Disorders of Acid-Base and Potassium Balance Part 3

Treatment

Treatment of metabolic acidosis is aimed at correcting both acidemia and the underlying disorder. The likelihood that alkali administration is needed and that it will be effective depends on the blood pH, the compensatory mechanisms, and the underlying condition.

Until the arterial blood pH falls below 7.15 to 7.20, the adverse effects of acidemia are usually compensated for by elevated plasma catecholamines [see Diagnosis, above]. To maintain adequate buffer reserves, alkali therapy should be considered to keep [HCO3-] higher than 10 to 12 mEq/L. Alkali administration is usually unnecessary, however, if the acidosis is likely to resolve spontaneously (as in lactic acidosis after a grand mal seizure). Therapy must take into consideration the underlying cause of metabolic acidosis (see below).

Chronic renal failure Of interest is the finding that the acidemia tends to be more severe in nondiabetic patients with chronic renal failure than in diabetic patients with similar degrees of renal insufficiency.15 This difference may be the result of more efficient extrarenal bicarbonate generation in diabetic patients.

As long as the metabolic acidosis is mild, many adults with renal failure are not treated with alkali replacement, partly because of a concern that sodium bicarbonate will exacerbate the volume expansion and hypertension that are commonly present.

Some studies, however, have suggested several reasons for the use of alkali replacement therapy in patients with renal fail-ure.16-18 Among the reasons are the likelihood that acidemia can enhance the breakdown of skeletal muscle, reduce albumin synthesis, and (by activating the complement system) contribute to tubulointerstitial injury and that bone buffering of hydrogen ions can lead to bone resorption.19 These findings have led some physicians to advocate the early use of alkali therapy to maintain plasma [HCO3-] above 22 mEq/L. Definitive studies are still needed, however, to ascertain the benefit of such treatment.


Physicians should be aware that electrolyte studies in he-modialysis patients who use non-hospital-based dialysis units may be performed in central laboratories several hundred miles away. Compared with results of samples obtained locally, samples tested at the central laboratories appear to show consistent and clinically important decreases in serum [HCO3-].20 This in vitro enhancement of metabolic acidosis should be confirmed before treatment is initiated.

Lactic acidosis Correcting the underlying disorder is the primary therapy for lactic acidosis. Reversal of circulatory failure, hypoxemia, or sepsis reduces the rate of lactate production and enhances its removal.

As in chronic renal failure, the use of alkali therapy in lactic acidosis is controversial.21,22 The principal rationale for bicarbonate administration is the potential maintenance of normal cardiovascular homeostasis. This possible advantage must be weighed against deleterious side effects, such as volume overload, hypernatremia, and alkalosis (when excessive bicarbonate is administered).

Clinical studies also suggest that sodium bicarbonate therapy may not improve either the blood pH or the survival of patients with lactic acidosis. The administration of sodium bicarbonate has been observed to reduce cardiac performance in patients with cardiac disease, congestive heart failure, or acute myocar-dial infarction.23 This lack of efficacy possibly results from an associated increase in net lactic acid production, although hyperos-molality of the administered alkali solution also may be important. The finding that dichloroacetate (which promotes the conversion of pyruvate into the Krebs cycle instead of permitting its conversion to lactate) can lower lactate levels and raise the blood pH in patients with lactic acidosis without seeming to improve survival lends further support to the concept that treating the underlying cause of lactic acidosis is more important than treating the acidemia. In view of these observations, a reasonable approach may be to administer bicarbonate to maintain the arterial blood pH at higher than 7.15 and the plasma bicarbonate at higher than 10 mEq/L. If one of the complications of bicarbonate therapy ensues, the benefit of continuing alkali therapy should be reevaluated.

Therapy for D-lactic acidosis must address the two underlying factors that contribute to the overproduction of D-lactic acid: the presence of intestinal bacteria that produce D-lactic acid and the enhanced delivery of carbohydrates to the colon secondary to short bowel syndrome. Treatment usually consists of antibiotic therapy to decrease the number of D-lactate-producing organisms. Alkali therapy has been used but has not been evaluated in a controlled fashion.

