Chapter 22. Electrolyte Disorders

Electrolyte disorders can be some of the most complex and subtle clinical conditions facing the critical care or emergency medicine physician. A healthy degree of suspicion coupled with vigilant electrolyte monitoring is necessary in order to avoid missing these disorders. This is particularly true as many electrolyte disorders occur secondary to other severe disease states.

Sodium disorders are commonly encountered in clinical practice. Both hyponatremia and hypernatremia have multiple underlying causes and may develop acutely or chronically. Patients with acute or severe sodium disorders may be critically ill and require rapid, aggressive correction of their sodium abnormality, while aggressive treatment of the chronic, compensated hyponatremic or hypernatremic patient may cause dangerous fluid shifts. It is critical that the emergency physician understand how to identify, classify, and treat sodium disorders.

Hyponatremia

Introduction

Hyponatremia is defined as a serum sodium less than 135 mEq/L. It is commonly found in both inpatients and outpatients.1 Even mild hyponatremia in outpatients is correlated with a poor outcome.2 Groups at particular risk for hyponatremia include hospitalized patients, elderly patients, and patients recently started on thiazide diuretics.3

Presentation

The severity of symptoms due to hyponatremia depends on the rate of sodium decline as well as the absolute level. Mildly hyponatremic patients are often asymptomatic. Moderately hyponatremic patients (Na of 125–130 mEq/L) may experience nausea, headache, malaise, and myalgias and have depressed tendon reflexes. Severe hyponatremia (Na <125 mEq/L) causes mental status changes; seizure, coma, and death occur at sodium levels of below 120 mEq/L. Acute hyponatremia occurs over less than 48 hours and is likely to cause neurologic manifestations secondary to cerebral edema. In chronic hyponatremia, neurologic effects are less likely as there has been time for compensation and brain size remains normal. This compensatory response puts the patient at risk for a demyelinating syndrome if sodium is corrected too rapidly.

Evaluation

Evaluation of hyponatremia consists of a stepwise narrowing of the differential diagnosis based on history, physical exam, and laboratory testing1,4–7 (see Figure 22-1).

Hyperosmolar or Iso-Osmolar Hyponatremia

Hyponatremia in the absence of a hypoosmolar state is referred to as pseudohyponatremia. Hyperosmolar hyponatremia occurs when large amounts of an osmotically active substance (such as mannitol, glucose, or intravenous [IV] contrast dye) draw water into the vasculature and dilute sodium concentration. For pseudohyponatremia secondary to hyperglycemia, a correction factor of 1.6 mEq/L sodium decrease for every 100 mg/dL rise in glucose is often used; however, experimental evidence indicates that a correction factor of 2.4 mEq/L may be more accurate.8

Iso-osmolar hyponatremia may be due to severe hyperlipidemia or hyperproteinemia, which causes a lab artifact in which serum water content is overestimated. Large volume irrigation with a sodium-free fluid such as sorbitol, which is frequently used in transurethral prostatic resection, can cause either an iso-osmolar or hypoosmolar hyponatremia.9

Hypoosmolar Hyponatremia

Hypoosmolar hyponatremia is due to the net gain of free water. This may occur when water intake overwhelms the excretory capacity of the kidney or when there is an inability to suppress antidiuretic hormone (ADH) secretion secondary to true volume depletion, effective circulating volume depletion, or a primary inappropriate release of ADH.

Evaluation of the patient with hypoosmolar hyponatremia begins with an assessment of volume status based on vital signs including orthostatics, skin turgor, mucus membranes, JVD, and presence or absence of edema and ascites. Because physical exam findings can be unreliable in assessing volume status,10 the patient's history, comorbidities, medications, and laboratory values should be incorporated into the assessment as well.

Hypovolemic hyponatremia occurs when sodium depletion exceeds free water depletion. Hypovolemia prompts ADH release and thirst, so the patient consumes and retains free water, leading to worsened hyponatremia.

Relative sodium depletion may occur by renal or extrarenal losses of sodium and water; distinguishing between the two is based on history and measurement of urine sodium. Low urine Na (<20 mEq/L) indicates that the kidneys are appropriately reabsorbing sodium, so losses are extrarenal. Causes of extrarenal loss include vomiting, diarrhea, and cutaneous loss in burn victims. Patients presenting to the ED with hyponatremia most commonly fall into this category.11 If urine sodium is high (>20 mEq/L), then sodium is being lost renally out of proportion to water losses. The most common cause is diuretics, and the emergency physician must be particularly alert for this condition in older patients recently started on hydrochlorothiazide. Other causes include sodium-wasting nephropathy or hypoaldosteronism.

Cerebral salt wasting syndrome (CSW) is a cause of hypovolemic hyponatremia that occurs after head injury or neurosurgical procedures. Distinguishing it from the syndrome of inappropriate antidiuretic hormone (SIADH), which may occur in the same clinical context, is critical as the treatment of CSW (administration of isotonic saline) may worsen SIADH. An assessment of the patient's volume status may help distinguish the two: patients with CSW are more likely to appear hypovolemic, whereas patients with SIADH are more likely to be euvolemic.6 In addition, CSW patients typically produce dilute urine with a high flow rate, unlike the concentrated, low-flow urine of SIADH. Spot measurement of urinary sodium cannot distinguish between these two diseases. Additional assessment of volume status, such as degree of hemoconcentration, BUN and creatinine levels, and central venous pressure, should be considered as well.12

Euvolemic hyponatremia occurs when there is free water gain and/or retention with minimal sodium loss. The SIADH is diagnosed when there is a hypoosmolar hyponatremia with an inappropriately concentrated urine (>100 mOsm/kg), clinical euvolemia, absence of diuretic use, and normal renal, cardiac, adrenal, and thyroid functions. It is a frequent cause of hyponatremia in hospitalized patients and may be due to malignancy, pulmonary disease, CNS disease, or drugs13 (see Table 22-1). Hypothyroidism and adrenal insufficiency may cause a similar clinical picture.

