Clinical Syndromes of Metabolic Acidosis

Hepatorenal Interactions

As illustrated in Fig. 59.1 , the proton load produced by hepatic amino acid metabolism will titrate extracellular bicarbonate resulting in production of CO ₂ that is eliminated by the lungs. To maintain plasma bicarbonate constant, the kidney has two pivotal tasks. First, it must reclaim the bicarbonate present in the glomerular ultrafiltrate (filtered load of bicarbonate) and, second, it must regenerate bicarbonate to replete the bicarbonate consumed during endogenous acid production. The first task is accomplished by virtually complete tubular reabsorption or reclamation of the filtered bicarbonate along the nephron (predominantly in the proximal tubule). The kidney regenerates bicarbonate (bicarbonate neogenesis) by excretion of protons either as ammonium ( ${\mathrm{NH}}_4^{+}$ ) or as titratable acid (TA) in amounts sufficient to offset those protons produced daily by oxidation of dietary foodstuffs. The amount of renal proton excretion or, equivalently, bicarbonate neogenesis/regeneration is measured as net acid excretion (NAE), which reflects the sum of ${\mathrm{NH}}_4^{+}$ and titratable acid excretion minus any urinary bicarbonate excreted (that which escapes tubular reabsorption):

The rate of renal hydrogen ion secretion is thus expressed as the sum of excreted hydrogen ions (NAE) and filtered bicarbonate reabsorbed:

where

Excretion of ${\mathrm{NH}}_4^{+}$ requires the interaction of hepatic and renal glutamine metabolism and intact renal ammoniagenesis. Urea synthesis in peripheral hepatocytes consumes equal amounts of bicarbonate and ${\mathrm{NH}}_4^{+}$ and thus is without impact on acid–base balance. However, a small fraction of ${\mathrm{NH}}_4^{+}$ is diverted to glutamine synthesis in hepatocytes located in the center of the hepatic lobule and such hepatic production of glutamine has been argued to be an important homeostatic substrate for renal ${\mathrm{NH}}_4^{+}$ /net acid excretion. Acidemia stimulates and alkalemia inhibits glutamine synthesis. Under certain conditions, acidemia has been demonstrated to be associated with decreased urea synthesis thereby increasing the supply of ${\mathrm{NH}}_4^{+}$ for glutamine synthesis, raising the question that the liver might control the renal acid excretory response to metabolic acidosis. Nevertheless, a significant role for acute metabolic acidosis in altering urea production rate in vivo has not been found when tested robustly in rats. In addition, in humans, a negative correlation between ureagenesis and plasma ${\mathrm{HCO}}_3^{-}$ concentration was observed over a wide range of ${\mathrm{HCO}}_3^{-}$ concentrations that were altered both chronically and acutely ( Fig. 59.2 ). Thus, the ureageneic process per se may even be maladaptive for acid–base regulation in humans and does not appear to have a discernible homeostatic effect on acid–base equilibrium. It is likely that other metabolic pathways operate in metabolic acidosis to offset the accelerated ${\mathrm{HCO}}_3^{-}$ consumption of ureagenesis (e.g., oxidation of glutamate or other alkali generating reactions), but this issue has not been addressed experimentally either in animals or humans.

Glutamine supply to the kidneys is also increased by enhanced release of glutamine from skeletal muscle and increased intestinal glutamine absorption, thus providing higher levels of glutamine in the splanchnic circulation. In aggregate, acidemia increases glutamine synthesis or release and delivery of glutamine to the kidney.