Ketoacidosis The trend toward normal-anion-gap metabolic acidosis during recovery from ketoacidosis has an important effect on the rate at which acidemia can be corrected. A maximal rate of 500 ml of fluid an hour appears to be effective replacement therapy for patients with ketoacidosis. Significantly higher rates of fluid administration can result in a decreased anion gap without an increase in the plasma [HCO3-]. Once this occurs, correction of the acidosis requires regeneration of bicarbonate by the kidney, a process that may take at least 2 to 3 days24; consequently, more vigorous fluid administration (above 500 ml/hr) may delay recovery from acidosis. The rate of fluid administration should be slowed after intravascular volume compromise— manifested by reduced blood pressure or increased plasma creat-inine and urea nitrogen concentrations—has been corrected.

Although exceptions occur, the administration of sodium bicarbonate is not usually necessary in ketoacidosis, because there appears to be no difference in mortality between patients treated with sodium bicarbonate and control subjects. Bicarbonate therapy may be a risk factor for the subsequent development of cerebral edema in this setting.25

Salicylate, ethylene glycol, and methanol intoxication Sal-icylate removal is enhanced by urinary alkalinization (to maintain the blood pH between 7.45 and 7.5), which increases urinary excretion even in the setting of dialysis.26 The usual treatment of ethylene glycol or methanol intoxication is to administer an agent that will reduce conversion of the nontoxic alcohol to its toxic metabolic by-products and to use dialysis in the presence of tissue damage or acidemia. Although ethyl alcohol has been administered historically, fomepizole (4-methylpyrazole) is the only potent inhibitor of alcohol dehydrogenase that has been studied prospectively and that has been approved by the Food and Drug Administration for this condition.

Hyperchloremic acidoses The therapeutic approach to the patient with diarrhea and metabolic acidosis depends upon the severity of the two disturbances. Alkali administration is not necessary with mild to moderate reductions in the plasma [HCO3-] if the diarrhea is controlled (thereby minimizing further bicarbonate loss) and renal function is normal (thereby allowing acid excretion to increase). However, some patients will require treatment with bicarbonate.

The acidemia in type 1 RTA can be corrected with bicarbonate or with a bicarbonate precursor, such as citrate. The usual requirement is 1 to 3 mEq/kg/day. Correcting the acidemia reduces tubular citrate reabsorption, which leads to an increase in urinary excretion and a decrease in the tendency toward nephrolithiasis and nephrocalcinosis. Usually, a potassium salt, such as potassium citrate, is administered because it corrects the potassium deficit as well.

In adults, treatment of type 2 RTA is aimed at the underlying cause (e.g., multiple myeloma). Because the acidemia is typically mild, alkali therapy may not be necessary. Correction of acidemia is appropriate in children because in this age group, acidemia is more likely to impair growth and contribute to metabolic bone disease.

Type 4 RTA, if mild, may not require treatment. Aldosterone replacement with fludrocortisone (0.1 to 1 mg/day) may increase acid secretion by lowering the serum [K+]; however, many patients cannot tolerate the side effects (e.g., edema and hypertension) associated with this therapy. In these individuals, correction of serum [K+] may be achieved by administration of a loop diuretic, dietary potassium restriction, or the elimination of drugs that promote hyperkalemia. When fludrocortisone therapy is required, concomitant administration of a loop diuretic may limit the development of high blood pressure and fluid retention.

Metabolic alkalosis

Primary metabolic alkalosis is characterized by an elevated plasma [HCO3-] and an arterial pH greater than 7.42. When there is concomitant metabolic acidosis, however, the blood pH may be increased, decreased, or normal. Furthermore, hyper-bicarbonatemia alone is not diagnostic of primary metabolic al-kalosis, because it may also represent the appropriate physiologic response to chronic respiratory acidosis. These conditions can usually be easily distinguished by measuring the arterial blood pH, which is reduced in respiratory acidosis.

Causes of Metabolic Alkalosis

Metabolic alkalosis is a relatively common clinical problem that is most often induced by diuretic therapy or the loss of gastric secretions as a result of vomiting or nasogastric suction.28 Two abnormalities must be present for metabolic alkalosis to develop and to be sustained. First, there must be an initial increase in the plasma [HCO3-] caused by hydrogen loss in gastrointestinal secretions or in the urine, hydrogen movement into cells, alkali administration, or volume contraction around a relatively constant amount of extracelluar bicarbonate (called a contraction alkalosis) [see Table 5]. Second, one of three factors (in the absence of advanced renal failure) must be present to sustain the high plasma [HCO3-] once the initiating event has been terminated: effective circulating volume depletion, chloride depletion, and hypochloremia or hypokalemia [see Additional Factors Contributing to Metabolic Alkalosis, below].