Euvolemic hyponatremia may also occur in the patient with normal kidney function if the amount of free water consumed is excessive (often >4 L per day). Etiologies include psychogenic polydipsia or beer potomania. An appropriately dilute urine (<100 mOsm/kg) is present as the kidney attempts to excrete free water.

Hypervolemic hyponatremia occurs in patients who are total body volume overloaded but have low effective arterial volume secondary to heart failure, cirrhosis, renal failure, or nephrotic syndrome. Hyponatremia portends a poor prognosis in heart failure patients14 and patients with cirrhosis.15

Treatment

Treatment of hyponatremia depends on both the etiology and the severity of the disease. Osmotic demyelination syndrome is a dreaded complication of treatment resulting when administration of hypertonic solution causes rapid movement of water out of brain cells. The emergency physician must balance the concern for this entity with the need to treat suspected cerebral edema.

The critically ill hyponatremic patient who presents with seizure or coma should be treated aggressively to rapidly correct sodium to a safe level (>120 mEq/L). This is accomplished by administration of hypertonic saline for the first 3–4 hours after presentation or until symptoms improve. The goal rate of correction is initially 1.5–2 mEq/L/h; once symptoms improve, the rate of correction should be slowed, and should not exceed 10 mEq/L over the first 24 hours. Severe hyponatremic symptoms are likely to be due to a rapid fall in sodium without time for the brain to compensate, so rapid correction in these patients is less likely to cause a demyelinating syndrome than rapid correction in chronically hyponatremic patients.

Treatment of the stable hyponatremic patient should be based on etiology. Hypovolemic patients require isotonic fluids, while euvolemic patients are treated with free water restriction. Hypervolemic patients may require diuresis in addition to free water restriction. Conivaptan, an oral V1A/V2 receptor antagonist, has been shown to increase serum sodium in patients with euvolemic or hypervolemic hyponatremia.16

The stable hyponatremic patient should be corrected at a rate of no more than 0.5 mEq/L/h and no more than 10–12 mEq/L/24h6 to minimize risk of demyelination. The change in serum sodium estimated to result from administration of 1 L of IV fluid is calculated by the following equation:

Correction factor:

The sodium content of common IV solutions is noted in Table 22-2. Serum sodium should be measured frequently during treatment, as calculations have limited accuracy and response to therapy is variable.17

Hypernatremia

Introduction

Hypernatremia is defined as a serum sodium >145 mEq/L. The underlying etiology is inadequate free water intake. This condition is rare in alert people with an intact thirst mechanism; it is more prevalent in those who rely on others for their water intake. Patients over age 60 may be particularly at risk, since the protective responses of thirst and ADH release are blunted in older age. Most outpatients who present with hypernatremia are at the extremes of age, while hospitalized patients or those with altered mental status are at risk for hypernatremia regardless of age.18

Presentation

Hypernatremia may present with weakness, agitation, muscle twitching, hyperreflexia, ataxia, lethargy, and coma. As with hyponatremia, symptoms in hypernatremia correlate with severity and rate of sodium change. In acute or severe hypernatremia, brain shrinkage may lead to vascular rupture with cerebral bleeding. In chronic hypernatremia, the brain adapts over time by accumulation of electrolytes, so neurologic symptoms may be less pronounced; however, this adaptive response complicates management because rapid correction of hypernatremia may lead to cerebral edema.18

Evaluation

As with hyponatremia, an assessment of the patient's volume status will help to clarify the etiology of hypernatremia (see Figure 22-2).

Hypovolemic hypernatremia occurs when free water depletion exceeds sodium depletion. Similar to hypovolemic hyponatremia secondary to relative sodium depletion, relative free water depletion may have a renal or extrarenal cause. Renal losses are due to an osmotic diuresis caused by hyperglycemia, mannitol, or a postobstructive state. In this case, free water is lost in a dilute urine (urine osmolality <700 mOsm/kg) with concomitant renal loss of sodium (urine Na >20 mEq/L). Extrarenal losses may be due to diarrhea, NG suction, vomiting, third spacing, or skin losses as in a burn patient; urine is appropriately concentrated (>700 mOsm/kg) and renal sodium is low (<10 mEq/L).

Euvolemic hypernatremia is caused by free water loss without significant sodium loss. The cause may be renal due to diabetes insipidus (DI) or extrarenal.

DI, in which the kidney does not concentrate the urine appropriately (urine osmolality <700 mOsm/kg), will cause hypernatremia if there is insufficient access to free water. Central DI is due to decreased ADH release secondary to head trauma, neurosurgery, infiltrative disease, or an idiopathic cause. Administration of exogenous ADH will allow these patients to concentrate the urine again. In nephrogenic DI, the collecting ducts are resistant to ADH, so exogenous administration is ineffective. This condition is frequently drug related, with the most common causes being lithium, foscarnet, and clozapine. Other causes include hypercalcemia, hypokalemia, low-protein diet, and release of ureteral obstruction.19

Extrarenal causes include insensible losses or hypodipsia, and present with an appropriately concentrated urine (>700 mOsm/kg).