In the proximal tubule, two deamidation/deamination steps, both positively regulated by low pH (glutamine deaminase, glutamate dehydrogenase) yield two ${\mathrm{NH}}_4^{+}$ and two ${\mathrm{HCO}}_3^{-}$ molecules. ${\mathrm{NH}}_4^{+}$ and ${\mathrm{HCO}}_3^{-}$ are then separated, ${\mathrm{NH}}_4^{+}$ is excreted in the urine and ${\mathrm{HCO}}_3^{-}$ is added to the extracellular fluid. Therefore, glutamine catabolism in the kidney leads to excretion of protons (as ${\mathrm{NH}}_4^{+}$ ) and reabsorption of equal amounts of ${\mathrm{HCO}}_3^{-}$ . Excess glutamine, not taken up by the kidney, is returned to the liver and deamidated/deaminated, a process that stimulates ureagenesis, inhibits further glutamine synthesis and thus constitutes a feed-back loop regulating glutamine (and thus potential ${\mathrm{NH}}_4^{+}$ and bicarbonate) supply to the kidney. In contrast to the kidney, ${\mathrm{HCO}}_3^{-}$ and ${\mathrm{NH}}_4^{+}$ are not physically partitioned or differentially excreted in the liver, but metabolized with a stoichiometry of 1:1 during ureagenesis. It is thus renal control of ${\mathrm{NH}}_4^{+}$ excretion, not hepatic control of glutamine vs. urea synthesis that relates changes in plasma acid–base composition to changes in ${\mathrm{NH}}_4^{+}$ metabolism and thus to possible effects on acid–base balance.

Role of the Gut

The gut affects acid–base homeostasis not only by absorbing sulfur-containing amino acids, but also by absorbing potential base. Organic anions in the gut are intrinsic to a whole food diet and are also derived in situ from dietary carbohydrates, protein and fat. They are absorbed (constituting potential base), or excreted in feces (around 10 to 60 mmol/day) or intestinally metabolized to organic acids by bacteria with subsequent reaction with ${\mathrm{HCO}}_3^{-}$ in ileum and colon. The amount of enterally absorbed base from a human whole food diet has been estimated to be about 50% of the magnitude of endogenous acid production.

The amount of base absorbed can be estimated as the difference between diet and stool anion gaps. In both diet and feces, alkali content is estimated as the sum of noncombustible cations (Na ⁺ +K ⁺ +Ca ²⁺ +Mg ²⁺ ) minus the sum of noncombustible anions (Cl ⁻ +1.8 P), where 1.8 represents the average valence of phosphate at pH 7.4.

The same arithmetic version of the urinary anion gap ([Na ⁺ +K ⁺ +Ca ²⁺ +Mg ²⁺ ]−[Cl ⁻ +1.8 P]) has also been shown to be correlated directly and positively with the diet-stool anion gap in metabolically controlled human subjects. This would allow an estimate of gastrointestinal alkali absorption without measuring stool composition. However, the urinary anion gap cannot reliably estimate gut base absorption under conditions in which net systemic organic acid production is altered or in which net bone accretion or dissolution prevails (i.e., typical clinically encountered conditions) as this construct of the urinary anion gap reflects the combined effects of cellular and bone buffering, endogenous organic acid production as well as gastrointestinal base absorption.

There is only weak evidence that intestinal absorption might adapt to acidosis. An insignificant decrement in the stool anion gap of 7 mEq/day was noted during the first six days of acid loading (potential secondary systemic gain of alkali) in human subjects, but this decrement subsequently decreased to only 2.5 mEq/day during prolonged acidosis. The measured 2.5 mEq/day increment in gut base absorption in metabolic acidosis is a very small response in comparison to the simultaneous increment in renal net acid excretion (from 44 to 247 mEq/day). Thus, quantitative considerations and the fact that the estimation of the stool anion gap rests completely on an unknown valence for phosphate, and that only 5% of the fecal phosphate is dialyzable (suggesting that most of stool phosphate is either intracellular or sequestered in calcium/magnesium salts) are the most important arguments against an adaptive role of the gut in acidosis.

Role of Bone

Bone is a large reservoir of exchangeable base as calcium carbonate and dibasic phosphate. Exogenous and endogenous metabolic acid loads, both acute and chronic, liberate calcium and base (carbonate) from bone. Calcium is subsequently lost in the urine.