Gastrointestinal hydrogen loss Gastric juice contains a high concentration of hydrochloric acid and a lesser amount of potassium chloride. Each 1 mEq of hydrogen lost generates 1 mEq of bicarbonate. In normal persons, gastric hydrogen secretion does not lead to metabolic alkalosis, because it is matched by pancreatic bicarbonate secretion that is stimulated as the acid enters the duodenum. There is no stimulus to bicarbonate secretion, however, when vomiting or nasogastric tube drainage prevents the hydrogen ions from reaching the duodenum. Vomiting can be surreptitious in some cases, such as in patients with eating disorders. In some patients, laxative abuse can also lead to a metabolic alkalosis, most likely resulting from a hypokalemia-induced increase in urinary ammonia production, which in turn leads to increased ammonium chloride excretion in the urine.

Table 5 Causes of Metabolic Alkalosis

Disorder

Cause

Hydrogen loss

Gastrointestinal losses

Removal of gastric secretions (vomiting or nasogastric suction)*

Chloride-losing diarrheal states

Laxative abuse

Renal losses

Loop or thiazide diuretics*

Mineralocorticoid excess

Posthypercapnic alkalosis

Hypercalcemia (including milk-alkali

High-dose intravenous penicillin derivatives

Bartter syndrome

Intracellular hydrogen shift

Hypokalemia

Bicarbonate retention

Administration of alkali (as either a bicarbonate or a bicarbonate precursor with massive blood transfusions or with absorbable antacids)+

Contraction alkalosis

Diuretics*

Loss of gastrointestinal secretions having high [Cl-] and low [HCO3-] when compared with plasma (usually with vomiting)*

*These are the most common causes.

+For alkalosis to be maintained, renal bicarbonate excretion also must be impaired by either reduced filtration or enhanced reabsorption by the proximal tubule.

Renal hydrogen loss An inappropriate renal acid loss may occur when there is an increase in hydrogen ion secretion by the distal nephrons. Mineralocorticoids, including aldosterone, act here both by directly stimulating the secretory H+-ATPase pump and by making the tubular lumen more electronegative through the stimulation of sodium reabsorption [see Figure 6]. Distal potassium secretion is also enhanced in this setting and results in concurrent hypokalemia.

Metabolic alkalosis associated with loss of renal hydrogen may be caused by mineralocorticoid excess, administration of loop or thiazide diuretics, posthypercapnic alkalosis, or hypercalcemia (e.g., milk-alkali syndrome) [see Table 5]. Metabolic alkalosis also occurs in Bartter and Gitelman syndromes. These two disorders produce electrolyte abnormalities similar to those caused by diuretic therapy because they are associated with defects in the transporters in the loop of Henle and the distal tubule, respectively, that inhibit mechanisms also inhibited by loop and thiazide diuretics [see Bartter Syndrome and Gitelman Syndrome, below]. Any cause of mineralocorticoid excess, such as primary aldos-teronism, can result in metabolic alkalosis. Primary aldostero-nism is generally accompanied by hypertension and hy-pokalemia. In contrast, untreated patients with secondary aldos-teronism caused by congestive heart failure or cirrhosis usually do not present with metabolic alkalosis or hypokalemia. In such cases, the effect of aldosterone is counteracted by decreased distal sodium delivery (unless diuretics are administered) and reduced urine volume; these factors limit the quantity of acid and potassium secreted into, as well as excreted in, the final urine.

 In metabolic alkalosis, the high ratio of a-intercalated to ^-intercalated cells decreases. Unlike a-intercalated cells, which regenerate bicarbonate and add it to the venous blood, ^-intercalated cells promote urinary HCO3- excretion by exchanging HCO3- for Cl-in the glomerular filtrate. As a result, chloride administration serves an important function in the treatment of most persons with metabolic alkalosis.

