion in hypokalemic metabolic acidosis may lead to “paradoxic aciduria” and low urine pH wherein H⁺ is preferentially exchanged for Na⁺ in response to aldosterone. Clinically, hypokalemia is associated with neuromuscular excitability, decreased peristalsis or ileus, hyporeflexia, paralysis, rhabdomyolysis, and arrhythmias. Electrocardiographic changes include flattened T waves, a shortened PR interval, and the appearance of U waves. Arrhythmias associated with hypokalemia include premature ventricular contractions; atrial, nodal, or ventricular tachycardia; and ventricular fibrillation. Hypokalemia increases responsiveness to digitalis and may precipitate overt digitalis toxicity. In the presence of arrhythmias, extreme muscle weakness, or respiratory compromise, intravenous K⁺ should be given. If the patient is hypophosphatemic ([PO₄^3–] < 2 mg/dL), a phosphate salt may be used. The first priority in the treatment of hypokalemia is the restoration of an adequate serum [K⁺]. Providing maintenance amounts of K⁺ is usually sufficient; however, when the serum [K⁺] is dangerously low and K⁺ must be administered intravenously, it is imperative that the patient have a cardiac monitor. Intravenous K⁺ should generally not be given faster than at a rate of 0.3 mEq/kg/h. Oral K⁺ supplements may be needed for weeks to replenish depleted body stores.

Hyperkalemia—due to decreased renal K⁺ excretion, mineralocorticoid deficiency or unresponsiveness, or K⁺ release from the ICF compartment—is characterized by muscle weakness, paresthesias, and tetany; ascending paralysis; and arrhythmias. Electrocardiographic changes associated with hyperkalemia include peaked T waves, widening of the QRS complex, and arrhythmias such as sinus bradycardia or sinus arrest, atrioventricular block, nodal or idioventricular rhythms, and ventricular tachycardia or fibrillation. An EKG should be obtained when significant hyperkalemia is suspected. If the serum [K⁺] is less than 6 mEq/L, discontinuing K⁺ supplementation may be sufficient if there is no ongoing K⁺ source, such as cell lysis, and if urine output continues. If the serum [K⁺] is greater than 6 mEq/L or if potentiating factors such as renal failure are present, more aggressive therapy is needed (Table 23–10). If electrocardiographic changes or arrhythmias are present, treatment must be initiated promptly. Initial treatment consists of cardiac membrane stabilization and rapid intracellular shifting of K⁺ intracellularly. Intravenous calcium gluconate will rapidly ameliorate depolarization and may be repeated after 5 minutes if electrocardiographic changes persist. Calcium should be given only with a cardiac monitor in place and should be discontinued if bradycardia develops. Administering Na⁺ and increasing systemic pH with bicarbonate therapy will shift K⁺ from the ECF to the ICF compartment, as will therapy with a β-agonist such as albuterol. In nondiabetic patients, 0.5 g/kg of glucose over 1–2 hours will enhance endogenous insulin secretion, lowering serum [K⁺] 1–2 mEq/L. Administration of intravenous glucose and insulin may be needed as a simultaneous drip given over 2 hours with monitoring of the serum glucose level every 15 minutes.

The therapies outlined above provide transient benefits and K⁺ will remain elevated unless other interventions are used to decrease total body potassium. Therapy must be given to reduce K⁺ to normal levels by reestablishing adequate renal excretion using diuretics or optimizing urinary flow, using ion exchange resins such as sodium polystyrene sulfonate that act in the GI tract, or by dialysis.

When evaluating a disturbance in acid–base balance, the systemic pH, partial carbon dioxide pressure (Pco₂), serum HCO₃⁻, and anion gap must be considered. The anion gap, [Na⁺ – (Cl^– + HCO₃⁻)], is an expression of the unmeasured anions in the plasma and is normally 12 ± 4 mEq/L. An increase above normal suggests the presence of an unmeasured anion, such as occurs in diabetic ketoacidosis, lactic acidosis, and salicylate intoxication. Although the base excess (or deficit) is also used clinically, it is important to recall that this expression of acid–base balance is influenced by the renal response to respiratory disorders and cannot be interpreted independently (as in a compensated respiratory acidosis, wherein the base excess may be quite large).

METABOLIC ACIDOSIS

Metabolic acidosis is characterized by a primary decrease in serum [HCO₃⁻] and systemic pH due to the loss of HCO₃⁻ from the kidneys or gastrointestinal tract, the addition of an acid (from external sources or via altered metabolic processes), or the rapid dilution of the ECF with nonbicarbonate–containing solution (usually normal saline). When HCO₃⁻ is lost through the kidneys or gastrointestinal tract, Cl^– must be reabsorbed with Na⁺ disproportionately, resulting in a hyperchloremic acidosis with a normal anion gap. Thus, a normal anion gap acidosis in the absence of diarrhea or other bicarbonate-rich gastrointestinal losses suggests the possibility of renal tubular acidosis and should be evaluated appropriately (see Chapter 24). In contrast, acidosis that results from the addition of an unmeasured acid is associated with a widened anion gap. Examples are diabetic ketoacidosis, lactic acidosis, starvation, uremia, toxin ingestion (salicylates, ethylene glycol, or methanol), and certain inborn errors of organic or amino acid metabolism. Dehydration may also result in a widened anion gap acidosis as a result of inadequate tissue perfusion, decreased O₂ delivery, and subsequent lactic and keto acid production. Respiratory compensation is accomplished through an increase in minute ventilation and a decrease in Pco₂. The patient’s history, physical findings, and laboratory features should lead to the appropriate diagnosis.