There is evidence from human studies (reviewed in Lemann et al. ) that when acid production is increased experimentally, renal net acid excretion does not increase as much as acid production. The difference in acid neutralization (i.e., the positive acid balance) is attributed to bone buffering of retained protons (or liberation of bone base) ( Fig. 59.3 ). Similarly, renal net acid excretion does not decrease as much as acid production when base (KHCO ₃ ) is administered. Under these experimental conditions, subjects are likely in small negative acid or positive base balance, with incorporation of base into bone. Positive acid balance is accompanied by equivalent cation loss (loss of Ca ⁺⁺ and possibly K and/or Na from bone), negative acid balance by equivalently positive cation balance (incorporation of Ca ⁺⁺ and K and/or Na). Scanning ion microprobe analysis indicates that bone buffer content (i.e., ${\mathrm{HCO}}_3^{-}$ and phosphate) is decreased significantly in mouse calvariae in in vitro models of both acute and chronic metabolic acidosis. This consumption of proton buffers may attenuate the severity of both acute and chronic metabolic acidosis, but does so at the expense of a decrease in bone mineral content.

Based on these and other observations that acidotic normal subjects and patients with renal failure that were shown to be in positive acid balance and yet were able to maintain stable plasma bicarbonate concentrations, it is assumed that bone contributes in a quantitative important way to proton buffering in response to acid loads. However, in the chronic acidosis of renal failure, quantitative considerations of the required large magnitude of bone alkali have raised the possibility that available bone alkali content is insufficient to offset a possible underexcretion of acid relative to that needed for daily acid balance. While it has been argued that increased gut organic anion absorption may account for this needed base, such a conclusion is not a compelling one when based on a computation of acid balance that depends on a constant and assumed phosphate valence for whole food diets and stool (see Role of Gut section).

Renal Regulation of Acid–Base Equilibrium

As discussed, the kidney maintains a stable [ ${\mathrm{HCO}}_3^{-}$ ]p through two processes: (1) ${\mathrm{HCO}}_3^{-}$ reclamation, the reabsorption of filtered ${\mathrm{HCO}}_3^{-}$ , and (2) ${\mathrm{HCO}}_3^{-}$ regeneration, the neogenesis of ${\mathrm{HCO}}_3^{-}$ that has been decomposed by the invasion of fixed acids into the extracellular fluid or ${\mathrm{HCO}}_3^{-}$ lost from the body through extrarenal or renal routes.

The relationship between the [ ${\mathrm{HCO}}_3^{-}$ ]p and renal ${\mathrm{HCO}}_3^{-}$ reclamation is shown in Fig. 59.4 . When the [ ${\mathrm{HCO}}_3^{-}$ ]p is reduced below the normal range, about 25 mEq/liter, the kidney reclaims all filtered ${\mathrm{HCO}}_3^{-}$ . As the [ ${\mathrm{HCO}}_3^{-}$ ]p increases toward normal, complete ${\mathrm{HCO}}_3^{-}$ reclamation continues until a critical ${\mathrm{HCO}}_3^{-}$ concentration is reached. This concentration is about 25 mmol/liter in normal humans’ “bicarbonate threshold,” and thus explains that normal subjects excrete only trivial quantities of bicarbonate in their urine (usually less than 5 mmol/day). Above the bicarbonate threshold, some filtered bicarbonate escapes reclamation and is excreted in the urine. As the [ ${\mathrm{HCO}}_3^{-}$ ]p increases further, ${\mathrm{HCO}}_3^{-}$ reclamation also increases, but not in proportion to the increment in the filtered ${\mathrm{HCO}}_3^{-}$ load. Therefore, despite greater reclamation, increasing quantities of ${\mathrm{HCO}}_3^{-}$ enter the urine. Finally, a [ ${\mathrm{HCO}}_3^{-}$ ]p is reached beyond which further increases elicit no greater ${\mathrm{HCO}}_3^{-}$ reclamation. At this [ ${\mathrm{HCO}}_3^{-}$ ]p (about 28 mEq/liter in normal humans), an apparent maximal reabsorptive rate, or apparent T _max for ${\mathrm{HCO}}_3^{-}$ , has been achieved. When the [ ${\mathrm{HCO}}_3^{-}$ ]p increases above the apparent T _max , the resultant increment in the filtered ${\mathrm{HCO}}_3^{-}$ load is entirely excreted into the urine.