Hypervolemic hypernatremia is caused by excessive sodium gain. It is frequently iatrogenic and may be due to excessive dosing of sodium bicarbonate, dialysis, salt tabs, or overcorrection with hypertonic saline.

Treatment

If the hypernatremic patient is unstable due to hypovolemia, volume resuscitation with normal saline should be aggressive until the patient is hemodynamically stable. After stabilization, fluids should be switched to 0.45% saline and correction rate monitored as described below. Although cerebral edema is a concern with rapid correction, the importance of correcting volume status outweighs the risk of this side effect. Additionally, these patients probably became hypernatremic acutely, so there has been less time for compensation and therefore lower risk of inducing cerebral edema. There should be a low threshold to get a head CT on the hypernatremic patient with neurologic deficit, as cerebral bleeding secondary to brain shrinkage and traction on cerebral vessels may occur.

As with hyponatremia, treatment of the stable hypernatremic patient will depend on the suspected etiology and volume status. The euvolemic patient may be treated with hypotonic saline; if central DI is suspected, vasopressin may be used. Hypervolemic hypernatremic patients require loop diuretics and free water replacement.

The stable hypernatremic patient should be corrected gradually to minimize the risk of cerebral edema, with a goal rate of 0.5 mEq/L/h and a maximum decrease of 10 mEq/L over a 24-hour period.1 Expected sodium change with 1 L of fluid may be calculated by the same formula used for hyponatremia:

Correction factor:

Total free water deficit may be calculated by the following formula:

The sodium content of common IV solutions is noted in Table 22-2. No more than half the free water deficit should be corrected in the first 24 hours, with the remainder corrected over the next 2–3 days. This calculation does not account for ongoing insensible losses. As in the correction of hyponatremia, predictive calculations may not be accurate, so sodium should be frequently monitored. Neurologic status should be assessed frequently, as acute changes may indicate the development of cerebral edema.

Potassium disorders are the most frequently observed electrolyte disorders of hospitalized patients,20 and probably of patients seen in the emergency department as well.21 Given the predominantly intracellular distribution of potassium in the body, and in turn the inability to truly correct serum potassium levels acutely, treatment for disorders of potassium is by necessity a protracted endeavor.

Total body potassium stores average 50–55 mEq/kg body weight.22 Of this, approximately 98% is contained intracellularly, and about 75% of the intracellular potassium is found in muscle tissue.21 The remaining 2% is extracellular, and only about 0.4% of total body potassium can be found in the plasma. The concentration of potassium in the plasma is maintained in a fairly narrow range of 3.5–5.0 mmol/L. The intracellular concentration of potassium averages about 150 mmol/L. The rather steep gradient of potassium across the cell membrane is maintained by the sodium–potassium ATPase pump. It is predominantly this gradient of potassium across the cell membrane that maintains resting cell membrane potential.

Alterations in the ratio of intracellular potassium concentration to extracellular potassium concentration, [K+]c\[K+]e, in turn alter the resting membrane potential.22 The excitability of a membrane is defined as the difference between the resting and threshold potentials. If alterations to the [K+]c\[K+]e occur that increase the difference between the resting and threshold potentials, a decreased excitability is seen. Likewise, with changes in [K+]c\[K+]e that decrease the difference between the resting and threshold potentials, an increased excitability is seen.

The effects of hypokalemia or hyperkalemia on resting membrane potential, and in turn the degree to which symptoms or complications occur, are more directly related to the [K+]c\[K+]e than to the serum potassium concentration. A gradual reduction in total body potassium, where the intracellular concentration is decreased proportional to the reduction in extracellular concentration, will alter membrane excitability less than an acute change in serum potassium, where the intracellular concentration is not allowed to equilibrate. For this reason, alterations that occur due to transcellular shift of potassium into or out of cells are more likely to manifest symptoms than those that result from gradual losses or accumulations.

Hypokalemia

Hypokalemia is defined as a serum potassium concentration of less than 3.5 mmol/L. Relatively mild hypokalemia, between 3.0 and 3.5 mmol/L, is often well tolerated by healthy people. In individuals with heart disease, particularly heart failure, even mild hypokalemia has been demonstrated to increase morbidity and mortality.23–25 More severe hypokalemia may be associated with generalized symptoms of fatigue, weakness, and constipation. At levels under 2.5 mmol/L, muscle necrosis can occur, and at levels less than 2.0 mmol/L, an ascending paralysis, including respiratory failure, can develop. While usually not arrhythmogenic in healthy individuals, hypokalemia can induce dysrhthmias in those with underlying heart disease. Hypokalemia is well known to exacerbate the arrhythmogenic properties of digoxin.25

Hypokalemia can arise from deficient intake of potassium, increased excretion of potassium, or a transcellular shift of extracellular potassium into cells. See Table 22-3 for a list of potential causes of hypokalemia.

The typical American diet contains a surfeit of potassium, and, as such, hypokalemia due to decreased intake is rare. In starvation states, or more often in critically ill patients without adequate diet repletion, it is possible to elicit a potassium-deficient state in a matter of days. Adequate repletion usually prevents this.