The ingestion of unknown toxins or the possibility of an inborn error of metabolism (see Chapter 36) must be considered in children without an obvious cause for a widened anion gap acidosis. Unfortunately, some hospital laboratories fail to include ethylene glycol or methanol in their standard toxicology screens, so assay of these toxins must be requested specifically. This is of critical importance when therapy with fomepizole (4-methylpyrazole) must be considered for either ingestion—and instituted promptly to obviate profound toxicity. Ethylene glycol (eg, antifreeze) is particularly worrisome because of its sweet taste and accounts for a significant number of toxin ingestions. Salicylate intoxication has a stimulatory effect on the respiratory center of the CNS; thus, patients may initially present with respiratory alkalosis or mixed respiratory alkalosis and widened anion gap acidosis.

Most types of metabolic acidosis will resolve with correction of the underlying disorder, improved renal perfusion, and acid excretion. Intravenous NaHCO₃ administration may be considered in the setting of metabolic acidosis when the pH is less than 7.2 or the [HCO₃⁻] is less than 6 mEq/L, but only if adequate ventilation is ensured. The dose (in milliequivalents) of NaHCO₃ may be calculated as

Weight (kg) × Base deficit × 0.3

and given as a continuous infusion over 1 hour. The effect of NaHCO₃ in lowering serum potassium and ionized calcium concentrations must also be considered and monitored.

METABOLIC ALKALOSIS

Metabolic alkalosis is characterized by a primary increase in [HCO₃⁻] and pH resulting from a loss of strong acid or gain of buffer base. The most common cause for a metabolic alkalosis is the loss of gastric juice via nasogastric suction or vomiting. This results in a Cl⁻-responsive alkalosis, characterized by a low urinary [Cl^–] (< 20 mEq/L) indicative of a volume-contracted state that will be responsive to the provision of adequate Cl^– salt (usually in the form of normal saline). Cystic fibrosis may also be associated with a Cl^–-responsive alkalosis due to the high losses of NaCl through the sweat, whereas congenital Cl^–-losing diarrhea is a rare cause of Cl^–-responsive metabolic alkalosis. Chloride-resistant alkaloses are characterized by a urinary [Cl^–] greater than 20 mEq/L and include Bartter syndrome, Cushing syndrome, and primary hyperaldosteronism, conditions associated with primary increases in urinary [Cl^–], or volume-expanded states lacking stimuli for renal Cl^– reabsorption. Thus, the urinary [Cl^–] is helpful in distinguishing the nature of a metabolic alkalosis, but must be specifically requested in many laboratories because it is not routinely included in urine electrolyte panels. The serum [K⁺] is also low in these settings (hypokalemic metabolic alkalosis) owing to a combination of increased mineralocorticoid activity associated with volume contraction, the shift of K⁺ to the ICF compartment, and preferential reabsorption of Na⁺ rather than K⁺ to preserve intravascular volume. A hypokalemic alkalosis seen in the setting of primary mineralocorticoid excess would be expected to be associated with systemic hypertension clinically, as observed in an adrenal adenoma and some forms of monogenic hypertension, including Liddle syndrome and apparent mineralocorticoid excess.

RESPIRATORY ACIDOSIS

Respiratory acidosis develops when alveolar ventilation is decreased, increasing Pco₂, and lowering systemic pH. The kidneys compensate for respiratory acidosis by increasing HCO₃⁻ reabsorption, a process that takes several days to fully manifest. Patients with acute respiratory acidosis frequently demonstrate air hunger with retractions and the use of accessory respiratory muscles. Respiratory acidosis occurs in upper or lower airway obstruction, ventilation-perfusion disturbances, CNS depression, and neuromuscular defects. Hypercapnia is not as detrimental as the hypoxia that usually accompanies these disorders. The goal of therapy is to correct or compensate for the underlying pathologic process to improve alveolar ventilation. Bicarbonate therapy is not indicated in a pure respiratory acidosis, because it will worsen the acidosis by shifting the equilibrium of the carbonic acid–bicarbonate buffer system to increase Pco₂.

RESPIRATORY ALKALOSIS

Respiratory alkalosis occurs when hyperventilation results in a decrease in Pco₂ and an increase in systemic pH. Depending on the acuity of the respiratory alkalosis, there may be an associated compensatory loss of bicarbonate by the kidneys manifested as a low serum bicarbonate level and a normal anion gap that may be misinterpreted as a normal anion gap acidosis if all acid–base parameters are not considered. Patients may experience tingling, paresthesias, dizziness, palpitations, syncope, or even tetany and seizures due to the associated decrease in ionized calcium. Causes of respiratory alkalosis include psychobehavioral disturbances, CNS irritation from meningitis or encephalitis, salicylate intoxication, and iatrogenic over ventilation in patients who are mechanically ventilated. Therapy is directed toward the causal process. Rebreathing into a paper bag will decrease the severity of symptoms in acute hyperventilation.