In addition to ${\mathrm{HCO}}_3^{-}$ reclamation (reabsorption), the kidney must also regenerate the ${\mathrm{HCO}}_3^{-}$ , that has been decomposed by the entry of fixed acids into the extracellular fluid (ECF) or lost in urine or stool. The kidney regenerates ${\mathrm{HCO}}_3^{-}$ by excreting acid. The vast majority of protons are excreted in the form of ${\mathrm{NH}}_4^{+}$ or titratable acid as the free proton concentration at the minimal urinary pH of 4.5 is less than 0.1 mmol/liter. The net effect of renal acidification can be measured as net acid excretion.

Net Acid excretion= ${\mathrm{NH}}_4^{+}$ +Titratable Acid− ${\mathrm{HCO}}_3^{-}$

Quantitatively, changes in renal ${\mathrm{NH}}_4^{+}$ excretion account for most of the changes in renal acid excretion both in response to changes in endogenous acid production and in the adaptation to acid–base disturbances. In adaptation to changes in acid load, ${\mathrm{NH}}_4^{+}$ excretion is regulated at multiple levels including glutamine supply to the kidney, production of NH ₃ in the proximal tubule (ammoniagenesis), secretion of NH ₃ into the tubular lumen and various transport processes of NH ₃ / ${\mathrm{NH}}_4^{+}$ along the nephron. (For a discussion of these mechanisms, see Chapter 54, Chapter 56 )

Phosphate and organic anions such as citrate, acetate, beta-hydroxybutyrate and creatinine contribute as filtered buffers to titratable acid excretion. Amongst these substances, the buffer pair di- and mono-basic phosphate ( ${\mathrm{HPO}}_4^{-}{/H}_2{\mathrm{PO}}_4^{-}$ ) ₂ is the most important buffer due to its pKa of 6.8 and urinary excretion values on the order of 20 to 50 mmol/day. Titration of the other potentially titratable species is severely limited due to their pKa values residing below 5.0. The importance of phosphate as a urinary proton buffer is illustrated by the effects of both neutral phosphate loading and depletion. Increases in neutral phosphate load significantly enhance net acid excretion and are able to generate as well as maintain metabolic alkalosis even without a concomitant increase in renal cation resorption or sodium avidity. Although phosphate delivery to distal acidification sites has been demonstrated to be rate limiting for renal acid secretion in acidotic dogs, under phosphate loading conditions, the generation of phosphate-induced secondary hyperparathyroidism appears to provide the major stimulus to enhanced renal acid excretion. Phosphate depletion reduces both phosphate and titratable acid excretion and induces positive hydrogen balance. It is unclear whether these changes are best explained by a rate-limiting role of luminal phosphate for renal acid excretion or by the observed renal ${\mathrm{HCO}}_3^{-}$ wasting.

Determination of Net Acid Excretion

Although enzymatic ammonium assays in serum are widely performed in clinical medicine and this assay method is valid in urine, ammonium values are seldom requested in urine for reasons that have not been reported. As an alternative to requesting an ammonium value in urine, a simplified version of the urinary anion gap is frequently used as an indirect index of ammonium excretion:

The rationale is that in chronic metabolic acidosis, ammonium excretion is elevated, is a cation, and balances part of the negative charge of the major urinary cation, chloride. Thus, the urinary anion gap should become progressively negative as the rate of ammonium excretion increases in response to acidosis or to acid loads. However, the urinary anion gap is not a quantitative index of ammonium excretion. Its potential role in the clinical differential diagnosis of hyperchloremic metabolic acidosis will be discussed in subsequent sections.

Calculation of titratable acid requires determination of urinary pH (UpH), blood pH (BbpH), urine phosphate, estimation of the pKa (near 6.8) and an arithmetic computation:

Although use of a pKa phosphate value of 6.8 is sufficient for most clinical purposes, the pK’ of phosphate can be corrected for ionic strength and pH for improved accuracy.

Ventilatory Response to Acidosis: Homeostatic and/or Maladaptive?