Low serum potassium levels can be caused by total body depletion brought about by excessive potassium losses not matched by intake. Potassium losses typically occur via either renal or gastrointestinal routes. Renal potassium losses associated with diuretic use are the most common cause of hypokalemia. Nasogastric drainage can cause hypokalemia by depleting chloride levels. Concomitant magnesium depletion exacerbates renal losses, as low magnesium levels inhibit the kidney's ability to reabsorb potassium at the distal tubule. Assessment of urinary chloride levels can help to distinguish between causes of renal potassium losses: an elevated urine chloride (>25 mEq/L) is associated with magnesium depletion or diuretic-induced losses, while a low urine chloride (<15 mEq/L) is associated with nasogastric drainage or alkalosis-induced potassium losses.

Gastrointestinal losses of potassium arise due to diarrhea. The potassium concentration of stool is about 75 mEq/L. In states of excessive stool volume loss, potassium losses can quickly add up.

Several factors can promote a transcellular shift of potassium from the extracellular space into the cells, leading to a low serum potassium level despite normal total body potassium stores, including hypothermia, alkalosis, and multiple medications—such as insulin and any of the sympathomimetics. Profound hypothermia may also present with immediate or delayed hyperkalemia as the result of tissue death.

Many medications induce hypokalemia. In the ED or ICU, β-agonists are the most common culprit, although their effect at therapeutic doses is minimal, typically resulting in reductions in serum potassium levels of less than 0.5 mEq/L.26 Other medications known to cause hypokalemia include loop and thiazide diuretics, osmotic diuretics, carbonic anhydrase inhibitors, adrenocortical steroids, natural penicillins, aminoglycosides, and amphotericin B.27 See Table 22-3 for a list of potential causes of hypokalemia.

Treatment of the Hypokalemic Patient

Treatment of the hypokalemic patient should begin with identification and correction of any causes for intracellular shifts of potassium. Magnesium levels should be checked and hypomagnesemia corrected prior to attempts at potassium repletion, as hypomagnesemia will preclude effective potassium repletion. If true potassium depletion is thought to exist, potential causes should be identified, beginning with a thorough review of the patient's medication profile. If causes of ongoing potassium loss can be mitigated, do so. It is often the case that treatments iatrogenically causing potassium depletion are required due to the patient's underlying disease processes. In these cases, attempts should be made to establish adequate ongoing potassium supplementation in order to prevent repeat episodes of hypokalemia. Repletion of potassium must be done relatively slowly allowing for stable equilibration between the intracellular and extracellular compartments. Too rapid repletion of potassium will cause rapid rises in serum potassium concentrations, and in turn acutely increase the [K+]c\[K+]e. IV repletion of potassium is routinely performed using potassium chloride solutions at a rate not in excess of 20 mEq/h. In cases of severe symptomatic potassium depletion, rates as high as 100 mEq/h have been reported as being used without ill effect.28 Due to the irritating properties of the hyperosmotic potassium chloride solutions, infusion via a large central vein is preferred.

Oral potassium supplementation is better suited to ongoing supplementation than to rapid repletion. Oral administration has the advantage that the rate of GI absorption inherently limits rapid changes in serum potassium concentrations. Oral potassium chloride salts can be used, or, in the context of concomitant hypophosphatemia, potassium phosphate salts.

Given the difficulty in estimating total body potassium deficit from the serum potassium level, it should come as no surprise that estimating the total amount of repletion required is difficult at best. In addition, the serum concentration to total body potassium stores relationship is not a linear one. Ongoing losses of potassium during attempts at repletion make calculating total repletion doses required more difficult. Due to the large intracellular distribution of potassium and the necessity of slow repletion, several days of repletion are often necessary in order to normalize serum concentrations safely, and the best way to track the amount of repletion required is to monitor repeated serum potassium concentrations. Due to the nonlinear nature of the serum potassium concentration to total body potassium levels, early repletion is likely to have minimal effects on serum concentration while smaller amounts of repletion will have more dramatic effects on serum concentration as the patient's total body potassium concentrations approach normal. As a general rule, however, a total of 175 mEq of potassium, given over several doses, should be needed for every 0.5 mEq/L decrease in the serum potassium level.29

Hyperkalemia

Elevated serum potassium levels are potentially arrhythmogenic due to their destabilizing effect on the myocardial cell membrane. See Table 22-4 for a list of potential causes of hyperkalemia.

Hyperkalemia may be due to inability to excrete excess potassium, as occurs in renal failure, or due to a release of intracellular potassium into the extracellular space. Release of intracellular potassium stores can occur either as a result of a transcellular shift, as mentioned above, or as a result of cell ischemia and necrosis. Acutely ischemic skeletal muscle or gut, given the high concentration of potassium in muscle cells, can cause dramatic releases of potassium. Complications due to the rapid development of hyperkalemia following reperfusion of ischemic tissue are often cited as a proximate cause of death following crush injury. Hyperkalemia is rarely caused by excessive intake, except when associated with renal failure. The effects of hyperkalemia can be seen transiently if too rapid repletion of potassium is undertaken, without allowing time for the extracellular and intracellular compartments to equilibrate. Multiple medications can induce hyperkalemia, including cation-exchange resins, citrate, spironolactone/amelioride, triamterene, trimethoprim, digoxin, angiotensin-converting enzyme inhibitors, angiotensin receptor blockers, heparin, and succinylcholine.27 See Table 22-4 for a list of potential causes of hyperkalemia.