Figure 6 In metabolic alkalosis, the high ratio of a-intercalated to ^-intercalated cells decreases. Unlike a-intercalated cells, which regenerate bicarbonate and add it to the venous blood, ^-intercalated cells promote urinary HCO3- excretion by exchanging HCO3- for Cl-in the glomerular filtrate. As a result, chloride administration serves an important function in the treatment of most persons with metabolic alkalosis.

When patients are treated with either thiazide or loop diuretics, adequate distal delivery of sodium chloride and increased secretion of aldosterone occur. The increase in distal hydrogen ion secretion and volume contraction, if the diuresis has been large, contribute to the development of metabolic alkalosis.

Chronic respiratory acidosis brings about an appropriate increase in hydrogen secretion, as the rise in the plasma [HCO3-] raises the pH toward normal. Rapid lowering of the PaCO2, usually by mechanical ventilation, leads to metabolic alkalosis, as the patient is left with an elevated plasma [HCO3-]. This abnormality is called a posthypercapnic metabolic alkalosis.

Hypercalcemia Hypercalcemia increases renal tubular bicarbonate reabsorption. Significant metabolic alkalosis in hypercal-cemic patients, however, is more commonly seen in those with milk-alkali syndrome. In milk-alkali syndrome, an increased alkaline load (caused by the ingestion of calcium carbonate) and hypercalcemia-induced renal failure increase bicarbonate production and diminish bicarbonate excretion. Metabolic alkalosis can be observed in any severe hypercalcemic state that is caused by volume contraction and by impaired sodium chloride and potassium chloride absorption resulting from tubular injury.

Intracellular hydrogen shift In addition to being caused by hydrogen loss, metabolic alkalosis can be caused by a shift of hydrogen into the cells. Both vomiting and diuretic therapy directly induce potassium and hydrogen loss. Hypokalemia produces a transcellular shift in which potassium leaves cells to replete extracellular stores; to maintain electroneutrality, hydrogen enters the cells. This shift not only raises the extracellular pH but also decreases the intracellular pH; the latter promotes proximal tubular bicarbonate reabsorption and distal hydrogen ion secretion.

Alkali administration The administration of sodium bicarbonate at a dosage as high as 1,000 mEq/day does not normally induce metabolic alkalosis in normal persons, because the excess bicarbonate is rapidly excreted in the urine. If the ability to excrete bicarbonate is impaired, however, metabolic alkalosis can occur when a very large quantity of bicarbonate or a bicarbonate precursor (e.g., lactate, citrate, or acetate) is administered, as occurs with citrate in large-volume blood transfusions.

Contraction alkalosis Contraction alkalosis develops when loss of relatively large volumes of bicarbonate-free fluid occurs. In this setting, the plasma [HCO3-] rises because the extracellular volume contracts around a relatively constant quantity of extracellular bicarbonate.

The most common cause of contraction alkalosis is administration of a loop diuretic to induce rapid fluid removal in a patient with marked edema. Similarly, contraction alkalosis occurs under other conditions in which fluid with a high chloride concentration and a low [HCO3-] is lost. Among these causes are receipt of thiazide diuretics; loss of gastric secretions (even in patients with achlorhydria); sweat losses in patients with cystic fi-brosis; and diarrhea in some patients with villous adenomas or congenital chloridorrhea.

Additional factors contributing to metabolic alkalosis In the absence of advanced renal failure, one of three factors must be present to sustain high plasma [HCO3-]: effective circulating volume depletion, chloride depletion, or hypokalemia.

Both the fall in the glomerular filtration rate (GFR) and the associated sodium avidity seen with hypovolemia limit the excretion of sodium bicarbonate. Most bicarbonate is reabsorbed in the proximal tubule [see Figure 2]. An important stimulus for enhanced reabsorption in this nephron segment is increased activity of the Na+-H+ antiporter in the tubular cell membranes. Volume contraction promotes Na+-H+ exchange in this nephron segment, in part through the release of angiotensin II. The hydrogen ions secreted into the lumen combine with filtered bicarbonate, ultimately leading to a higher rate of transport back into the tubular cells. More bicarbonate is returned to the venous blood at the level of the collecting tubules, partly under the influence of secondary aldosteronism. Volume depletion does not cause metabolic alkalosis without selective chloride losses. For example, blood loss does not increase the [HCO3-], because the chloride losses are identical to those in an equal volume of plasma.