The most characteristic clinical manifestation of metabolic acidosis is hyperventilation. Because the depth of ventilation increases to a much greater degree than the respiratory rate, hyperventilation may not be clinically apparent until acidemia becomes severe. Then, patients may experience labored breathing and their chief complaint is dyspnea.

Metabolic acidosis rapidly causes secondary hyperventilation, which reduces the PaCO ₂ and tends to increase the blood pH toward normal. This secondary respiratory response is usually fully developed in 12–24 hours. Peripheral chemoreceptors in the carotid and aortic bodies and central chemoreceptors on the ventral surface of the medulla trigger the respiratory response to metabolic acidosis. Some controversy still exists as to which chemoreceptors are more important.

During acute metabolic acidosis (initial six hours after an acid load) PaCO ₂ falls by 0.85 mm Hg for every millimole-per-liter decrease in [ ${\mathrm{HCO}}_3^{-}$ ]p ( Fig. 59.5 ). In response to sustained acidosis, the respiratory response is further stimulated and is maximal by 24 hours. Then, PaCO ₂ is decreased by about 1.1 mm Hg for each millimole-per-liter fall in [ ${\mathrm{HCO}}_3^{-}$ ]p in acidotic human subjects. Several other helpful guides can be used at the bed side to determine whether respiratory compensation is appropriate: (1) metabolic acidosis should reduce the PaCO ₂ so that it approximates the decimal digits of the pH. For example, when metabolic acidosis reduces the pH to 7.30, the pCO ₂ should be about 30 mm Hg. Maximal hyperventilation in patients with extreme acidemia will reduce the PaCO ₂ to about 10 mm Hg. (2) Another useful rule is that the PaCO ₂ should approximate the [ ${\mathrm{HCO}}_3^{-}$ ]p+15. 9. (3) In uncomplicated metabolic acidosis the decrease in PaCO ₂ after full adaptation may also be calculated as follows:

The respiratory response to metabolic acidosis may be slightly different in patients with different forms of metabolic acidosis. For example, it has been suggested that lactic acidosis generates greater hyperventilation and a lower PaCO ₂ than equally severe ketoacidosis. This could be the result of a lower intracellular pH in the patients with lactic acidosis. However, others find that secondary hyperventilation in lactic acidosis is not different from other metabolic acidoses.

A chronically low PaCO ₂ (hypocapnia) lowers the [ ${\mathrm{HCO}}_3^{-}$ ]p as a result of transient bicarbonaturia or decreased renal net acid excretion. In dogs, experimental hypocapnia caused the kidney to decrease [ ${\mathrm{HCO}}_3^{-}$ ]p (transient renal acid retention) even in the context of metabolic acidosis with preexisting hypobicarbonatemia. The decrease in [ ${\mathrm{HCO}}_3^{-}$ ] was strictly proportional to the induced hypocapnia over a wide range of PaCO ₂ values and [ ${\mathrm{HCO}}_3^{-}$ ]p (fall in [ ${\mathrm{HCO}}_3^{-}$ ]p of 0.54 mmol/liter per 1-mm Hg fall in PaCO ₂ ). The hypocapnia-induced decrement in [ ${\mathrm{HCO}}_3^{-}$ ]p was sufficient to increase blood acidity. These findings demonstrated that the renal response to metabolic acidosis is not primarily geared at the defense of pH and that secondary hypocapnia elicits a maladaptive renal response. In contrast to dogs, however, humans exhibit an alkalemic response to hypocapnia (increase in blood pH) ( Fig. 59.6 ) irrespective of the presence or absence of preexisting metabolic acidosis ([ ${\mathrm{HCO}}_3^{-}$ ]p between 25 and 10 mmol/liter), despite the finding that hypocapnia also caused transient renal acid retention. The reason for this species difference is that the inhibitory effect of hypocapnia on renal acid excretion (decrease of [ ${\mathrm{HCO}}_3^{-}$ ]p by 0.41 mmol/liter per 1-mm Hg fall in PaCO ₂ ) is too small in humans to offset the direct alkalemic effect of the decrease in PaCO ₂ as dictated by the ${\mathrm{HCO}}_3^{-}$ Henderson-Hasselbalch equation. However, if the slope of line for the relation of the [ ${\mathrm{HCO}}_3^{-}$ ]p on PaCO ₂ extends to [ ${\mathrm{HCO}}_3^{-}$ ]p below 10 mmol/liter, it can be predicted that at the lowest end of the range of values for [ ${\mathrm{HCO}}_3^{-}$ ]p, hypocapnia might worsen acidemia in humans as well as dogs. Figure 59.7 illustrates these differences in the prediction of blood [H ⁺ ] when chronic hypocapnia is superimposed on chronic metabolic acidosis when a [ ${\mathrm{HCO}}_3^{-}$ ]/PaCO ₂ slope of 0.54 is used (as reported for dogs; dashed lines) or when a [ ${\mathrm{HCO}}_3^{-}$ ]/PaCO ₂ slope of 0.41 is used as reported for normal human subjects.

Metabolic and Endocrine Consequences of Metabolic Acidosis

Effects of Metabolic Acidosis on Sodium Homeostasis

It is well known that both acute and chronic metabolic acidosis are associated with natriuresis. Both proximal and distal tubular sodium reabsorption appear to be inhibited. The proximal effect is the result of downregulated organic-anion stimulated NaCl absorption, while ${\mathrm{HCO}}_3^{-}$ reabsorption is enhanced as a consequence of increased expression and activity of Na ⁺ /H ⁺ exchange (sodium–hydrogen exchanger isoform 3, NHE-3). The natriuresis of acidosis induces extracellular volume depletion and increased renin-angiotensin-aldosterone activity, which significantly counteracts and limits the acidosis-induced negative sodium balance and weight loss. Inhibition of angiotensin II action by the AT-1 receptor antagonist losartan resulted in decreased renal net acid excretion in pre-existing metabolic acidosis. Therefore, distal nephron angiotensin II activity is a coregulator of the renal-tubular response to metabolic acidosis and, thereby, is important in determining the severity of acidosis. Increased aldosterone is also expected to limit the severity of metabolic acidosis by stimulation of distal nephron acidification. However, inhibition of aldosterone action by spironolactone exacerbated human metabolic acidosis by an extrarenal mechanism (increased endogenous acid production precluding an analysis of its effect on renal acidification. Thus, the quantitative role of acidosis-induced hyperaldosteronism in the tubular response to metabolic acidosis and the resultant severity of metabolic acidosis remain to be determined.

Effects of Metabolic Acidosis on Potassium Homeostasis

Both extrarenal cellular potassium shifts and the renal regulation of potassium transport have been examined in metabolic acidosis. Contrary to a widely expressed opinion, there is no compelling evidence—either in acute mineral or acute organic acidosis—for a demonstrable increase in plasma potassium in response to acidosis. Under carefully controlled conditions, arterialized venous plasma potassium increased nonsignificantly by only +0.02 mmol/liter per mmol/liter decrease in plasma [ ${\mathrm{HCO}}_3^{-}$ ] in acute, NH ₄ Cl-induced metabolic acidosis in normal human subjects. Previous studies in dogs and humans (cited in Wiederseiner et al. ) had described both unchanged and increased plasma potassium concentrations albeit with unusually large variances. Failure to detect hyperkalemia in the most recent and some of the previous reports does not preclude, however, the existence of acidosis-induced net K ⁺ /H ⁺ exchange across cell membranes with efflux of K ⁺ from cells. The finding of a hyperinsulinemic response (with concomitant decrease in glucagon concentration and unchanged catecholamine levels) in response to acidosis suggests that an insulin response counterregulates any acidemia-induced cellular potassium efflux, resulting in stable plasma potassium concentrations. This occurred despite an increase in the insulin/glucose ratio, suggesting the insulin resistance in acute acidosis may not involve transcellular potassium balance. A similar mechanism may operate in acute organic acidosis. In the case of organic acidosis, organic anions may enter cells along with protons thus obviating exit of potassium. In conclusion, acidosis-induced hyperinsulinemia seems to prevent acidosis-induced hyperkalemia. The plasma potassium response in acutely acidotic subjects with insulin deficiency or preexisting insulin resistance needs further investigation. In the clinical setting, therefore, the plasma potassium response must always be interpreted with a diligent analysis of the presence of factors in addition to acidemia that are known to influence potassium distribution between intra- and extracellular spaces such as changes in osmolality, sympathoadrenergic activity, aldosterone, and the activity of glucoregulatory hormones (insulin and glucagon).

Acute metabolic acidosis (less than two hours) initially reduces renal potassium excretion, based in large part on effects in the CCT. Experimental reduction of luminal pH in the isolated perfused CCT results in diminished K secretion. The effect of acute acidification of CCT principal cells is believed to be mediated by the effect of cellular acidity to decrease the open probability of potassium channels in the apical membrane. Metabolic acidosis of longer duration is a potent stimulus of renal K excretion and may lead chronically to significant hypokalemia and substantial potassium depletion. This increase in potassium excretion is probably the consequence of increased volume/sodium delivery to the distal tubule and collecting duct and hyperaldosteronism.

Effect of Metabolic Acidosis on Divalent Ion, PTH, and 1,25 (OH) ₂ D Metabolism

Metabolic acidosis profoundly affects calcium and phosphate metabolism resulting in calcium loss from bone in association with hypercalciuria. Hypercalciuria is the result of an increase in filtered load and decreased tubular reabsorption of calcium, the cellular mechanisms of which are poorly understood although calcium reabsorption is correlated with luminal [ ${\mathrm{HCO}}_3^{-}$ ] in the distal tubule. Important clinical sequelae of the resultant negative calcium balance are a metabolic bone disease with poorly characterized features of osteomalacia/low formation rate/high resorption rate and calcium nephrolithiasis. Metabolic acidosis induces hypophosphatemia in association with increased renal phosphate clearance and increased fractional excretion of phosphate; in brief, metabolic acidosis induces renal phosphate depletion. The mechanism of decreased tubular phosphate reabsorption in the proximal tubule is likely to be complex. High ambient proton concentration was shown to stimulate phosphate transport by a glucocorticoid-dependent, posttranslational mechanism. Isohydric decreases in [ ${\mathrm{HCO}}_3^{-}$ ]/PaCO ₂ , on the other hand, stimulated phosphate transport in OK-cells via a transcriptional effect. In addition, metabolic acidosis has been shown to decrease [IGF-1]s in humans, providing a plausible mechanism for decreased renal phosphate reabsorption. Thus, metabolic acidosis seems to affect renal regulation of phosphate reabsorption both directly (via effects of acid–base changes on phosphate transport) and indirectly via endocrine changes (increased glucocorticoid activity, decreased IGF-1 levels). The relative importance of these mechanisms remains to be elucidated.

Acute and chronic metabolic acidosis was demonstrated to induce renal magnesium wasting and hypomagnesemia. The cellular mechanisms of acidosis-induced renal magnesium wasting are largely unknown, although extracellular acidity has been shown to decrease magnesium uptake in mouse distal tubular cells.

In animals, metabolic acidosis was found to decrease [1,25(OH) ₂ D]s, an effect generally attributed to decreased activity of renal 1-alpha-hydroxylase. However, chronic metabolic acidosis was demonstrated repeatedly to increase [1,25(OH) ₂ D]s (by stimulation of its production rate) and to concomitantly decrease [PTH]s in humans. The effects of metabolic acidosis on ionized calcium concentration (hypercalcemia not observed or very mild in humans, but prevalent in rats) and on the severity of phosphate depletion/hypophosphatemia seem to differ among species. Thus, it is likely that the changes in [1,25(OH) ₂ D]s and [PTH]s observed are primarily determined by the occurrence or the severity of acidosis-induced hypercalcemia, which has been shown to override other potent stimuli of 1,25(OH) ₂ D production including phosphate depletion unless the latter is quite severe.

It is interesting to speculate that the elevated [1,25(OH) ₂ D]s in response to metabolic acidosis could serve a homeostatic role, or, that elevated [1,25(OH) ₂ D]s could contribute to the normal acid excretory response to an acid load/acidosis. This question merits investigation in as much as vitamin D deficiency was shown to result in metabolic acidosis in chicks and chronic 1,25(OH) ₂ D administration results in metabolic alkalosis (in part of renal origin) in thyroparathyroidectomized dogs.

In summary, metabolic acidosis in humans induces (1) hypercalciuria due to release of calcium from bone and decreased renal tubular calcium reabsorption, (2) renal phosphate depletion and hypophosphatemia, (3) renal magnesium wasting, and (4) increases [1,25(OH)2D]s and decreases in intact [PTH]s.

Effect of Metabolic Acidosis on Bone

The effects of metabolic acidosis on bone mineral content—that is, reduced levels of mineral sodium, potassium, carbonate, calcium, and phosphate—are consistent with the role of bone as a proton buffer. In mouse calvariae in vitro , exposure to acid media (low ${\mathrm{HCO}}_3^{-}$ ) induces initial physicochemical mineral dissolution. Cell-mediated bone resorption follows and is characterized by both increased osteoclastic and inhibited osteoblastic activities paralleled by alterations in the expression of a number of osteoclastic and osteoblastic genes. In particular, increased osteoblastic RNA expression of the osteoclast activator RANK-ligand with unchanged expression of its soluble decoy receptor, osteoprotegerin, may provide a central mechanism by which metabolic acidosis enhances osteoclastogenesis and osteoclast activity. The effect is mediated—at least in part—by acidosis-induced enhancement of osteoblastic prostaglandin E2 (PGE2) secretion. Both RANK-L expression and cell-mediated calcium efflux from bone are inhibited by cyclooxygenase inhibition, confirming the central role of acidosis-induced PGE2 stimulation.

Since glucocorticoid activity is increased in metabolic acidosis and glucocorticoids are known to dramatically decrease bone mineral content, it is possible that acidosis-induced hyperglucocorticoidism might contribute to calcium efflux from bone. Surprisingly, however, in vitro cortisol inhibited rather than stimulated acid-induced, cell-mediated osteoclastic bone resorption through a decrease in osteoblastic PGE2 production.

Other Endocrine Effects of Metabolic Acidosis

Effects of Metabolic Acidosis on Renal Citrate Metabolism

The effects of metabolic acidosis on renal citrate metabolism have important clinical consequences and are, therefore, briefly discussed here. Citrate is derived from carbohydrate metabolism and contains three negatively charged carboxyl groups. Complete oxidation of citrate thus generates three ${\mathrm{HCO}}_3^{-}$ ions per mole and thus increased renal reabsorption and metabolism of citrate would be expected to serve a homeostatic role in the adaptation to metabolic acidosis. Citrate is freely filtered at the glomerulus and reabsorbed and metabolized almost exclusively in the proximal tubules ( Fig. 59.8 ), with these processes determining the amount of urinary citrate excretion. In metabolic acidosis, tubular reabsorption is increased and urinary excretion decreased due to both increased protonation of trivalent to divalent citrate, the substrate for the sodium/citrate cotransporter in the proximal tubule and increased expression and activity of Na ⁺ /citrate cotransporter NaDC1. In addition, cellular (i.e., cytoplasmic) metabolism of citrate ( Fig. 59.8 ) is increased in the proximal tubule. Enhanced cytoplasmic metabolism in response to metabolic acidosis is mediated by a specific increase in renal ATP citrate lyase activity and expression. Mitochondrial citrate metabolism is also stimulated in metabolic acidosis as shown by the increased activity of mitochondrial aconitase, the first step in mitochondrial citrate metabolism. These findings are clinically important because of citrate’s role in calcium complexation, inhibition of stone formation and prevention of nephrocalcinosis. Indeed, a high proportion of patients with nephrolithiasis have low urinary citrate levels. A high-protein intake (which increases endogenous acid production) decreases urinary citrate excretion in rats by renal mechanisms described previously.

Miscellaneous Clinical Effects of Metabolic Acidosis