The immediate concern following the identification of hyperkalemia is the need to stabilize myocardial cell membranes in order to prevent arrhythmic complications. A 12-lead EKG should be obtained, and examined for evidence of cell membrane instability. EKG abnormalities associated with mild hyperkalemia include large-amplitude T waves, and “peaked” or “tented” T waves. Moderate hyperkalemia can result in PR interval prolongation, decreased P-wave amplitude or disappearance, QRS complex widening, and conduction blocks with escape beats. Severe hyperkalemia can result in a sine-wave pattern, ventricular fibrillation, and eventually asystole30 (see Figure 22-3).

If hyperkalemia has produced electrocardiographic evidence of myocardial instability, administration of calcium-containing salt solution should be undertaken immediately, as calcium will help stabilize the myocardial membrane, while subsequent efforts are made to normalize the serum potassium concentrations. Calcium is commonly available in calcium chloride and calcium gluconate solutions. Calcium chloride should only be infused via a large central line, as it is very hyperosmolar. Administration of 10 mL of 10% calcium chloride or 30 mL of 10% calcium gluconate will have effects within minutes, but those effects will only last 30–60 minutes, requiring that other acute interventions be undertaken in that interval.

Treatments for hyperkalemia can be divided into those that temporize the hyperkalemia by inducing an intracellular shift of potassium and those that remove total body potassium. The former may be more appropriate if the hyperkalemia is suspected to be due to transcellular shifts of potassium as opposed to a total body excess of potassium.

Aggressive hydration with isotonic normal saline IV solution will help dilute the serum potassium, as well as promote diuresis and renal excretion of potassium if the patient is not oliguric. In the context of ischemia, aggressive resuscitation will help minimize ongoing potassium release by restoring perfusion.

Sodium bicarbonate induces an intracellular shift of potassium by inducing relative alkalemia. Because sodium bicarbonate and calcium salt solutions can precipitate when coadministered, some authors have advocated not using sodium bicarbonate if calcium salt administration is being performed. The risks of coadministration of sodium bicarbonate and calcium salts can be minimized by administering them through separate IV lines and/or temporally spacing their respective administration if the advantages of sodium bicarbonate administration are felt to be warranted.

β-Agonists cause an intracellular shift of potassium. Because β-agonists have arrhythmogenic properties of their own, some caution may be advised in their utilization, particularly in the presence of hyperkalemic EKG changes. The use of β2-specific agonists, and albuterol in particular, can lower the serum potassium concentration acutely.

Insulin causes an intracellular shift of potassium. Because of insulin's obvious effect on serum blood sugars, it usually requires the coadministration of a dextrose-containing solution to prevent hypoglycemia. Typical dosing is 10 U of regular insulin administered with one or two ampoules of 50% dextrose in water (25–50 g D50).

Sodium polystyrene sulfonate (SPS) is a gastrointestinal-binding resin that exchanges sodium cations for potassium cations in the intestinal lumen, resulting in the fecal excretion of potassium. It does not work acutely as it requires the progressive equilibration of potassium across the intestinal mucosa. Effects are usually seen within 4–6 hours. Faster onset may be seen with the use of a retention enema of SPS. It is used more typically in the chronic renal failure patient in order to help prevent hyperkalemia. When initiated early in the acutely hyperkalemic patient, SPS may help prevent later rebound hyperkalemia after the effect of other acute interventions has begun to wane.

Diuretics are a frequent cause of hypokalemia, so it should be no surprise that they are employed to counter hyperkalemia. Non–potassium-sparing diuretics, most commonly furosemide, induce renal excretion of potassium. When using diuretics to remove excess potassium, replace urinary fluid output with isotonic IV fluid.

Hemodialysis will acutely clear elevated serum potassium levels as well as normalize hyperkalemia-inducing metabolic acidosis, and is the treatment of choice for acute severe hyperkalemia. The rate of clearance will vary depending on the dosing of dialysis and the diasylate chosen. Patient's electrolytes should be monitored closely following hemodialysis, as rebound hyperkalemia often occurs as the intracellular and extracellular compartments reequilibrate. Repeat runs of hemodialysis or a steady state of continuous renal replacement therapy may be required. As the initiation of hemodialysis often requires a significant amount of time, even in the well-resourced facilities, its availability should not preclude the early application of other more immediately available interventions.

Magnesium is the second most prevalent intracellular cation, next to potassium. It serves as a cofactor for enzymatic reactions involving adenosine triphosphate (ATP), including the sodium–potassium ATPase pump responsible for maintaining cell membrane potential. Magnesium also regulates the transport of calcium into smooth muscle cells.

Total body magnesium stores are approximately 24 g in the adult. Only about 1% of this is located in the extracellular fluid compartment, making imputation of the total body magnesium status from serum measures very difficult.31

Magnesium Deficiency

Effects of Magnesium Deficiency

There are no characteristic clinical findings specific for magnesium deficiency. Hypokalemia and/or hypocalcemia suggest the possibility of magnesium depletion, as low magnesium levels will impair renal absorption of potassium and the secretion of parathyroid hormone (PTH), respectively.

Cardiac dysrhythmias are associated with magnesium depletion, as magnesium is a cofactor in the sodium–potassium ATPase pump. Depletion of magnesium can result in depolarized myocytes and predispose to tachydysrhythmias. Magnesium depletion will accentuate the effects of digitalis toxicity, as both agents have their effect on the sodium–potassium ATPase membrane pump.32 This is of particular concern as patients on digitalis are often also prescribed diuretics. Magnesium deficiency is associated classically with torsades de pointes, where the rapid administration of magnesium is a first-line therapy.