Chloride depletion can both promote bicarbonate regeneration and decrease distal bicarbonate secretion. Bicarbonate generation in the a-intercalated cells in the cortical collecting tubule is mediated by hydrogen ion secretion via H+-ATPase pumps in the luminal membrane [see Figure 6]. Passive cosecretion of chloride is required to maintain electroneutrality. Intracellular bicarbonate is returned to the systemic circulation in exchange for chloride.

The -intercalated cells in the cortical collecting tubule (which increase in number when metabolic alkalosis develops) are able to secrete bicarbonate directly by reversing the location of the transporters [see Figure 6]. Thus, the Cl-HCO3- exchangers are located in the luminal membrane, leading to bicarbonate secretion into the tubular lumen; the H+-ATPase pumps are located in the basolateral membrane. Although the activity of these cells is appropriately enhanced by alkalemia in an attempt to excrete the excess bicarbonate, the associated fall in the tubular fluid [Cl-] diminishes the favorable inward gradient for chloride, thereby reducing bicarbonate secretion. However, chloride depletion in humans is almost always associated with effective volume depletion.

Hypokalemia directly increases bicarbonate reabsorption by at least two different mechanisms. First, the fall in plasma [K+] shifts potassium from hydrogen into cells. The ensuing intracel-lular acidosis stimulates hydrogen secretion and bicarbonate re-absorption in the proximal and collecting tubules. Second, distal hydrogen and potassium secretion is mediated in exchange for luminal sodium [see Figure 6]. In states of potassium depletion, the rate of hydrogen secretion in exchange for sodium increases. As a result, hypokalemia and hyperaldosteronism, which stimulate hydrogen ion secretion, can have a potentiating effect on the development and maintenance of metabolic alkalosis.

Diagnosis

Clinical manifestations Some individuals with metabolic alkalosis have severe cramping, paresthesias, or even tetany, but others with similar electrolyte levels do not; the reason for this difference in presentation is unclear. Other clinical findings are the result of the underlying etiology (e.g., hypertension with primary hyperaldosteronism).

The diagnosis of metabolic alkalosis is usually evident from the patient’s history of vomiting or receipt of diuretic therapy. In some cases, however, no cause for the metabolic alkalosis is apparent. In such a setting, the most likely diagnosis is surreptitious vomiting caused by an eating disorder, use of diuretics, or one of the causes of mineralocorticoid excess (e.g., primary al-dosteronism). The first two factors induce effective volume depletion, whereas primary hyperaldosteronism is usually associated with mild volume expansion as a result of the stimulatory effect of aldosterone on renal sodium reabsorption.

Several findings on physical examination may suggest surreptitious vomiting as the cause, including dental erosion from repeated exposure to acid gastric secretions and the presence of ulcers and calluses on the dorsum of the hand caused by sticking a finger in the back of the throat to induce vomiting.

Laboratory tests Measurement of the [Na+] in a random urine specimen is used in many conditions to distinguish between volume depletion ( urinary [Na+] usually < 20 mEq/L) and euvolemia ( urinary [Na+] > 40 mEq/L). However, metabolic al-kalosis is one of the conditions in which volume depletion may not lead to a low urinary [Na+]. The capacity to retain sodium in this setting may be antagonized by the need to excrete bicarbonate (as the sodium salt) in an attempt to correct the alkalosis. In such cases, a random urinary [Cl-] determination is more useful.

Sodium wasting is most likely to occur during the first few days of vomiting, when the plasma [HCO3-] and thus the filtered bicarbonate load are increased. Early in the course of vomiting, the ability to enhance bicarbonate reabsorption has not yet occurred. The net effects are a high urinary [Na+] and urinary [K+] and a urine pH of greater than 7.0 caused by bicarbonaturia.