Magnesium's effect on smooth muscle has led to its use as an adjunctive therapy in severe asthma, although to what degree magnesium depletion might be contributory to the exacerbation of that disease state is unclear.

Causes of Magnesium Deficiency

Like potassium, low magnesium levels may be due to intracellular shifts of magnesium, or due to renal or gastrointestinal losses. Non–potassium-sparing diuretics, and in particular the loop diuretics, impair magnesium reabsorption and result in renal losses. Lower GI losses due to diarrhea result in loss of magnesium in the stool. Unlike potassium, however, upper GI losses due to vomiting or gastric drainage rarely lead to significant magnesium losses. Drug-induced causes of renal magnesium loss other than diuretics include use of aminoglycosides, amphotericin, pentamidine, cisplatin, and cyclosporine.27 Digitalis, insulin, and epinephrine can all cause intracellular shifts in magnesium. Absent excessive losses, low magnesium levels are not frequently due to poor dietary intake in developed countries. An exception to this has been in the alcoholic population, where magnesium deficiency due to poor diet may impair attempts at thiamine repletion, as magnesium is a cofactor in the metabolism of thiamine into thiamine pyrophosphate.33 See Table 22-5 for a list of potential causes of magnesium deficiency.

Diagnosis of Magnesium Deficiency

To even a greater degree than with potassium, the serum magnesium level is a poor measure of total body magnesium stores. The majority of magnesium is intracellular or within bone, and of the extracellular magnesium, much is protein bound and inert. Serum magnesium levels do not distinguish between ionized and bound forms of magnesium, and unlike calcium, specific ionized magnesium assays are not routinely available in most labs.34

Low serum levels of magnesium almost always reflect deficiency and warrant repletion, but many patients with deficiencies of magnesium have normal levels of serum magnesium. In the absence of renal disease or of renal losses of magnesium, the urine magnesium level may be a useful measure.

Often the only indication that magnesium may need repletion is the clinical suspicion brought about by identifying a predisposing condition. Patients on loop diuretics, those with refractory hypokalemia or hypocalcemia, anyone with an osmotic diuresis, those with lower GI losses due to diarrhea, and alcoholics undergoing thiamine repletion may all warrant repletion of magnesium based on clinical suspicion alone, regardless of otherwise normal serum magnesium levels.

Treatment of Magnesium Deficiency

Repletion of magnesium, as with potassium, must be done over time, although for different reasons. Once magnesium is infused, the shift from extracellular to intracellular fluid compartment is gradual, and renal excretion of perceived excess magnesium begins almost immediately after infusion. As such, a single dose of IV magnesium is likely to only have an effect on serum concentrations for about 30 minutes unless followed up by a steady-state administration over time while intracellular stores are repleted. Oral magnesium agents are useful in the outpatient environment to supplement diet and prevent deficits due to diuretic use, but are rarely adequate to replete deficits in the acute setting.

The IV solution of 50% magnesium sulfate (MgSO4) is the agent most commonly available. This solution contains 4 mEq/mL of elemental magnesium. The 50% MgSO4 solution is very hyperosmolar (4,000 mOsm/L), and should be diluted 5:1 to a 10% solution in normal saline before infusing.

Magnesium-depleted patients should be treated differently based on the severity of their depletion.34,35 For mild asymptomatic hypomagnesemics, assume a total magnesium deficit of 1–2 mEq/kg. Because approximately 50% of the repleted magnesium will be lost in urine before intracellular equilibration, replete twice the anticipated deficit. Administer 1 mEq/kg over the first 24 hours, and then 0.5 mEq/kg per day over the next 3–4 days. If enteral access exists, oral repletion may be appropriate.

For moderate hypomagnesemia (<1 mEq/L), add 6 g of MgSO4 to 250 mL of normal saline and infuse over 3 hours, and then 5 g of MgSO4 in 250 mL of normal saline over the next 6 hours, and finally 5 g of MgSO4 every 12 hours for the next 5 days.

For severe life-threatening hypomagnesemia, or repletion in the context of torsades de points or seizure activity, infuse 2 g of MgSO4 IV over 2–5 minutes. This dose may be repeated. Follow this with 5 g MgSO4 in 250 mL of normal saline over the next 6 hours, and then an additional 5 g of MgSO4 every 12 hours for 5 days.

Magnesium Excess

Elevated magnesium levels are rarely a problem outside the context of renal failure, as excess magnesium is excreted renally with great facility. Symptomatically, hyporeflexia is seen at serum levels around 4 mEq/L, first-degree AV block at levels around 5 mEq/L, complete heart block at levels around 10 mEq/L, and cardiac arrest at levels around 13 mEq/L.36 It would be rare to achieve clinically problematic elevated levels of magnesium in the context of renal failure without first encountering clinically significant hyperkalemia, unless an untoward level of magnesium intake were being undertaken. Hemodialysis is the treatment of choice for malignant hypermagnesemia. The administration of calcium salts, either calcium chloride or gluconate, can be used to temporize the conduction delays due to excess magnesium while hemodialysis is arranged.

Calcium is the most abundant electrolyte in the human body, but the vast majority of it (99%) is found in bone. The portion of calcium in the serum, which is measured on routine calcium concentration assays, exists partially as albumin or other protein-bound calcium, partially as chelated calcium, and partially as ionized calcium. Only the ionized calcium is metabolically active and of interest clinically. Unfortunately, routine lab values fail to distinguish between the different forms of serum calcium. Since variations in albumin concentration, as well as variations in the degree of calcium binding to albumin, directly affect the ionized calcium proportion, attempts at imputing the ionized calcium level from the total serum level of calcium are difficult at best. A number of calculations have been proposed to adjust the serum calcium concentration based on the serum albumin concentration; none of them reliably work in the acutely ill patient.37 The only way to meaningfully appreciate the concentration of active ionized calcium is to directly measure the ionized calcium with an ion-specific probe. Fortunately, direct serum measurement of ionized calcium levels can now be routinely performed in most labs in a timely fashion. Normal values for serum ionized calcium are between 1.1 and 1.3 mmol/L (4.5–5.0 mg/dL).

Hypocalcemia

Effects of Hypocalcemia

Hypocalcemia induces increased excitability of muscle tissue, leading to cardiac irritability and muscle twitching (evidenced by the oft-referenced but not sensitive Chvostek's and Trousseau's signs). Ionized hypocalcemia also impairs strength in muscle contractions, due to calcium's role in actin/myosin chain interactions. The result is progressive twitching leading eventually to tetany of skeletal muscles and hyperreflexia. Tetany of the laryngeal muscles may occur, creating an airway emergency. Cardiovascular effects include increased excitability and ectopy along with decreased myocardial function.

Causes of Hypocalcemia

Hypocalcemia is often seen in the acutely ill patient. While disorders of parathyroid function are the most common causes of hypocalcemia in the outpatient, they are rarely the culprit in the critically ill patient. Hypocalcemia in the critically ill patient is often multifactorial. See Table 22-6 for a list of potential causes of ionized hypocalcemia.

Medications that can induce hypocalcemia include fluoride poisoning, bisphosphonates, calcitonin, amphotericin B, cimetidine, ethanol, foscarnet, citrate, albumin, heparin, phenytoin, rifampin, aminoglycosides, loop diuretics, isoniazid, and propothiouracil.27

Hypomagnesemia can blunt calcitriol production as well as blunt the end-organ response to calcitriol. For this reason, repletion of magnesium in the case of low magnesium levels or refractory ionized hypocalcemia is recommended. Hyperphosphatemia can lead to chelation of ionized calcium in renal failure. Chelation of ionized calcium also occurs during massive blood transfusions, where large cumulative infused amounts of citrate (used as anticoagulant in stored blood products) can drop the ionized calcium levels. For this reason, monitoring and repletion of ionized calcium should be performed as necessary during massive transfusions, particularly in the context of refractory hypotension.38

Sepsis and the systemic inflammatory response syndrome (SIRS) are associated with hypocalcemia, probably due to decreased PTH and calcitriol production.39 No clinical benefit has been seen to ionized calcium repletion in sepsis if patients are asymptomatic, and it is not clear if sepsis-related ionized hypocalcemia is protective or deleterious.

Treatment of Hypocalcemia

Treatment of ionized hypocalcemia is 2-fold: (1) identify and treat the underlying disorder that gave rise to the ionized hypocalcemia; (2) replete symptomatic or severe (<0.8 mmol/L) ionized hypocalcemia emergently. Calcium only requires urgent repletion if symptomatic or if approaching critically low levels. Previous studies have suggested no advantage to repletion of asymptomatic ionized hypocalcemia above levels of 0.65 mmol/L,40 while one recent study indicated increased adverse outcomes only in patients with ionized calcium levels below 0.8 mmol/L.41 Calcium administration is not benign, particularly in the context of tissue hypoxia, where calcium administration can aggravate cellular injury.42 Rapid infusion of calcium solutions can cause bradycardia, hypotension, and vasodilatation. When indicated, IV repletion should be undertaken with either calcium gluconate or calcium chloride. Calcium gluconate and calcium chloride each comes in 10-mL vials, each containing 100 mg of its respective compound. Calcium chloride, however, has three times the elemental calcium of calcium gluconate (27 mg/mL [1.36 mEq/mL] vs. 9 mg/mL [0.46 mEq/mL], respectively). Calcium chloride is significantly more hyperosmolar than calcium gluconate (2,000 mOsm/L vs. 680 mOsm/L), and so should only be infused through a large central line. Both solutions should be diluted in normal saline or dextrose 5% in water prior to administration. Calcium will equilibrate between the extracellular and intracellular space following infusion, and the immediate results seen following infusion will wane within 30 minutes of administration unless a follow-up infusion is started.40 A total of 200 mg of elemental calcium (corresponding to approximately 8 mL of 10% calcium chloride, or 22 mL of 10% calcium gluconate) may be required to increase the ionized calcium 0.1 mmol/L.

Hypercalcemia

Hypercalcemia, defined as an increase in ionized calcium greater than 2.6 mmol/L, is rare in critically ill patients. The most common cause of hypercalcemia in the ED, as in outpatients, is primary hyperparathyroidism, which can be diagnosed with a PTH level. When present in critically ill patients, hypercalcemia is most often associated with underlying malignancy, although it may also be seen in other disorders that result in increased bone resorption such as sarcoidosis or with prolonged immobilization. A few medications can cause hypercalcemia, including thiazide diuretics, lithium, and vitamin D or A supplementation.27

Effects of Hypercalcemia

Mild hypercalcemia is often asymptomatic. Gastrointestinal symptoms include ileus, constipation, nausea, and vomiting due to smooth muscle relaxation. Patients often suffer from severe lethargy, dehydration due to polyuria, and stupor. Cardiac features include shortened QTc, broadened T waves, and first-degree AV block. There is very poor correlation between serum calcium levels and the severity of symptoms.43

Treatment of Hypercalcemia

IV fluid hydration and diuresis to promote renal excretion are indicated if the patient is symptomatic, or in the case of ionized calcium levels greater than 3.5 mmol/L.44 If underlying malignancy is suspected to exist, use of salmon calcitonin or bisphosphanates may be useful. Hydrocortisone may be useful in the case of multiple myeloma.45

Phosphorus (PO4) exists predominantly in bone, and as free ionic PO4 intracellularly. PO4 is critical in all ATP-related energy-requiring cellular activities such as glycolysis and high-energy phosphate bond formation. Normal serum levels of PO4 are 2.5–5.0 mg/dL, or 0.8–1.6 mmol/L.

Hypophosphatemia

Hypophosphatemia can be the result of an intracellular shift of PO4, an excess of PO4 secretion, or a deficiency in PO4 intake.

Effects of Hypophosphatemia

Hypophosphatemia is usually clinically silent unless very profound. The effects of hypophosphatemia are all related to cellular energy production, including a decrease in cardiac output observed in heart failure patients, a theoretical decrease in skeletal muscle strength, and reports of difficulty weaning from ventilators due to respiratory muscle weakness.46,47 Multiple reports of cardiac arrest, cardiovascular collapse, or profound encephalopathy associated with profound hypophosphatemia exist in the emergency medicine literature, many associated with intracellular shifts of PO4 during high-dose insulin therapy for DKA.48–50

Causes of Hypophosphatemia

The transit of glucose into cells is an active process in which PO4 is cotransported into the cell. As such, both the feeding of a patient who has been nutritionally deprived and the aggressive use of insulin for tight euglycemic control can cause dramatic intracellular shifts of glucose, and in turn PO4. Any time nutritional support is started on a patient after a period of deprivation, gradual increases in caloric intake along with frequent assessment of serum PO4 levels should be undertaken.

Alkalosis can cause intracellular shift of PO4, presumably due to increased glycolysis that accompanies increased intracellular pH. This is seen to a far greater degree in respiratory alkalosis than in metabolic alkalosis, and accounts for the increased incidence of hypophosphatemia seen in chronic obstructive pulmonary disease (COPD) patients.51

The use of β-receptor sympathomimetics is associated with a transient intracellular shift of PO4, although the clinical significance of this is not clear.52 The same response to sympathetic tone may account for the hypophosphatemia seen in sepsis or SIRS.

The use of antacid compounds, such as sucralfate or aluminum hydroxide, can bind phosphate in the upper gastrointestinal tract and impair its absorption.

Other medications that can cause hypophosphatemia include glucocorticoids, insulin, and, when in overdose, acetaminophen, aspirin, and theophylline.27

Osmotic dieresis, as a result of either hyperglycemia or administration of an osmotic diuretic, can impair renal reabsorption, and lead to urinary wasting of phosphate.

See Table 22-7 for a list of potential causes of hypophosphatemia.

Treatment of Hypophosphatemia

PO4 can be repleted in IV formulation with either sodium phosphate (93 mg/mL [3 mmol/mL] PO4, 4.0 mEq/mL sodium) or potassium phosphate (93 mg/mL [3 mmol/mL] PO4, 4.4 mEq/mL potassium). Patients presenting with a serum PO4 of less than 2.0 mg/dL should receive 15 mmol of sodium phosphate in 100 mL of sodium chloride, infused over 2 hours. If a concomitant hypokalemia exists, potassium phosphate can be substituted at an equal dose. If follow-up serum PO4 levels at 6 hours remain less than 2 mg/dL, identical repeat doses can be administered up to 45 mmol of PO4 in a 24-hour period.53 If no central venous access is available, the above doses should be diluted in a total of 250 mL of normal saline in order to prevent phlebitic complications of infusing a hyperosmolar solution. Oral PO4-containing solutions, such as K-Phos or Neutra-Phos, cannot be used effectively for the large doses of PO4 required to replete severe hypophosphatemia (<1.0 mg/dL), as they tend to promote diarrhea, but can be used to maintain PO4 levels once repletion via IV is complete, or to replete mild asymptomatic phosphate deficits.54 Repletion of mild to moderate hypophosphatemia in the ED may be accomplished with the administration of 2–3 g of sodium phosphate or potassium phosphate, most formulations of which contain 8 mmol of PO4 per packet or capsule. Daily PO4 intake requirements, absent excessive losses, are about 1,200 mg (38 mmol) per day orally or about 800 mg (25 mmol) per day IV for a 70-kg adult.

Hyperphosphatemia

Hyperphosphatemia occurs as the result of renal failure or the widespread necrosis of cells, as in ischemia/reperfusion, rhabdomyolysis, or tumor lysis. The principal concern is for the formation of calcium–PO4 insoluble complexes and subsequent profound hypocalcemia. Treatment is with IV hydration, and the administration of PO4-binding agents via the gastrointestinal tract, such as sucralfate, aluminum-containing antacids, or calcium acetate. Hemodialysis, while rarely necessary, may be employed to promote PO4 clearance in the renal failure patient.55