As a result, the urinary [Na+] is not necessarily an accurate reflection of the patient’s volume status in metabolic alkalosis. The presence of underlying hypovolemia can be detected more accurately by the finding of a urinary [Cl-] below 25 mEq/L. The appropriate chloride conservation is caused both by volume depletion and by hypochloremia induced by chloride losses in gastric secretions. The urinary chloride concentration may be inappropriately elevated, however, if a defect in chloride reabsorption is present. Such a defect most commonly occurs in patients who receive diuretic therapy. Thus, in patients with metabolic alkalosis caused by vomiting, the urinary chloride concentration is typically low; it is higher when diuretics are administered. In both circumstances, the urinary [Na+] may be elevated.

Bartter syndrome and Gitelman syndrome Bartter and Gitelman syndromes are disorders of sodium chloride reabsorp-tion in the loop of Henle and the distal tubule, respectively.29 Bartter syndrome is a rare disorder that causes hypokalemic metabolic alkalosis. Because the urinary [Na+] and urinary [Cl-] are usually higher than 25 mEq/L, surreptitious diuretic use is the major disorder to be considered in the differential diagnosis. Classic Bartter syndrome generally presents in early life and may be associated with delayed growth and mental retardation. The spectrum of findings, including hypercalciuria, is most compatible with a primary defect in sodium chloride reabsorption in the medullary thick ascending limb of the loop of Henle.

Gitelman syndrome is a more benign condition that may be inherited as an autosomal recessive disease; it may not be diagnosed until late childhood or even adulthood. In contrast to patients with Bartter syndrome, who have a defect in urinary concentrating ability (normal function of the loop of Henle is needed to generate a high interstitial osmotic gradient), patients with Gitelman syndrome can demonstrate normal urinary concentrating ability and have hypocalciuria. This finding suggests that the defect resides in the distal tubule, because similar findings are observed in patients treated with thiazide diuretics. Bartter and Gitelman syndromes are diagnosed only after diuretic use—the other, much more common, cause of these findings— has been excluded.

Treatment

In patients with true volume depletion caused by vomiting, nasogastric suction, villous adenomas, or diuretic therapy, metabolic alkalosis can be corrected by the administration of sodium chloride.30 The administration of potassium chloride to patients with concurrent hypokalemia also contributes to correction of the alkalemia. Measuring the urinary [Cl-] may be useful in determining when volume depletion has been corrected; once renal perfusion has been restored, this value should be above 40 mEq/L.

Edematous states Therapy is different for edematous patients with metabolic alkalosis usually caused by heart failure, cor pulmonale, or advanced liver disease. In these disorders, sodium chloride is contraindicated because it increases the degree of edema. Administration of the carbonic anhydrase inhibitor acetazolamide (beginning with 250 mg p.o., q.d. or b.i.d., or 125 mg I.V., q.d. or b.i.d.), however, may be particularly effective. This drug preferentially inhibits proximal tubular reabsorp-tion of sodium bicarbonate, thereby correcting both the alkalosis and the fluid overload.

A potential side effect of therapy with a carbonic anhydrase inhibitor is the development or worsening of hypokalemia. Although hypokalemia can be treated with potassium supplements, an alternative approach to therapy is to administer a potassium-sparing diuretic (e.g., spironolactone) instead of a carbonic anhydrase inhibitor. Potassium-sparing diuretics impair sodium reabsorption in the collecting tubules and, as a result, limit further potassium and hydrogen secretion [see Figure 6]. In patients with advanced liver disease, the aldosterone antagonist spironolactone or eplerenone may be the most effective diuretic.

On extremely rare occasions, metabolic alkalosis may be so severe that administration of hydrochloric acid is required to correct the problem. The standard 0.1N (decinormal) hydrochloric acid solution contains 100 mEq H+/L. Because of the corrosive nature of hydrochloric acid solution, administration of hydrochloric acid should be used only when other measures to correct metabolic alkalosis have failed. The hydrochloric acid solution must always be administered into a central vein.

Bartter syndrome and Gitelman syndrome The tubular defect in patients with Bartter syndrome or Gitelman syndrome cannot be corrected. As a result, treatment is directed at minimizing the electrolyte and metabolic abnormalities. The combination of an NSAID, including cyclooxygenase-2 inhibitors (because prostaglandin levels are secondarily increased), and a potassium-sparing diuretic can raise the plasma [K+] toward normal and largely reverse the metabolic alkalosis. Most patients, however, require continued oral potassium and magnesium supplements because drug therapy is rarely completely effective.

Next post:

Previous post: