Brenner and Rector's The Kidney, 8th ed.

CHAPTER 24. Interpretation of Electrolyte and Acid-Base Parameters in Blood and Urine

K.S. Kamel   M.R. Davids   S.H. Lin   M.L. Halperin

  

 

Water and Sodium, 757

  

 

Polyuria, 757

  

 

Defense of the Extracellular Fluid Volume, 764

  

 

Potassium and Metabolic Alkalosis, 765

  

 

Dyskalemias, 765

  

 

Metabolic Alkalosis, 772

  

 

Metabolic Acidosis, 774

  

 

Conclusion, 781

An analysis of laboratory data in blood and urine is essential to make accurate diagnoses and to design optimal therapy for patients with disturbances of water, sodium (Na+), potassium (K+), and acid-base homeostasis. [1] [2] Our clinical approach and interpretation of these tests rely heavily on an understanding of basic concepts in renal physiology. Hence we begin each section with concepts that help to identify the most important factor(s) that indicate how the kidneys regulate the excretion of these substances and then discuss the tools that utilize the laboratory data to help make a correct diagnosis; consults are presented to illustrate the utility of these tools. At the end of each section, all of the information is integrated to design a clinical approach.

Two principles will hold true throughout this chapter. First, there are no normal values for the urinary excretion of water and electrolytes because normal subjects in steady state excrete all ions that are consumed and not lost by non-renal routes. The urine also contains the major nitrogenous metabolic waste, urea—its rate of excretion depends largely on protein intake. Second, data should be interpreted with respect to the prevailing stimulus and the “expected” renal response. In this regard, urine collections done over short periods of time are more valuable than 24-hour urine collections because they more closely reflect the renal response to the prevailing stimulus at that time.

WATER AND SODIUM

In this section, we illustrate how to use information about the composition and volume of the urine in the differential diagnosis and management of disorders causing polyuria, an abnormal intracellular fluid (ICF) volume, and/or an abnormal extracellular fluid (ECF) volume.

Polyuria

There are three reasons why polyuria may be present; a water diuresis, an osmotic diuresis, and/or a renal medullary concentrating defect.

Water Diuresis

Concept SW-1

To move water across a membrane, there must be a channel that allows water to cross that lipid membrane and a driving force (a difference in concentration of “effective” osmoles).

Water Channels

Vasopressin is released when the concentration of Na+ in plasma (PNa) is >136 mmol/L. This hormone causes the insertion of aquaporin-2 water channels (AQP2) into the luminal membrane of the late distal nephron; AQP2 permit water to be reabsorbed when there is an osmotic driving force.[3] Even in the absence of vasopressin, there is a small degree of water permeability in the inner medullary collecting duct (MCD) (basal water permeability).[4]

Driving Force

Water will be drawn from a compartment with a lower to one with a higher “effective” osmolality; the magnitude of the force is enormous (∼19 mm Hg per mOsm/kg H2O per osmol/L difference). In the renal cortex, fluid with an osmolality of ∼100 mOsm/L enters the late distal convoluted tubule. When AQP2 are present in their luminal membranes, water is reabsorbed because the osmotic pressure difference is ∼200 mOsm/L (interstitial osmolality equals the plasma osmolality (Posm), which is ∼300 mOsm/L for easy math). Hence the osmotic driving force is ∼3800 mm Hg (19 mm Hg × 200 mOsm/L). In the renal inner medulla, the “effective” interstitial osmolality rises ∼twofold (from 300 mOsm/L to 600 mOsm/L). Because this driving force is even larger, water will be absorbed rapidly until osmotic equilibrium is achieved.

Tools: Water Diuresis

Urine Flow Rate

When AQP2 are not present in the luminal membrane (absence of vasopressin actions), the urine volume will be equal to the volume of filtrate delivered to the late distal nephron ( Fig. 24-1 ). In subjects consuming a typical western diet, the distal flow rate is high and the peak urine flow rate is 10 ml/min to 15 ml/min (∼14 L/day to 21 L/day). If the urine volume is considerably <14 L/day, seek a reason for a low distal delivery of filtrate.

000926

000519

FIGURE 24-1  Major concept in the control of the excretion of water. The barrel-shaped structure represents the late distal nephron. The major concept is shown to the left of the vertical dashed line. When vasopressin acts, water channels (AQP2, shown by the circle) are inserted in the luminal membrane of collecting duct cells, making these nephron segments permeable to water. The driving force to reabsorb water is a high “effective” osmolality in the medullary interstitial compartment. In a water diuresis (top right portion), AQP2 are absent; hence the urine flow rate is regulated by the distal volume delivery. In contrast, during an osmotic diuresis (lower right portion), AQP2 are present so the urine flow rate is dependent on the number of “effective” osmoles delivered to this nephron segment and the osmolality in the medullary interstitial compartment.

000519

 

Urine Osmolality (Uosm)

The Uosm is equal to the number of excreted osmoles divided by the urine volume. When the osmole excretion rate is 800 mosmol/day and vasopressin is absent, the Uosm will be 50 mOsm/L when the 24-hour urine volume is 16 L/day. A change in the Uosm in this setting reflects a change in the osmole excretion rate and/or the volume of filtrate delivered to the late distal nephron (which determines the urine volume), rather than a change in the concentrating ability of the kidney. In the above example, the Uosm will be 100 mOsm/L when the 24-hour urine volume is 8 L and the rate of excretion of osmoles has not changed. In both of these settings, there is still an enormous difference in osmolality between the luminal and interstitial fluid compartments due the absence of AQP2.

Osmole Excretion Rate

This is simply the product of the Uosm and the urine flow rate. In subjects eating a typical western diet, this excretion rate is 600 mosmol/day to 900 mosmol/day. Electrolytes and urea each account for half of these osmoles. In absence of AQP2, a change in the rate of excretion of osmoles does not cause a change in urine volume; rather, there is a change in the Uosm. Nevertheless, the osmole excretion rate should be calculated in a patient with a water diuresis because it will influence the urine flow rate if vasopressin is given and acts.

Consult SW-1: Does This Patient Have Polyuria?

A 22-year-old female lives in a southern climate. She is concerned about her body image and runs several miles per day. To avoid “dehydration”, she drinks large volumes of water, despite the absence of thirst. She consumes a low-salt diet because this helps her maintain her desired weight. She sought medical advice because she voided frequently, passing large volumes of urine each time. On the past two visits, her laboratory results were very similar; her PNawas 130 mmol/L, the 24-hour urine volume was 5 L, and the Uosm was 80 mOsm/kg H2O.

Questions

Does this patient have polyuria?

What risks might you anticipate with respect to her PNa?

Discussion

Does This Patient Have Polyuria?

There are two definitions of polyuria:

Conventional definition: Polyuria is present when the urine volume is >2.5 or 3 L/day.

Physiology-based definition: Polyuria is present when the urine volume is “higher than expected” in a specific setting.

Using the conventional definition, polyuria is present in this patient. The polyuria is due to a water diuresis because the Uosm is low ( Fig. 24-2 ). Because hyponatremia is present, polyuria is due to primary polydipsia.

000857

000519

FIGURE 24-2  Approach to the patient with polyuria due to a water diuresis. To enter this flow chart, the patient must have a large urine volume with an osmolality that is distinctly lower than the Posm. A bullet symbol denotes the final diagnostic categories.

000519

 

Using the physiology-based definition, the “expected” urine flow rate in a normal adult that lacks vasopressin actions (PNa is 130 mmol/L) should be at least 10 ml/min or ∼14 L/day.[5] Hence a urine volume of 5 L/day in this setting is a low urine volume; in fact, she has a diminished ability to excrete water. Because of a low “effective” ECF volume due to ongoing losses of Na+ and Cl- in sweat, there would be an increased reabsorption of Na+ upstream in the nephron, lowering the distal delivery of filtrate. The combination of this low distal delivery and the presence of basal water permeability in the medullary collecting duct, even in the absence of detectable levels of vasopressin in plasma,[4] leads to a diminished ability to excrete water, what we call “trickle-down” hyponatremia.[6] In support of presence of a low distal delivery of filtrate, her osmole excretion rate is low (80 mOsm/L × 5 L/day = 400 mosmol/day Vs the usual 600-900 mosmol/day) which suggests that there is a low rate of excretion of NaCl.

What Risks Might You Anticipate with Respect to Her PNa?

Because this is a chronic condition, she is in balance and water intake must equal water loss. One route for water loss is her urine (5 L/day); she also has a large loss of water in sweat (volume is not known, perhaps 2 L/day to 3 L/day). Hence she has a daily intake and loss of 7 L to 8 L of water.

With such a large throughput of water, she could easily develop a large positive balance of water. This will cause acute hyponatremia, which leads to acute brain cell swelling. There are three possible causes for this large positive balance of water: first, she may drink >8 L of water on a given day (her water intake is not driven by thirst); second, she may lose less water in sweat (e.g., she did not run that day); third, water excretion may decrease suddenly due to a non-osmotic stimulus for the release of vasopressin (e.g., pain, nausea, anxiety, drugs such as “Ecstasy”[7]).

A different risk is osmotic demyelination, which may develop if she had a large negative balance of water and thereby, too rapid a rise in her PNa, especially if she had a much lower PNa for several days. Moreover, patients with a poor dietary intake are at greater risk of developing osmotic demyelination.[8] Because her urine flow rate is determined by the rate of delivery of filtrate to the distal nephron, this delivery may increase if she had a high salt intake (e.g., she ate pizza with anchovies) or was given an infusion of isotonic saline (see Fig. 24-1 ).

Consult SW-2: What Is “Partial” About Partial Central Diabetes Insipidus?

A 32-year-old healthy male had a recent basal skull fracture. His urine output is ∼4-L/day—this is a consistent finding. The first morning PNa is 143-mmol/L, his Uosm is 200 mOsm/kg H2O in the 24-hour urine collection, and vasopressin levels in plasma were undetectable. When he was given dDAVP, his urine flow rate decreased to 0.5 ml/min and the Uosm rose to 900 mOsm/kg H2O. Two other facts, however, deserve further analysis. First, he was thirsty in the morning when he woke up. Second, his sleep was not interrupted by a need to urinate. In fact his Uosm was ∼425 mOsm/kg H2O in several overnight urines. In random urine daytime collections, his Uosm was 900 mOsm/kg H2O; his PNa at those times was 137 mmol/L. Moreover, his urine flow rate fell to 0.5 ml/min and his Uosm rose to 900 mOsm/kg H2O after an infusion of hypertonic saline.

Questions

Is this a water diuresis?

What are the best options for therapy?

Discussion

Is This A Water Diuresis?

Uosm and urine flow rate: Because his Uosm was 200 mOsm/kg H2O and the urine volume was 4 L/day, this was a water diuresis with a normal osmole excretion rate (800 mosmol/day) (see Fig. 24-2 ).

Response to dDAVP: He had an adequate renal response to dDAVP because his Uosm rose to 900 mOsm/kg H2O when this hormone was given—this ruled out nephrogenic diabetes insipidus (DI). Hence he has central DI. Because his urine volume was 4 L/day and not 10 L/day to 15 L/day, the diagnosis was “partial central DI”.

Interpretation

Although the diagnosis of central DI was straightforward, there were two facts that have not yet been interpreted. First, because he was thirsty, his “osmostat” and thirst center as well as the fibers connecting them appear to be functionally intact ( Fig. 24-3 ). Similarly, because he could excrete concentrated urine (his overnight Uosm was 425 mOsm/kg H2O) when his PNa was 143 mmol/L, his vasopressin release center was also functioning, but only when there was this “stronger stimulus” for the release of this hormone. Therefore a possible site for his lesion was destruction of some but not all of the fibers linking his “osmostat” to his vasopressin release center ( Fig. 24-4 ).[9] This could also explain why polyuria was not present overnight. (He stopped his oral water intake several hours prior to going to sleep.) This challenged the diagnosis of partial central DI, or at least our concept of what that diagnosis really implies.

000528

000519

FIGURE 24-3  Lesion causing a water diuresis. The three circles represent areas in the hypothalamus; the top one labeled “stat” is the sensor (“osmostat”), which detects changes in cell volume in response to a change in the PNa. These cells are linked to the thirst center (lower left) and to the vasopressin (AVP) release center (lower right). Non-osmotic stimuli also influence the release of vasopressin. Vasopressin causes the insertion of water channels in the late distal nephron (lower barrel-shaped structure). The diseases that can lead to an abnormally high urine output are shown to the right.

000519

 

000500

000519

FIGURE 24-4  Lesion in the CNS causing partial central DI. The three circles represent areas in the hypothalamus, the top one labeled “stat” is the sensor (“osmostat”), the circle on the lower left is the thirst center, and the circle on the lower right is the vasopressin (AVP) release center. The X symbol represents the hypothetical lesion that leads to severing of some but not all of the fibers connecting the “osmostat” to the vasopressin release center.

000519

 

Primary Polydipsia

On first evaluation, because his PNa was high enough to stimulate the release of vasopressin (143 mmol/L), primary polydipsia was not present at this time. In contrast, his Uosm was consistently ∼90 mOsm/kg H2O and his PNa was 137 mmol/L during the daytime; this suggests that primary polydipsia was present while he was awake. Its basis probably reflects a “learned behavior” to avoid the very uncomfortable feeling of thirst. This interpretation provides a rationale to understand the natural history of, and importantly the options for management for his partial central DI.

What Are The Best Options for Therapy?

The major point here is that a higher PNa could stimulate the release of vasopressin. There are two ways to raise the PNa: an input of NaCl or a water deficit. The patient selected oral NaCl tablets to raise his PNa to control his daytime polyuria because of its rapid and reproducible onset. Moreover, this therapy avoids the risk of acute hyponatremia, which may occur if he was given dDAVP and drank an excessive quantity of water. In contrast, he selected water deprivation to raise his PNa overnight to permit him to have undisturbed sleep; he was able to tolerate the thirst that developed.

Osmotic Diuresis with The “Expected” Medullary Osmolality

Concept SW-2

When AQP2 water channels are inserted into the luminal membrane of the late distal nephron, the Uosm is equal to the medullary interstitial osmolality.

The volume of an osmotic diuresis is directly proportional to the rate of excretion of osmoles and inversely proportional to the medullary interstitial osmolality. Because AQP2 are not present in late distal nephron in the absence of vasopressin, the rate of excretion of osmoles does not influence the urine volume in this setting. Therefore there cannot be an osmotic diuresis and a water diuresis in the same patient at the same time (see Fig. 24-1 ).

Concept SW-3

The medullary interstitial osmolality falls during an osmotic diuresis because there is less reabsorption of Na+ and Cl- in the medullary thick limb of the loop of Henle (mTAL), largely due to the fact that fluid entering the mTAL has a lower Na+ concentration.[10]

The “expected” medullary interstitial osmolality is ∼600 mOsm/kg H2O at somewhat high osmole excretion rates and values are closer to the Posm at much higher osmole excretion rates.[11]

Tools: Osmotic Diuresis

When evaluating the basis for a large urine volume in a patient with an osmotic diuresis, measure the concentrations of all major solutes in the urine ( Fig. 24-5 ).

000496

000519

FIGURE 24-5  Approach to the patient with polyuria due to an osmotic diuresis. To enter this flow chart, the patient must have a large urine volume with an osmolality that is distinctly higher than the Posm. A bullet symbol denotes the final diagnostic categories.

000519

 

Osmole excretion rate: This rate should be much >1000 mosmol/day or 0.5 mosmol/min in an adult during an osmotic diuresis.

Uosm: The “expected” Uosm should be more than the Posm; the absolute value depends on the osmole excretion rate and the medullary interstitial osmolality.

Nature of the urine osmoles: This should be determined by measuring the rate of excretion of the individual osmoles in the urine. As a quick test, however, deduce which solute may be responsible for the osmotic diuresis by measuring the concentrations of likely compounds in plasma (e.g., glucose and urea) and determine if mannitol or a lavage fluid solute was infused. Nevertheless, patients rarely are given a sufficiently large amount of mannitol to be the sole cause of a sustained osmotic diuresis. A saline- induced osmotic diuresis may occur if there was a large infusion of saline or in a patient with cerebral or nephrogenic salt wasting. To diagnose a state of salt wasting, there must be an appreciable excretion of Na+ at a time when the “effective” arterial blood volume is definitely contracted.

Source of the urine osmoles: In a patient with a glucose or urea-induced osmotic diuresis, it is important to decide whether these osmoles were derived from catabolism of endogenous proteins.

Consult SW-3: An Unusually Large Osmotic Diuresis in A Diabetic Patient

A 50-kg, 14-year-old female has a long history of poorly controlled type 1 diabetes mellitus because she does not take insulin regularly. In the past 48 hours, she was thirsty, drank large volumes of fruit juice, and her urine volume was very high. On physical examination, there was no evidence to imply that her ECF volume was appreciably contracted. The urine flow rate was 10 ml/min over this 100-minute period. Other lab data include: pH 7.33, PHCO3 24 mmol/L, PAnion gap-16-mEq/L, PK 4.8 mmol/L, PCreatinine 1.0 mg/dL (88 mmol/L) (close to her usual values), BUN 22 mg/dL (PUrea 8 mmol/L), and hematocrit 0.45. Of note, there was no decrease in her glucose concentration in plasma (PGlucose) despite the excretion of a large amount of glucose ( Table 24-1 ).


TABLE 24-1   -- Data for Consult SW-3

 

 

Admission

100 min

 

 

Plasma

Urine

Plasma

Urine

Glucose

mg/dL (mmol/L)

1260 (70)

5400 (300)

1260 (70)

5400 (300)

Na+

mmol/L

125

50

123

50

Osmolality

mOsm/L

320

450

316

450

 

 

 

Questions

What is the basis of the polyuria?

What dangers do you anticipate for this patient?

Discussion

What Is the Basis of the Polyuria?

Osmole excretion rate: The product of her Uosm (450 mOsm/L) and urine flow rate (10 ml/min) yields an osmole excretion rate of 4.5 mosmol/min, a value that is ninefold higher than the usual value in an adult (0.5 mosmol/min) (see Fig. 24-5 ).

Uosm: The Uosm of 450 mOsm/kg H2O indicates that this polyuria is due to an osmotic diuresis. In addition, this Uosm is lower than “expected”, probably reflecting the very high osmole excretion rate, which caused a larger fall in her medullary interstitial osmolality.

Urine flow rate: The urine flow rate was extremely high for two reasons: the very high osmole excretion rate and the lower than “expected” osmolality in her medullary interstitial compartment.

Nature of the urine osmoles: Because her GFR is modestly low and her PGlucose was extremely high (1260 mg/dL, 70 mmol/L), her filtered load of glucose will be markedly higher than the maximum tubular capacity for its reabsorption; hence, this is likely to be a glucose-induced osmotic diuresis (confirmed by a UGlucose that was 350 mmol/L).

Source of the urine osmoles: Of special emphasis, her PGlucose did not decline despite such a high rate of excretion of glucose. To put it into quantitative terms, the total content of glucose in her ECF compartment is 12.6 g (1260 mg/dL × 10 × 10 L ECFV/1000) while she excreted 54 g of glucose (5400 mg/dL × 10 × 1 L/1000). Therefore, she excreted an amount of glucose that is fourfold more than all the glucose she had in her ECF compartment. Accordingly, to maintain this degree of hyperglycemia, she needed an enormous input of glucose over a short period of time. The only likely source of such a large input of glucose was the glucose that was retained in her stomach. As a reference, 1 L of apple juice contains ∼135 g of glucose. For ingested glucose to fuel an osmotic diuresis, this patient would need a rapid rate of gastric emptying. Although the usual effect of hyperglycemia is to slow gastric emptying, for some reason this did not occur in this patient.[12]

What Dangers Do You Anticipate for This Patient?

Low ECF volume: Because glucose is an “effective” osmole in the ECF compartment, it helped to maintain her ECF volume. If she had discontinued her ingestion of glucose-containing beverages long before arriving in hospital, her ECF volume would now be obviously contracted because she would have excreted a large proportion of the glucose in her ECF compartment (e.g., at a rate faster than glucose entry from the GI tract).

Cerebral edema: Brain cell swelling may occur if there is a significant fall in her “effective” Posm (2 PNa + PGlucose in mmol/L terms).[13] This could occur if she excretes urine with a higher “effective” osmolality than the Posm. This risk would be even greater if she had changed her intake to water rather than sugar-containing beverages.

Concept SW-4

Not all osmoles are equal in their ability to increase the urine volume.

The osmoles that cannot achieve equal concentrations in the lumen of the MCD and in the medullary interstitial compartment are called “effective” osmoles; they dictate what the urine flow rate will be when the MCD is permeable to water. Urea, on the other hand, may be an “ineffective” urine osmole in some circumstances and an “effective” urine osmole in other circumstances. Because cells in the papillary collecting duct have urea transporters in their membrane when vasopressin acts, urea is usually an “ineffective” osmole (same concentration on both sides of that membrane) and it does not cause water to be excreted. The net result of excreting a small extra amount of urea is a higher Uosm, but not a higher “effective” Uosm or a higher urine flow rate.[14] Therefore, it is more correct to say that the urine flow rate is directly proportional to the number of non-urea or “effective” urine osmoles and inversely proportional to their concentration in the medullary interstitial compartment (see following equation).

Urine flow rate = # “Effective” urine osmoles/[“effective” urine osmoles]

In contrast, when the rate of excretion of urea rises by a large amount, urea might not be absorbed fast enough to achieve equal concentrations on both sides of a membrane. Hence, urea may become an “effective” osmole in the inner MCD and obligate the excretion of water. The analysis is not always that simple because urea is a partially “effective” urine osmole if the rate of excretion of electrolytes is low. [14] [15]

Tools

Urea Appearance Rate

The rate of appearance of urea can be determined from the amount of urea that is retained in the body plus the amount excreted in the urine over a given period of time. The former can be calculated from the rise in the concentration of urea in plasma (PUrea) and assuming a volume of distribution of urea equal to total body water (∼60% body weight).

Source of Urea

Close to 16% of the weight of protein is nitrogen. Therefore if 100 g of protein were oxidized, 16 g of nitrogen would be formed. Because the molecular weight of nitrogen is 14, there would be 1143 mmol of nitrogen produced. Since urea contains two atoms of nitrogen, 572 mmol urea are produced from the oxidation of 100 g of protein. In terms of lean body mass, because water is its main constituent (80% of weight), each kg has 800 g of water and 180 g of protein.[16] Therefore, breakdown of 1 kg of lean mass will produce ∼1000 mmol of urea. One can use this calculation to determine if the source of urea was exogenous or from endogenous breakdown of protein.

Consult SW-4: Osmotic Diuresis with An Emphasis on The Rate of Excretion of Urea

A 70-kg male had a recent bone marrow transplant. He was given large doses of corticosteroids. His 24-hour urine volume was 6 L/day and his Uosm was 500 mOsm/kg H2O. He did not receive mannitol, his PGlucose was 180 mg/dl (10 mmol/L), and his BUN was 210 mg/dL (PUrea was 75 mmol/L).

Questions

What is the cause of polyuria?

What is the major aim of therapy with respect to urea excretion?

Discussion

What is the cause of the polyuria?

Osmole excretion rate: There is an osmotic diuresis because his osmole excretion rate was 3000 mosmol/day (6 L urine/day and Uosm of 500 mOsm/kg H2O).

Nature of the Osmoles

His PGlucose was too low for high rates of glucosuria. On the other hand, his PUrea was high enough (75 mmol/L) to produce a sufficient quantity of filtered urea to cause the osmotic diuresis—this was confirmed later when his UUreawas measured (400 mmol/L). Because his urine volume was 6 L, he excreted ∼2400 mmol of urea that day.

Source of the Urine Urea

These 2400 mmol of urea would be produced from the oxidation of 420 g of protein. On that day, he was given 60 g of protein by nasogastric tube so he oxidized approximately 360 g of endogenous protein. This excretion of urea represents the catabolism of 2 kg of lean body mass (360 g/180 g/kg). If this were to continue, he would ultimately undergo marked muscle wasting. This can cause a problem because of compromised respiratory muscle function leading to bronchopneumonia. Furthermore, this catabolic state could affect his immunological defense mechanisms.[17]

What Is the Major Aim of Therapy?

Once this metabolic problem is recognized, therapy must be more vigorous at the nutritional level. First, more exogenous calories and protein must be given. Second, therapy could include anabolic hormones such as high-dose insulin (with glucose to avoid hypoglycemia) or anabolic steroids and/or the provision of nutritional supplements such as glutamine to minimize endogenous protein catabolism.[17] Third, one should be cautious about continued use of high doses of the catabolic hormone, glucocorticoids.

Concept SW-5

If there is no change in the number of osmoles in the ICF compartment, the volume of the ICF compartment is inversely proportional to the PNa.

When the PNa rises, the cell volume shrinks and when the PNa falls, the cell volume swells ( Fig. 24-6 ). The volume of brain cells is most important in this regard because the brain is constrained by the rigid skull.

000501

000519

FIGURE 24-6  The PNa reflects the ICF volume. The solid circle represents a cell. The ICF compartment contains macromolecular anions (P-) that are restricted to this compartment along with its major effective osmole, K+. The osmoles restricted to the ECF compartment are Na+ and its attendant anions (Cl- and HCO3-). Urea is not an effective osmole because there is a urea transporter that permits urea to achieve near-equal concentrations in the ECF and ICF compartments. There is osmotic equilibrium because water can cross the cell membrane rapidly through water channels (AQP1) in the cell membrane. The PNa is inversely proportional to the ICF volume (e.g., in acute hyponatremia shown on the left). Exceptions to this rule are when there are other “effective” osmoles in the ECF compartment (e.g., glucose, mannitol) or when the number of “effective” osmoles changes in the ICF compartment (e.g., brain cell volume regulation in chronic hyponatremia [shown on the right], or during a seizure [not shown]).

000519

 

Tools

Tonicity Balance

To decide what the basis is for a change in the PNa and to define the proper therapy to return the PNa, ECF and ICF volumes to their normal value, separate balances for water and Na+ must be calculated.[18] To perform this calculation, one must examine the volume and electrolyte composition of all the fluids ingested and infused and that of all outputs over the period when the PNa changed ( Fig. 24-7 ). In practical terms, a tonicity balance can be performed only in a hospital setting where inputs and outputs are accurately recorded. With regard to the output, this can be restricted to the urine in an acute setting. In a febrile patient or one who has been in the ICU for a prolonged period, balance calculations will not be as accurate because sweat losses are not measured. For example, if the balance for water is recorded in the hospital chart, the clinician can determine why the PNa changed.[18]

000508

000519

FIGURE 24-7  Calculation of a tonicity balance. The rectangle represents the body with its concentration of Na+ in the ECF compartment at the beginning (written above this rectangle) and at the end (written below this rectangle) of this period where the tonicity balance is calculated. The input of Na+ + K+ and of water are shown on the left and their outputs are shown on the right of this rectangle. The essential data to calculate a tonicity balance are shown to the left of the vertical dashed line and actual values from Consult SW-5 are shown to the right of that line.

000519

 

When calculating a Na+ balance, be careful not to multiply the concentration of Na+ in the terminal potion of a long urine collection period by the total urine volume because the urine composition may have changed during this time.

Calculation of An Electrolyte-Free Water Balance

To perform this calculation, one must also know the volume and the concentrations of Na+ + K+ in the input and in the urine. The first step is decide how much water is needed to convert all of the Na+ + K+ into an isotonic saline solution (e.g., 150 mmol in 1 L of water in molal terms if the Posm is in the normal range). For example, if the concentration of Na+ + K+ in a urine sample were lower than in plasma, the remainder of the volume is called electrolyte-free water. Alternatively, had there been residual Na+ + K+, the excretion of the remaining electrolytes is given the very confusing name of “negative” electrolyte-free water. As shown in Table 24-2 , an electrolyte-free water balance cannot distinguish between negative balances of water and positive balances for Na+ as the cause of the rise in the PNa. Accordingly, we do not use electrolyte-free water balances as they do not reveal the basis of the change in the PNa and hence do not help with the design of therapy to return the volume and composition of the ECF and ICF compartments to their normal values.


TABLE 24-2   -- Comparison of an Electrolyte-free Water Balance and a Tonicity Balance

 

Na + K (mmol)

Water (L)

EFW (L)

Therapy from Balances

 

 

 

 

EFW

Tonicity

Consult SW-5

 Input

450

3

0

+2 L

+0 L water

 Output

150

3

2

? Na+

-300 mmol Na+

 Balance

+300

0

-2

 

 

Case if 4 L of isotonic saline were infused and the urine output was unchanged

 Input

600

4

0

+2 L

-1 L

 Output

150

3

2

? Na+

-450 mmol Na+

 Balance

+450

+1

-2

 

 

Case if no intravenous fluid was administered and the urine output was unchanged

 Input

0

0

0

+2 L

+3 L

 Output

150

3

2

? Na+

+150 mmol Na+

 Balance

-150

-3

-2

 

 

 

Three situations are described where the PNa rose from 140 mmol/L to 150 mmol/L. The only difference is the volume of isotonic saline infused over the time period of observation. In all three settings, there is a negative balance of 2 L of electrolyte-free water. Nevertheless, the goals of therapy to correct the hypernatremia are clear only after a tonicity balance is calculated.

EFW, electrolyte-free water.

 

 

 

Consult SW-5: A Water Diuresis and An Osmotic Diuresis in the Same Patient?

A craniopharyngioma was resected today from a 16-year-old male (weight 50 kg, total body water 30 L). During neurosurgery, his urine flow rate rose to 10 ml/min (3 L in 300 min) and his PNa rose from 140 mmol/L to 150 mmol/L. Over this period, he received 3 L of isotonic saline. His Uosm was 120 mOsm/kg H2O, and his urine Na+ + K+ concentration was 50 mmol/L. To confirm the diagnosis of central DI, he was given dDAVP; his urine flow rate fell to 6 ml/min, the UNa+K rose to 175 mmol/L, and his Uosm was 375 mOsm/kg H2O.

Questions

What is the “expected” renal response to dDAVP in this patient?

Why did his PNa rise from 140 mmol/L to 150 mmol/L during his large water diuresis?

Discussion

What is the “Expected” Renal Response to dDAVP?

The initial information that will be available to assess the response to dDAVP is his urine flow rate; it is directly proportional to the osmole excretion rate and inversely proportional to the medullary interstitial osmolality.

Osmole excretion rate: The osmole excretion rate before dDAVP is given is known—it is equal to the product of the Uosm (120 mOsm/kg H2O) and urine flow rate (10 ml/min), or 1.2 mosmol/min; this excretion rate is more than double the expected rate of 0.5 mosmol/min.

Uosm: There will be a delay before the actual value is known, but we can predict what it might be; the “expected” value is ∼600 mOsm/kg H2O in an osmotic diuresis, but because the preceding large water diuresis would have “washed out” or lowered the osmolality in the medullary interstitial compartment, a reasonable prediction would be ∼400 mOsm/kg H2O.

Nature of the osmoles: Prior to the administration of dDAVP, 5 of 6 of the osmoles in the urine were Na+ + K+ salts (UNa + UK = 50 mmol/L, Uosm 120 mOsm/kg H2O). Because electrolytes are “effective” osmoles whereas urea is not usually an “effective” osmole when vasopressin acts, his “effective” osmole excretion rate is even higher than expected from the osmole excretion rate calculated earlier.

Summary: Based on all of the previous discussion, we would not be surprised to find a urine flow rate of ∼5 ml/min after dDAVP acts. In fact, once the effect of the anesthetic agent that diminished the tone of venous capacitance vessels abates, there will be an even higher “effective” osmole excretion rate (i.e., a larger natriuresis driven by the higher central venous pressure).

Why Did His PNa Rise from 140 mmol/L to 150 mmol/L during His Large Water Diuresis?

Tonicity balance: The patient received 3 L of water (as isotonic saline, but it is still 3 L of water in volume terms) and he excreted 3 L of urine (with a UNa + UK of 50 mmol/L, but it too is still water in volume terms). Accordingly, there is a nil balance of water.

Balance data for Na+ + K+ reveal that he received 450 mmol (3 L × 150 mmol/L) and he excreted only 150 mmol (3 L × 50 mmol/L). Hence he has a positive balance of 300 mmol of Na+ + K+. Dividing this surplus by the total body water (30 L) suggests that the rise in PNa should be 10 mmol/L, a value equal to the actual rise in the PNa. Therefore the basis for the rise in PNa was a gain of Na+ and not a loss of water. The proper treatment to restore body tonicity and the volume and composition of the ECF and ICF compartments is to induce a loss of 300 mmol of Na+.

Parenthetically, even if there were no measurements of Na+ or K+ in the urine, these values can be calculated for the period where one has the following measured values; the PNa at the beginning and end of the balance period, water balance for that period, and the quantity of Na+ + K+ infused; details of the calculation and its verification can be found in Ref. 19.

Integration: Clinical Approach to the Patient with Polyuria

In a patient with polyuria, a water diuresis is present when the Uosm is appreciably <250 mOsm/L (see Fig. 24-2 ). To determine if it is due to a lack of vasopressin (central DI) or an inability of vasopressin to act (nephrogenic DI), this hormone should be administered. Patients with circulating vasopressinase (e.g., tissue necrosis) respond to the administration of dDAVP, but fail to respond to the administration of vasopressin. As shown in Consult SW-5, the “expected” response should be adjusted for the calculated osmole excretion rate and the lower medullary interstitial osmolality due to prior medullary wash out. Another caveat in interpreting the urine data after the administration of dDAVP is that in absence of vasopressin action, the urine flow rate is determined by of the distal delivery of filtrate. Therefore, a lower urine flow rate and a higher Uosm may be observed if there is diminished distal delivery of filtrate because of a fall in the GFR. The clinical clues for this scenario are a fall in blood pressure and a fall in the osmole excretion rate.

To diagnose an osmotic diuresis, vasopressin must act, the Uosm must exceed the Posm, and the osmole excretion rate must be >1000 mosmol/day (see Fig. 24-5 ). Another factor that influences the urine flow rate in an osmotic diuresis is the medullary interstitial osmolality. To deduce which osmole is responsible for the osmotic diuresis, assess whether there is a high PGlucose or PUrea, whether sufficient amount of mannitol was infused, whether the patient has received a large amount of saline; and/or whether the patient has renal or cerebral salt wasting. The diagnosis of an osmotic diuresis has several important implications for the patient. First, it can induce a loss of Na+ in the urine, and thereby lead to contraction of the ECF volume (e.g., the patient with diabetic ketoacidosis [DKA]). Second, because the concentration of Na+ in the urine is usually much less than the PNa, the urinary loss can lead to the development of hypernatremia. Third, the source of the urine osmoles may be a high rate of excretion of glucose or urea that was derived from lean body mass (see “Consult SW-4”).

To know whether a water or an osmotic diuresis will lead to a rise (or fall) in the PNa and to design appropriate therapy, one should calculate separate balances for water and Na+ (a tonicity balance).

Concentrating Defect in the Renal Medulla

There are two factors that influence the urine flow rate when vasopressin acts, the number of “effective” osmoles in the luminal fluid of the inner medullary collecting duct and the osmolality of the fluid in the medullary interstitial compartment. To illustrate the importance of each of these factors, consider a patient who has sickle cell anemia. In this disorder, there is obstruction of the blood vessels by sickled cells deep in the inner medulla. As a result, there is necrosis of the renal papilla. Since the inner medullary collecting duct is the nephron site where vasopressin causes the insertion of urea transporters, patients with papillary necrosis cannot reabsorb urea at appreciable rates when vasopressin acts. Therefore, urea becomes an “effective” urine osmole in this setting and it obligates the excretion of water. In addition, if urea cannot be reabsorbed as an isosmotic solution in the inner medullary collecting duct, this will prevent the reabsorption of Na+ and Cl- from the thin ascending limb of the loop of Henle. Accordingly, the maximum medullary interstitial osmolality will be ∼600 mOsm/kg H2O. If this patient excretes 900 mosmol/day and all of the urine osmoles are “effective”, the minimum urine volume will be 1.5 L/day. Contrast these numbers with a subject who has a similar osmole excretion rate (1/2 urea, 1/2 electrolytes) who has a normal papilla, but a maximum Uosm that is also 600 mosmol/kg H2O. After vasopressin acts, there will be an insertion of urea transporters in the inner medullary collecting duct and the concentration of urea will be equal in the lumen of the inner medullary collecting duct and in the papillary interstitial compartment. Therefore, his rate of excretion of effective osmoles would be 1/2 of 900 mosmol/day or 450 mosmol/day. Accordingly, his daily urine volume would be 0.75 L (450 mosmol/600 mOsm/L), much less than in the patient with sickle cell anemia.

This example illustrates the complexities of a concentrating defect with a given maximum Uosm because if the lesion converts urinary urea into an “effective” urine osmole, the daily urine volume will increase ∼ twofold.

Defense of the Extracellular Fluid Volume

Concept SW-6

The volume of the ECF compartment is largely determined by its quantity of Na+.

The most reliable way to know how much Na+ is present in the ECF compartment is to measure the PNa and to multiply this value by the ECF volume. This requires a quantitative assessment of the ECF volume.

Concept SW-7

Na+ wasting can only be diagnosed when there is excretion of too much Na+; the term “wasting” implies that the ECF volume must be low.

To make a diagnosis of renal or cerebral salt wasting, one needs a quantitative estimate of the ECF volume.

Tools

Estimate the Extracellular Fluid Volume

The physical examination, the concentrations of K+, HCO3-, creatinine, urea, and urate in plasma as well as the fractional excretions of the latter two are useful at times to imply that the “effective” ECF volume may be contracted. Nevertheless, none of the above provides a quantitative estimate of the ECF volume. For the latter, we rely on the hematocrit (or a change in the hematocrit with therapy in the patient who has anemia or erythrocytosis) ( Table 24-3 ).


TABLE 24-3   -- Use of the Hematocrit to Estimate the Extracellular Fluid Volume

Hematocrit

Hemoglobin (g/L)

% Change in ECF Volume

0.40

140

0

0.50

171

-33

0.60

210

-60

 

The assumptions made when using this calculation are that the patient did not have anemia or erythrocytosis, that the RBC volume is 2 L, and the plasma volume is 3 L (blood volume 5 L). The formula is: hematocrit = RBC volume/(RBC volume + plasma volume).

ECF, extracellular fluid.

 

 

 

Sample calculation: In a normal adult, the usual hematocrit is 0.40; this represents a blood volume of 5 L (2 L of red blood cells (RBC) and 3 L of plasma; see following equation). Therefore, with a hematocrit of 0.50, there are still 2 L of RBC. Solving for “X”—the present blood volume—it is 4 L and the plasma volume is 2 L (+2 L RBC)—hence the plasma volume is reduced by 1 L from its normal 3 L value. Ignoring changes in Starling forces for simplicity, the ECF volume should have declined to approximately two thirds of its normal volume.

Hematocrit (0.40) = 2 L RBC/Blood volume (2 L RBC + 3 L plasma)

Hematocrit (0.50) = 2 L RBC/“X” L blood volume

Basis for the Low Extracellular Fluid Volume

Measuring the urine electrolytes can be very helpful to gain insights into the basis for a contracted ECF volume ( Table 24-4 ). As background, the expected response to a low “effective” arterial blood volume is to excrete as little Na+ and Cl- as possible. Because timed urine collections to calculate absolute rates of excretion of Na+ and Cl- are seldom obtained, clinicians should interpret the UNa and UCl in a spot urine sample. A low UNa or a low UCl (or both) does not necessarily indicate a low rate of excretion of Na+ and/or Cl- if the urine flow rate is high. To avoid this type of error, the UNa and UCl should be related to the urine creatinine concentration (UCreatinine).


TABLE 24-4   -- Urine Electrolytes in the Differential Diagnosis of Extracellular Fluid Volume Contraction

Condition

Urine Electrolyte

 

Na+

Cl-

Vomiting

 Recent

High[*]

Low[†]

 Remote

Low

Low

Diuretics

 Recent

High

High

 Remote

Low

Low

Diarrhea or laxative abuse

Low

High

Bartter or Gitelman syndrome

High

High

 

Adjust values for the urine electrolyte concentration when polyuria is present.

 

*

High = Urine concentration > 15 mmol/L.

Low = Urine concentration < 15 mmol/L.

 

A Low Rate of Excretion of Na+ and Cl-

This pattern may suggest a low intake of NaCl or that the effective arterial blood volume is contracted due to a loss of NaCl by a non-renal route (e.g., by sweating), or that there was a prior renal loss of Na+ and Cl- (e.g., remote use of diuretics).

A High Rate of Excretion of Na+ but Little Excretion of Cl-

In a patient with a low effective arterial blood volume, there is an anion other than Cl- being excreted with Na+. If the anion is HCO3- (the urine pH is alkaline), suspect recent vomiting. The anion could also be one that was ingested or administered (e.g., penicillin) in which case, the urine pH is usually less than 6.

A High Rate of Excretion of Cl- but Little Excretion of Na+

In this patient with a low “effective” arterial blood volume, there is a cation being excreted with Cl- other than Na+. Most often the cation is NH4+ and the setting is diarrhea or laxative abuse. Rarely the cation could be K+ if KCl was ingested.

The Excretions of Na+ and Cl- Are Not Low

In a patient who has a low effective arterial blood volume, a high rate of excretion of both Na+ and Cl- suggests that the patient has a deficit of a stimulator of the reabsorption of Na+ and Cl- (e.g., aldosterone), the presence of an inhibitor of the reabsorption of NaCl (e.g., a diuretic), or an intrinsic renal lesion that is similar to having the actions of a diuretic (discussed later in “Potassium and Metabolic Alkalosis”). The pattern of excretion of electrolytes throughout the day can also be very important. For example, if there are times when the UNa and UCl are both low, this suggests that the patient is taking a diuretic.

Fractional Excretion of Na+ or Cl-

On a typical western diet, the daily excretions of Na+ and Cl- are ∼150 mmol. Because a normal GFR is ∼180 L/day, the kidney filters ∼27,000 mmol of Na+ and 20,000 mmol of Cl- per day (adjusting the PNa and PCl per plasma water). Rather then expressing this function in fractional reabsorption terms (>99%), it is common to express them as fractional excretion terms (FENa ∼ 0.5%, FECl ∼ 0.75%). To make this calculation of the fractional excretion as simple as possible, the urine to plasma ratio for creatinine is used (see following equation).

FENa = 100 × (UNa/PNa)/(UCreatinine/PCreatinine)

There are three practical points to bear in mind when using the FENa or FECl. First, the excretions of Na+ and Cl- are directly related to the dietary intake of NaCl. Hence a low FENa or FECl may represent either a low effective arterial blood volume or a low intake of NaCl (or both). Second, the FENa or FECl may be high in a patient with a low “effective” arterial blood volume when there is an unusually large excretion of another anion (e.g., HCO3-) in the case of Na+ or another cation (e.g., NH4+) in the case of Cl-. Third, the numeric values for the FENa and the FECl will be twice as high in a euvolemic patient who consumes 150 mmol of NaCl daily if that patient has a GFR that is half of normal. Hence the clinical significance of these FENa and FECl numeric values must be adjusted for the GFR at the time when the measurements are made. In addition, there are problems with respect to accuracy when creatinine is used to measure the GFR. Nevertheless, the use of these parameters may be of value in the differential diagnosis of pre-renal azotemia versus acute tubular necrosis (ATN).[20] The advantage of using these tests in this setting over the use of the UNa and UCl is that the use of UCreatinine/PCreatinine adjusts these concentration terms for water reabsorption in the nephron.

Determine the Nephron Site with an Abnormal Reabsorption of Na+

Look for Failure to Reabsorb Other Substances

If a compound or ion that should have been reabsorbed in a given nephron segment is being excreted, one has presumptive evidence for a reabsorptive defect in that nephron segment. For example, if the defect is in the PCT, one might find glucosuria in the absence of hyperglycemia.

Compensatory Effects in Downstream Nephron Segments

When there is inhibition of the reabsorption of NaCl in upstream nephron segments, more NaCl is delivered to the CCD, where the reabsorption of Na+ may occur in conjunction with K+ secretion.

Consult SW-6: Assessment of the “Effective” Arterial Blood Volume

A 25-year-old female was assessed by her family physician because of progressive weakness. Although she admitted to being concerned about her body image, she denied vomiting or the intake of diuretics. Her blood pressure was 90/60 mm Hg, her pulse rate was 110 beats/min, and her jugular venous pressure was low. Acid-base measurements in arterial blood revealed a pH 7.39 and a PCO2 of 39 mm Hg. In results from venous blood, her PHCO3 was 24 mmol/L, PAnion gap was 17 mEq/L, PK was 1.9 mmol/L, hematocrit was 0.50, and her PAlbumin was 5.0 g/dL (50 g/L). The urine electrolytes prior to therapy were UNa 0 mmol/L, UCl 42 mmol/L, and UK 23 mmol/L.

Questions

How severe is her degree of ECF volume contraction?

What is the cause of her low ECF volume?

Discussion

How Severe Is Her Degree of extracellular Fluid Volume Contraction?

The elevated value for the hematocrit (0.50) provides quantitative information about her ECF volume (see Table 24-3 )[21]; her ECF volume is reduced by 33% if she did not have anemia prior to therapy. If anemia were present, her ECF volume would be even more reduced.

What Is the Cause of Her Low Extracellular Fluid Volume?

The low UNa implies that the “effective” arterial blood volume is low if the patient is not consuming a low salt diet. Nevertheless, the high UCl (42 mmol/L) does not necessarily indicate an intrinsic renal abnormality. Rather, the fact that her UCl exceeded the sum of her UNa + UK suggested that there was another cation in that urine, most likely NH4+.

Interpretation

Calculating the content of HCO3- in her ECF reveals that she had a deficit of NaHCO3 (see “Metabolic Acidosis” for more discussion). Loss of NaCl plus NaHCO3 via the GI tract was suspected as the cause of contracted ECF volume. The patient later admitted to the frequent use of a laxative. Hence the hypokalemia and contracted effective ECF volume are easily accounted for. Hypokalemia stimulated ammoniagenesis, raising the rate of excretion of the cation NH4+; this obligated the excretion of Cl- despite the presence of a low effective circulating volume.

POTASSIUM AND METABOLIC ALKALOSIS

We combine disorders of K+ homeostasis and metabolic alkalosis in this section because a deficiency of KCl plays an important role in the pathophysiology of metabolic alkalosis.

Dyskalemias

Hypokalemia and hyperkalemia are common electrolyte disorders in clinical practice that may precipitate life-threatening cardiac arrhythmias. Data from urine measurements provide essential evidence to establish their underlying pathophysiology and to suggest options for therapy.

Concept K-1

There are two factors that influence the movement of K+ across cell membranes: first, a driving force, which is the concentration difference for K+ and the electrical voltage across cell membranes and second, the presence of open K+ channels in cell membranes.

K+ are kept inside the cells by a net negative interior voltage because the Na-K-ATPase is an electrogenic pump—exporting 3 Na+ while importing 2 K+.[22] The Na-K-ATPase can cause more K+ ions to enter cells if there is a higher concentration of intracellular Na+ or if this ion pump is activated by β2-adrenergics, thyroid hormone, or insulin.[23] For the former to result in an increase in cell negative voltage, Na+ entry into cells must be electroneutral (e.g., via the Na+/H+ exchanger (NHE) ( Fig. 24-8 ). NHE is almost always inactive in cell membranes, but it becomes active if there is a high concentration of insulin and/or a high H+ concentration in the ICF compartment.[24]

000704

000519

FIGURE 24-8  Factors influencing the movement of K+ across cell membranes. The passive movement of K+ across cell membranes requires a driving force (voltage and/or a concentration difference) and an open K+ channel. The active transport of Na+ from cells is by the electrogenic Na-K-ATPase increases the intracellular negative voltage. The source of the intracellular Na+ is the electro-neutral entry of Na+ via NHE (activated by insulin or a high ICF [H+]) or the existing ICF Na+ content. β2 adrenergic agonists activate the Na-K-ATPase.

000519

 

Concept K-2

There is no normal rate of K+ excretion in the urine because normal subjects in steady state excrete all the K+ they eat and absorb from the GI tract.

In a patient with hypokalemia or hyperkalemia, the “expected” rate of K+ excretion is the one observed when normal subjects were deprived of K+ or given a large load of KCl.

Concept K-3

Chronic disorders of K+ homeostasis are due to abnormal rates of renal excretion of K+.

This regulation occurs primarily in the late cortical distal nephron including the cortical collecting duct (CCD). This process has two components—the secretion of K+ by principal cells and the rate of flow traversing the CCD.

Tools

Rate of Excretion of K+

The appropriate renal response is to excrete as little K+ as possible when there is a deficit of K+ (i.e., <15 mmol/day)[25] and to excrete as much K+ as possible when there is a surplus of K+ (i.e., >200 mmol/day).[26] A 24-hour urine collection is not necessary to assess the rate of excretion of K+. Taking advantage of the fact that creatinine is excreted at a near-constant rate throughout the day (200 mmol/min/kg body weight or 20 mg/min/kg body weight),[27] the same information about the rate of excretion of K+ from a 24-hour urine collection can be obtained by examining the UK/UCreatinine ratio in a spot-urine. Furthermore, The UK/UCreatinine has an advantage because the data are available very quickly and more relevant information is gathered because one knows the stimulus (PK) influencing K+ excretion at that time. On the other hand, it has a disadvantage because there is a diurnal variation in K+excretion,[28] but this does not negate the advantages. The expected UK/UCreatinine ratio in a patient with hypokalemia is <15 mmol K+/g creatinine (or 1.5 mmol K+/mmol creatinine) whereas this ratio should be >200 mmol K+/g creatinine (>20 in mmol K+/mmol creatinine) in a patient with hyperkalemia.

Calculate the Components of the Rate of Excretion of K+ in the terminal Cortical Collecting Duct

If the rate of K+ excretion is inappropriate for the presence of hypokalemia or hyperkalemia, both components of the K+ excretion formula (flow rate and [K+]) should be examined in terms of events in the terminal CCD (see following equation).[29]

K+ in the lumen of the CCD = [K+]CCD × Flow rateCCD

Flow rate in the terminal CCD: When vasopressin acts, the luminal osmolality is equal to the Posm and the flow rate in the terminal CCD is determined by the rate of delivery of osmoles because the luminal osmolality is relatively constant (equal to the Posm) and all the luminal osmoles are “effective” ones in this nephron site. Therefore one can obtain a minimum estimate of the rate of flow in terminal CCD by dividing the rate of excretion of osmoles by the Posm (see following equation) ( Fig. 24-9 ). This provides a minimum estimate of flow rate in terminal CCD because of reabsorption of some osmoles in the medullary collecting duct.

000509

000519

FIGURE 24-9  Non-invasive estimate of the flow rate and the [K+] in the terminal CCD. The barrel-shaped structures represent the CCD and the arrow below it represents the medullary collecting duct (MCD). As shown in the right side of the CCD, when vasopressin acts, the Posm and the osmolality in the luminal fluid in the terminal CCD are equal (e.g., 300 mOsm/kg H2O); hence the flow rate in the terminal CCD is determined by the rate of delivery of osmoles. In the example shown on the left side, the luminal K+ concentration is 40 mmol/L or 10-fold larger than the peritubular K+ concentration of 4 mmol/L. When 1 L of fluid traverses the MCD, 75% of this water is reabsorbed (and no K+ was reabsorbed or secreted in the MCD in this example). Hence the UK is fourfold higher (40 mmol/L to 160 mmol/L) as is the Uosm (300 to 1200 mOsm/kg H2O). To back-calculate the [K+]CCD, the UK is divided by the (U/P)osm.

000519

 

Flow rateCCD = (Urine flow rate × U osm)/Posm

The “usual” rate of osmole excretion is ∼0.5 mosmol/min or 600 mosmol/day to 900 mosmol/day so a minimum estimate of flow rate in the terminal CCD is 2 L/day to 3 L/day. The major osmoles in the terminal CCD are urea and Na+ plus Cl-. Hence a low flow rate in the terminal CCD could be due to a low delivery of urea (low intake of proteins) and/or of Na+ and Cl- (low effective circulating volume, low intake of salt). On the other hand, a high flow rate in the CCD could be due to inhibition of the reabsorption of NaCl in an upstream nephron segment (a high salt intake, use of a diuretic [osmotic or pharmacologic]), or diseases that lead to an inhibition of reabsorption of NaCl in an upstream nephron segment (e.g., Bartter syndrome or Gitelman syndrome). During a water diuresis, while the rate of flow in the CCD is high, the rate of excretion of K+ is not because vasopressin is required for K+ secretion in the CCD.[30]

[K+] in the lumen of the terminal CCD ([K+]CCD): A reasonable approximation of the [K+]CCD can be obtained by adjusting the UK for the amount of water reabsorbed in the MCD (see Fig. 24-9 ). The assumption here is that there is little K+ secretion or reabsorption occurs in the MCD (reasonable in most circumstances; see the following equation).

[K+]CCD = [K+]urine/(U/P)Osm

Transtubular [K+] Gradient (TTKG)

To calculate the TTKG, divide the [K+]CCD by the PK (see following equation). The TTKG offers an advantage over the [K+]CCD because it permits one to relate the [K+]CCD to the PK.[31] The expected value for the TTKG in a patient with hypokalemia due to a non-renal cause is <2, whereas the appropriate renal response to hyperkalemia is a TTKG > 7.

TTKG = [K+]CCD/PK

Establish the Basis for the Abnormal [K+]CCD

The driving force for the net secretion of K+ in the cortical distal nephron is a lumen-negative transepithelial voltage generated by the “electrogenic” reabsorption of Na+ (i.e., Na+ is reabsorbed faster than Cl-) ( Fig. 24-10 ). For secretion of K+, open channels for K+ must be in the luminal membrane.[32] In a patient with hypokalemia, a higher than expected [K+]CCD implies that the lumen-negative voltage is abnormally more negative and that open luminal ROMK channels are present in the CCD.[32] The former could be due to reabsorbing Na+ “faster” than Cl- in the CCD. The converse is true in a patient with hyperkalemia and a lower than expected [K+]CCD.

000717

000519

FIGURE 24-10  K+ secretion in the cortical collecting duct. The barrel-shaped structures represent the CCD and the rectangles represent principal cells. Na+ ions are reabsorbed via ENaC; their reabsorption is increased by aldosterone (the shaded enlarged circle). Net secretion of K+ occurs through their specific ion channel (ROM-K). Electroneutral reabsorption of Na+ and Cl- (shown to the left of the vertical dashed line) does not generate a lumen-negative voltage, which diminishes the secretion of K+. Electrogenic reabsorption of Na+ enhances the secretion of K+ (e.g., HCO3- and/or an alkaline luminal pH decreases the apparent permeability for Cl- in the CCD; this is shown to the right of the dashed vertical line).

000519

 

The clinical indices that help in this differential diagnosis of the pathophysiology of the abnormal rate of electrogenic reabsorption of Na+ in CCD are an assessment of the ECF volume and the ability to conserve Na+ and Cl- in response to a contracted effective arterial blood volume. The measurement of the activity of renin (Prenin) and the level of aldosterone in plasma (PAldosterone) are helpful in patients with hypokalemia ( Table 24-5 ).[29]


TABLE 24-5   -- Plasma Renin and Aldosterone to Assess the Basis of Hypokalemia due to a Fast Na Type Lesion

Adrenal gland lesion

Renin

Aldosterone

 Primary hyperaldosteronism or adrenal tumor

Low

High

 ACTH causes aldosterone synthesis (GRA)

Low

High

Kidney lesions

 

 

 Renal artery stenosis

High

High

 Malignant hypertension

High

High

 Renin-secreting tumor

High

High

 Liddle syndrome

Low

Low

 11β-HSDH fails to remove all cortisol

 

 

  Hereditary defect (AME)

Low

Low

  Inhibition (e.g., licorice ingestion)

Low

Low

  Saturated because of ectopic ACTH

Low

Usually low

 

For details, see text.

AME, apparent mineralocorticoid excess syndrome; 11β-HSDH, 11 β-hydroxysteroid dehydrogenase; GRA, glucocorticoid-remedial aldosteronism.

 

 

 

Consult K-1: Hypokalemia and A Low Rate of Excretion of K+

A 35-year-old obese, Asian male developed extreme weakness progressing to paralysis over a period of 12 hours. It was preceded by his routine exercise after eating a carbohydrate-rich meal. He has had three similar attacks of paralysis in the past 6 months, but there was no family history of hypokalemia, paralysis, or hyperthyroidism. He denied the intake of laxatives and diuretic use, but he did take amphetamines to induce weight loss. On physical examination, he was alert and oriented; blood pressure was 150/70 mm Hg, heart rate was 124 beats/min, and respiratory rate was 18 per minute. Symmetrical flaccid paralysis with areflexia was present in all four limbs. There were no other abnormal findings on examination. The pH and PCO2 in Table 24-6 were from arterial blood whereas all other data were from venous blood. The EKG showed prominent U waves. On subsequent evaluation, tests indicated thyroid function was normal.


TABLE 24-6   -- Data for Consult K-1

 

 

Blood

Urine

 

 

Blood

K+

mmol/L

1.8

10

pH

 

7.40

Creatinine

mg/dL

0.6

1 g/L

PCO2

mm Hg

40

Na+

mmol/L

140

HCO3-

mmol/L

25

Cl-

mmol/L

103

Glucose

mg/dL

84

 

 

 

Questions

Is there a medical emergency?

What is the basis of the hypokalemia?

What are the major options for therapy?

Discussion

Is There A Medical Emergency?

Because the EKG did not show significant changes due to hypokalemia and because respiratory muscle weakness was not evident from the arterial PCO2, there were no emergencies that required urgent therapy ( Fig. 24-11 ).

000514

000519

FIGURE 24-11  Initial approach to the patient with hypokalemia. In the initial approach, the major objectives are to rule out an emergency for which therapy must take precedence. The next step is to decide whether there is an important degree of shift of K+ into cells. One must follow this up with an examination of the renal response to hypokalemia (see Figure 24-12 ).

000519

 

What Is the Basis of the Hypokalemia?

Rate of excretion of K+: Because the time course was short, the major basis for his acute hypokalemia is a shift of K+ into cells. In support of this diagnosis, his UK/UCreatinine was <15 mmol/g creatinine (<1.5 in mmol/mmol terms).[33] There is a possible caveat—the low rate of excretion of K+ may represent the normal renal response to a prior K+ loss or an extra-renal loss of K+. The absence of an acid-base disorder suggested also that the major basis for his hypokalemia is an acute shift of K+ into cells ( Table 24-7 ).[33]


TABLE 24-7   -- Plasma Acid-Base Status and Hypokalemia

  

 

Patients with metabolic alkalosis

  

 

Vomiting, nasogastric suction, cationic resins, rare types of diarrhea

  

 

Diuretic use or abuse

  

 

Other disorders based on plasma renin activity

  

 

Settings with a low renin activity in plasma (e.g., primary aldosteronism, glucocorticoid remediable hypertension, Liddle syndrome, apparent mineralocorticoid excess syndrome, Cushing syndrome, licorice)

  

 

Settings with a high renin activity in plasma (e.g., renal artery stenosis, malignant hypertension, renin producing tumor, Bartter or Gitelman syndrome)

  

 

Patients with hyperchloremic metabolic acidosis

  

 

Overproduction of hippuric acid (e.g., toluene abuse)

  

 

GI loss of NaHCO3 (e.g., diarrhea, laxative abuse, fistula, ileus, ureteral diversion)

  

 

Reduced reabsorption of in the PCT (e.g., proximal RTA treated with large amounts of NaHCO3, use of acetazolamide)

  

 

Distal renal tubular acidosis

  

 

Low excretion of NH4+ (e.g., usually low net distal H+ secretion subtype, Sjögren syndrome, Southeast Asian Ovalocytosis with second mutation of Cl-/HCO3- anion exchanger)

 

 

 

Reasons for K+ to shift into cells: There were reasons to believe that K+ may have shifted into cells (e.g., a large carbohydrate intake [high insulin levels activate NHE]) and vigorous exercise (β2 adrenergic agonists stimulate Na-K-ATPase). Notwithstanding, the stimulus for K+ shift must be prolonged considering that his symptoms persisted for a long period of time. The absence of hypoglycemia made it unlikely that there was a prolonged excessive release of insulin. Long-term β-adrenergic stimulation can be caused by amphetamine that was used by this patient, or a large intake of caffeine, especially if the relevant cytochrome P450 for its metabolism is inhibited[34]; this was denied on careful questioning of the patient. There was no personal or family history of hypokalemic periodic paralysis (HPP) or hyperthyroidism.

Evidence for an adrenergic surge: On physical examination, there were findings supporting a high β-adrenergic state (e.g., tachycardia, systolic hypertension, and a wide pulse pressure). Other laboratory tests that suggest this diagnosis include the presence of hypophosphatemia with a low Uphosphate/Ucalcium ratio.[35]

Interpretation

The cause of the hypokalemia seemed to be an acute shift of K+ into cells due to a large adrenergic surge due to the use of amphetamines.[34] (More detailed information about causes for an acute shift of K+ into cells can be found in Ref. 33.)

What Are the Major Options for therapy?

Because the diagnosis was an acute shift of K+ into cells due to a large adrenergic surge due to the amphetamines, he was treated with a non-selective β-blocker and a moderate dose of K+ supplementation[36]; the PK returned to the normal range in 2 hours. Of great importance, large doses of K+ should not be given because of the risk of life- threatening rebound hyperkalemia when the reason for this shift of K+ abates.

Consult K-2: Hypokalemia and A High Rate of Excretion of K+

A 76-year-old Asian male developed progressive muscle weakness over the past 6 hours that became so severe that he became unable to move. He had no other neurological symptoms. He denied nausea, vomiting, diarrhea, or the use of diuretics or laxatives. Hypokalemia (PK 3.3 mmol/L) and hypertension were noted 1 year ago, but were not investigated further. His family history was negative for hypertension or hypokalemia. His blood pressure was 160/96 mm Hg and his heart rate was 70 per minute. Neurological examination revealed symmetric flaccid paralysis with areflexia, but no other findings. The laboratory data prior to therapy are shown in Table 24-8 . Subsequent measurements indicated that his Prenin and PAldosterone were low and his PCortisol was in the normal range.


TABLE 24-8   -- Data for Consult K-2

 

 

Blood

Urine

 

 

Blood

Urine

K+

mmol/L

1.8

26

pH

 

7.55

Na+

mmol/L

147

132

PCO2

mm Hg

40

Cl-

mmol/L

90

138

HCO3-

mmol/L

45

0

Creatinine

mg/dL

0.8

0.6 g/L

Osmolality

mOsm/L

302

482

 

 

 

The patient was treated initially with intravenous KCl; the weakness improved when his PK reached 2.5 mmol/L. He was continued on oral KCl supplementation. Two weeks later, his PK and blood pressure had returned to normal levels, while his body weight decreased from 78 kg to 74 kg.

Question

What is the cause of hypokalemia in this patient?

Discussion

Excretion of K+

The patient presented with an acute symptom—extreme weakness of both upper and lower limbs. There was little to support the diagnosis of HPP because he did not have previous attacks of paralysis and there was no evidence of thyrotoxicosis. Most importantly, his UK/UCreatinine was 5, a value that is fivefold higher than what is expected if the major basis for his hypokalemia was an acute shift of K+ into cells (see Fig. 24-11 ). Moreover, he had metabolic alkalosis. Hence his hypokalemia was largely due to a disorder that caused excessive loss of K+ into the urine.

Notwithstanding, his acute presentation with extreme weakness might be due to an acute shift of K+ into cells in conjunction with a chronic disorder that caused excessive excretion of K+. This component of an acute shift of K+into cells could have been induced by vigorous exercise and a large carbohydrate intake during breakfast prior to the onset of symptoms. Our approach to determine the pathophysiology of his hypokalemia will utilize the clinical tools discussed earlier and the step-by-step approach is illustrated in Figure 24-12 .

000668

000519

FIGURE 24-12  Renal causes of hypokalemia. In a patient with a high rate of excretion of K+, one must examine both the flow rate and the concentration of K+ in the luminal fluid exiting from the terminal CCD as discussed in the text.

000519

 

Assess the [K+]CCD

The [K+]CCD was very high in the presence of hypokalemia. This reflects a high lumen-negative voltage in the CCD, due to either a faster rate of reabsorption of Na+ or a slower rate of reabsorption of Cl- (see Fig. 24-9 ).

Establish the Basis for the Abnormal [K+]CCD

On clinical assessment, his ECF volume was not contracted and he had hypertension. Therefore the increased lumen-negative voltage in his CCD was likely due to a faster rate of reabsorption of Na+ (see Fig. 24-10 ). The differential diagnosis of disorders with a faster rate of reabsorption of Na+ in the CCD is guided by measurements of the Prenin and PAldosterone (see Table 24-5 ).

Interpretation

Because both PAldosterone and Prenin were suppressed, the differential diagnosis was between disorders in which cortisol acts as mineralocorticoid and those with an open ENaC despite the undetectable levels of aldosterone. A chest radiograph did not reveal a lung mass and cortisol levels were not elevated. Inherited disorders where ENaC is constitutively active (Liddle syndrome) seemed unlikely considering the patient's age. Although the patient denied consuming licorice or chewing tobacco, it turned out that an herbal sweetener he used to sweeten his tea contained large amounts of glycyrrhizic acid (the active principle in licorice).[37] Discontinuing this intake led to a normal PKand a fall in his blood pressure.

Consult K-3: Hyperkalemia in A Patient Taking Trimethoprim

A 35-year-old cachectic male with HIV developed Pneumocystis carinii jerovici (PJP). On admission, he was febrile; his ECF volume and all plasma electrolyte values were in the normal range. He was treated with clotrimazole (sulfamethoxazole and trimethoprim). Three days later, he was noted to have low blood pressure, his “effective” arterial blood volume was low, and his PK rose to 6.8 mmol/L. His urine volume was 0.8 L/day and his Uosm was 350 mOsm/L ( Table 24-9 ).


TABLE 24-9   -- Data for Consult K-3

 

 

Blood

Urine

 

 

Blood

Urine

K+

mmol/L

6.8

14

pH

 

7.30

Na+

mmol/L

130

80

PCO2

mm Hg

30

Cl-

mmol/L

105

63

HCO3-

mmol/L

15

0

Creatinine

mg/dL

0.9

0.8 g/L

BUN

mg/dL

14

420

 

 

 

Question

Why is hyperkalemia present?

Discussion

The steps to follow are provided in Figures 24-13 and 24-14 [13] [14]. Although an element of pseudohyperkalemia could be present in this cachectic patient, the presence of EKG changes indicated that he has true hyperkalemia.

000518

000519

FIGURE 24-13  Initial steps in the patient with hyperkalemia. In the initial approach, the major objectives are to rule out an emergency for which therapy must take precedence. The next step is to decide whether there is an important degree of shift of K+ out of cells. One must follow this up with an examination of the renal response to hyperkalemia (see Figure 24-14 ).

000519

 

000517

000519

FIGURE 24-14  Renal causes for hyperkalemia and a low rate of K+ excretion. In a patient with a hyperkalemia and a low rate of excretion of K+, one must examine both the flow rate and the concentration of K+ in the luminal fluid exiting from the terminal CCD as discussed in the text. NCC, Na, Cl, cotransporter.

000519

 

Rate of Excretion of K+

Because the rise in PK occurred over many days, one would be tempted to conclude that the major basis for the hyperkalemia was the low rate of excretion of K+.

Shift of K+ Out of Cells

Because the patient consumed little K+, a shift of K+ from cells rather than a large positive external balance for K+ should be the major cause of hyperkalemia.[29] The likely cause of this exit of K+ from cells could be cell necrosis, insulin deficiency, and/or hyperchloremic metabolic acidosis.[38] Insulin deficiency could be due to the α-adrenergic effect of adrenaline released in response to the low ECF volume.[39] Because the major basis of hyperkalemia is a shift of K+ from cells, it would be an error to induce a large loss of K+ when the total body K+ surplus is small. It is important to realize that he could also have a defect in K+ excretion.

Rate of Excretion of K+

His UK was 14 mmol/L and his rate of excretion of K+ was extremely low in the face of hyperkalemia (UK/UCreatinine was 17.5 mmol/g).

Flow Rate in His Terminal Cortical Collecting Ducts

This flow rate was low because his rate of excretion of osmoles was only 0.2 mosmol/min (Uosm 350 mOsm/kg H2O and the urine flow rate was 0.6 ml/min)—this is less than half the usual rate of excretion of osmoles (0.5 mosmol/min).[28] It was likely that he had a low rate of production of urea (low protein intake) and a low excretion of NaCl (low effective arterial blood volume and low intake of NaCl).

The [K+]CCD

His [K+]CCD was <2—this indicates a low rate of K+ secretion in his CCD, which implies a low lumen-negative voltage in the CCD due to less electrogenic reabsorption of Na+.

Establish the Basis for the Low [K+]CCD

Because he had a low effective arterial blood volume and a UNa and UCl that were inappropriately high in the presence of a contracted ECF volume, his low [K+]CCD was due to slower reabsorption of Na+ in the CCD ( Fig. 24-15 ).

000874

000519

FIGURE 24-15  Basis for the abnormal [K+] in the cortical collecting duct. The barrel-shaped structure represents the CCD. A lumen-negative voltage is generated by the electrogenic reabsorption of Na+. Slower pathways are indicated by smaller open circles with dashed arrows and faster ones by large open circles with bold arrows.

000519

 

The major groups of disorders that can cause a slower reabsorption of Na+ in the CCD are listed in Table 24-10 . Because he did not have a response to exogenous mineralocorticoids, the presumptive diagnosis was that his slower Na+ reabsorption was due to inhibition of ENaC by the trimethoprim that was used to manage his PJP.[40] Both his Prenin and PAldosterone (which became available later) were high as expected in this setting.


TABLE 24-10   -- Renal Causes of a Dyskalemia

  

1.   

Hyperkalemia

  

(i) 

Slower reabsorption of Na+ in the CCD

  

 

Low levels of aldosterone (e.g., Addison disease)

  

 

Blockade of the aldosterone receptor (e.g., spironolactone)

  

 

Low activity of ENaC (e.g., amiloride, trimethoprim)

  

(ii) 

Faster reabsorption of Cl- in the CCD

  

 

Gordon syndrome (e.g., WNK kinase mutations), drugs (e.g., cyclosporin)

  

2.   

Hypokalemia

  

(i) 

Faster reabsorption of Na+ in the CCD

  

 

High aldosterone levels

  

 

Cortisol acts as a mineralocorticoid

  

 

Low 11-β hydroxysteroid dehydrogenase (11-βHSDH) activity (AME)

  

 

Inhibitors of 11-βHSDH (e.g., licorice)

  

 

Very high cortisol levels (e.g., ACTH-producing tumor)

  

 

Constitutively active ENaC (e.g., Liddle syndrome)

  

 

Artificial ENaC (e.g., amphotericin B)

  

(ii) 

Slower reabsorption of Cl- in the CCD

  

 

Delivery of Na+ to the CCD with little Cl-

  

 

Higher delivery of Na+ and Cl- to the CCD when ENaC is active (e.g., Bartter or Gitelman syndrome, loop or thiazide diuretic plus a contracted ECF volume)

  

 

Inhibition of Cl- reabsorption in the CCD (e.g., bicarbonaturia)

 

AME, Apparent mineralocorticoid excess syndrome; CCD, cortical collecting duct.

Chronic kidney disease is not included in this table as the mechanism is not well-defined.

 

 

 

Interpretation

Renal salt wasting due to blockade of ENaC by trimethoprim led to the development of a contracted ECF volume. As a result, there was a shift of K+ out of cells probably because of inhibition of insulin release by α-adrenergics.[39]Because of the low ECF volume (and the low intake of proteins), there was a low rate of flow in the CCD. This low flow rate in the CCD, in addition to its effect to diminish the rate of K+ excretion, caused the trimethoprim concentration to be higher in the lumen of the CCD (same amount of trimethoprim in a smaller volume); hence trimethoprim became a more effective blocker of ENaC.

From a therapeutic point of view, the question arose as to whether trimethoprim should be discontinued. Because the drug was needed to manage his PJP, a means to remove its renal ENaC-blocking effect was sought. Increasing flow in the CCD using a loop diuretic plus the infusion of enough NaCl to re-expand his ECF volume should lower the concentration of trimethoprim in the lumen of the CCD (same amount of trimethoprim in a much larger volume). Inducing bicarbonaturia could also be considered to lower the concentration of H+ in the luminal fluid in the CCD and thereby the cationic form of the drug that blocks ENaC.[41]

Integration: Clinical Approach, Hypokalemia

The first step in the clinical approach to the patient with hypokalemia is to identify whether there is an emergency prior to therapy (cardiac arrhythmia and/or respiratory muscle weakness) and to anticipate and avoid dangers associated with therapy (e.g., administration of glucose leading to the release of insulin and a shift of K+ into cells).

The second step is to decide if a major basis for hypokalemia is an acute shift of K+ into cells (see Fig. 24-11 ). The time course, a family or past personal history of hypokalemia, and/or episodes of muscle weakness or paralysis, male gender, Asian ethnicity, as well as other issues from the history that suggest the presence of a cause for a K+ shift into cells are very helpful to make this decision. In addition, physical findings that suggest an adrenergic surge, the absence of an acid-base disorder (see Table 24-7 ), and a low UK/UCreatinine are excellent indicators to suggest that the diagnosis might be a condition that caused an acute shift of K+ into cells.

The third step is to examine the rate of K excretion using the UK/UCreatinine to identify renal causes for hypokalemia (see Fig. 24-12 ). A ratio >1.5 indicates a high rate of excretion of K+. At this point, a reason for this higher [K+]CCD should be sought—i.e., the reason why the voltage is more negative in the lumen of the CCD. This could be due to a faster rate of reabsorption of Na+ or a slower rate of reabsorption of Cl- in the CCD (see Fig. 24-15 ).

Patients with disorders that cause a faster reabsorption of Na+ often have hypertension because they have an expanded ECF volume. The UCl reflects their intake of salt. Measurements of Prenin and PAldosterone provide means to identify the cause in this group of patients (see Table 24-5 ).

Patients with disorders that cause a slower reabsorption of Cl- (than Na+) can be divided into three groups. First, there is a delivery of very little Cl- to the CCD (e.g., delivery of Na+ with HCO3- [recent vomiting] or a drug anion like penicillin); second, there is a decreased rate of reabsorption of Cl- in the CCD (e.g., possibly due to the effect of HCO3- or an alkaline luminal pH[42]); third, there is a combination of a high delivery of Na+ and Cl- to the CCD due to inhibition of their absorption in an upstream nephron segment together with a higher capacity for the reabsorption of Na+ than Cl- in the CCD (e.g., release of aldosterone in response to a low ECF volume). Patients with a slow Cl- type of lesion should have a high Prenin.

Integration: Clinical Approach, Hyperkalemia

It is imperative to recognize when hyperkalemia represents a medical emergency because therapy must then take precedence over diagnosis (see Fig. 24-13 ); this emergency is usually secondary to a cardiac arrhythmia. If there is no emergency present, the second step is to assess whether there is a component of pseudohyperkalemia (including hemolysis, megakaryocytosis, fragile leukemic cells, a K+ channel disorder in red blood cells,[43] and excessive fist clenching during blood sampling[44]). Even without fist clenching, pseudohyperkalemia can be present in cachectic patients because the normal T-tubule architecture in skeletal muscle can be disturbed thus permitting more K+ to be released into venous blood. (Normally K+ is released during muscle depolarization into a region that does not mix appreciably with the circulating volume.) The third step is to determine if hyperkalemia is acute and/or occurred in absence of a large intake of K+—if so, there is an important contribution of a shift of K+ out of cells (see “Consult K-3”). If a shift of K+ is likely, proceed to an analysis of factors that could either destroy cells or decrease the magnitude of the intracellular negative voltage.[38]

The fourth step is to examine the rate of excretion of K+. If this rate is considerably <200 mmol/g creatinine (or <20 mmol K+/mmol creatinine), it is inappropriately low for the presence of hyperkalemia. If so, the basis for the low rate of K+ excretion should be examined by analyzing the two components of the K+ excretion formula in terms of events in the terminal CCD. The flow rate in the terminal CCD should be assessed when vasopressin is acting (Uosm> Posm) by calculating the osmole excretion rate (see Fig. 24-9 ). If the [K+]CCD is low, seek the basis for a slower rate of electrogenic reabsorption of Na+ in the CCD. In patients with few remaining nephrons, the flow rate is so rapid per nephron that this may limit K+ secretion.

Slower Na+ Type of Lesion

These patients will have a low effective ECF volume, their UNa and UCl will not be very low, and their Prenin will be high. The basis could be a low PAldosterone if there is a rise in the [K+]CCD 2 hours after the administration of 100 mg of fludrocortisone in an adult; in addition, the UNa and UCl should fall to very low values. This diagnosis can be confirmed by finding a low PAldosterone. On the other hand, if the patient did not respond to exogenous mineralocorticoids, the presumptive diagnosis would be that his slower Na+ lesion would be due to blockade of the aldosterone receptor in principal cells (e.g., by drugs such as spironolactone), an aldosterone receptor problem (e.g., inherited disorders), or a reduced activity of the epithelial Na+ channel (ENaC) in principal cells (e.g., inherited disorder or blockade by cationic drugs such as amiloride, triamterene, or trimethoprim). If the Prenin is low in the presence of a low ECF volume, suspect that there is a defect in renin release from the juxtaglomerular apparatus.

Faster Cl- -Type of Lesion

The ECF volume in these patients tends to be expanded. Hypertension may be present if their blood pressure is more sensitive to blood volume than usual. The Prenin is suppressed and the PAldosterone is low considering that hyperkalemia is present. This type of lesion could be due to an increased activity of the thiazide sensitive NaCl cotransporter in the distal convoluted tubule (e.g., Gordon syndrome or type II pseudohypoaldosteronism due to a mutation involving WNK-1 or WNK-4 kinases[45]). The decreased delivery of Na+ and Cl- to the CCD, together with a diminished activity of ENaC (because of low aldosterone level in response to expansion of the ECF volume), lead to a rate of reabsorption of Cl- that cannot be less than that of Na+ in the CCD. These patients are expected to have a have a rise in [K+]CCD with the administration of a thiazide diuretic. A faster Cl- type of lesion may result from increased permeability for Cl- in CCD (e.g., the putative cause of hyperkalemia due to cyclosporin). A rise in the [K+]CCD when there is bicarbonaturia suggests the diagnosis of a Cl- shunt disorder.[46]

Metabolic Alkalosis

Metabolic alkalosis is a process that leads to a rise in the PHCO3 and the plasma pH. It is important to recognize that metabolic alkalosis is a diagnostic category with many different causes. Notwithstanding, metabolic alkalosis is included in this section on K+ disorders because a deficiency of KCl plays an important role in its pathophysiology.[47] The following fundamental concepts are central to our understanding of why metabolic alkalosis develops. They also provide the basis for our clinical approach to this diagnostic category, and in the design of optimal therapy.

Concept M Alk-1

Because concentration terms have numerators and denominators, there are two ways to raise the PHCO3: add more HCO3- or reduce the ECF volume ( Fig. 24-16 ).

000903

000519

FIGURE 24-16  Basis for a high concentration of HCO3- in the ECF compartment. The rectangle represents the ECF compartment. The concentration of HCO3- is the ratio of the content of HCO3- in the ECF compartment (numerator) and the ECF volume (denominator). The major causes for a rise in the content of HCO3- in the ECF compartment are a deficit of HCl or a deficit of KCl (upper portion of the figure). The major cause for a fall in the ECF volume is a deficit of NaCl. An intake of NaHCO3 (or Na+ with potential alkali/bicarbonate) is not sufficient on its own to cause a sustained increase in the content of HCO3- in the ECF compartment, except if there also is a mechanism to diminish the renal excretion of NaHCO3 (double bold lines on the left portion of the figure indicate the reduced renal output of NaHCO3). As shown in Figure 24-17 , a contracted ECF volume leads to the net addition of HCO3- to the ECF compartment.

000519

 

When metabolic alkalosis is caused by a deficit of NaCl, the content of HCO3- in the ECF compartment is not elevated until the venous PCO2 rises ( Fig. 24-17 ).[21]

000524

000519

FIGURE 24-17  Addition of HCO3- to the ECF compartment when the venous PCO2 is high. The oval represents a cell with its HCO3- and protein buffer systems. When the ECF volume is very contracted, rate of blood flow is reduced and more O2 is extracted from each liter of blood flowing through the capillaries; this raises both the capillary blood and ICF PCO2 (left portion of the figure). The higher PCO2 in these cells drives the synthesis of H+ and HCO3 ions; the H+ bind to intracellular proteins while the HCO3- ions are exported to the ECF (right portion of the figure). This process reverses when a large volume of saline is infused and the venous PCO2 declines.

000519

 

Concept M Alk-2

Electroneutrality must be present in every body compartment and in the urine. Therefore balances for Cl--salts must be defined in electroneutral terms (i.e., HCl, KCl, and/or NaCl Fig. 24-18 ). It follows that “Cl-depletion” is not an adequate term to describe the pathophysiology of metabolic alkalosis because it ignores the need for electroneutrality.

000887

000519

FIGURE 24-18  Pathophysiology of metabolic alkalosis due to a deficit of CL-salts. This Flow Chart is useful to understand how a deficit of each type of Cl--containing compound contributes to the development of metabolic alkalosis.

000519

 

Concept M Alk-3

Knowing balances for Na+, K+, and Cl- permits one to deduce why the PHCO3 rose and what changes occurred in the composition of ECF and ICF compartments (see Fig. 24-16 ).

Concept M Alk-4

The kidney plays an important role in the pathophysiology of metabolic alkalosis largely because this organ determines the balance for K+.

NaCl or HCl deficits, which can lead to a higher PHCO3, can also lead to a secondary deficit of KCl and hypokalemia. A deficit of K+ is associated with an acidified PCT cell pH and this can both initiate and sustain a high PHCO3 as a result of renal new HCO3- generation (higher excretion of NH4+), reduced excretion of dietary HCO3- in the form of organic anions,[48] and enhanced reabsorption of HCO3- in the PCT.

Contrary to the widely held impression, there is no renal tubular maximum for the reabsorption of HCO3-.[49] Rather HCO3- ions are retained unless their reabsorption is inhibited (low angiotensin II because of ECF volume expansion and/or an alkaline PCT cell pH[50]). Said a different way, ingesting NaHCO3 will not cause metabolic alkalosis because it expands the “effective” ECF volume, lowers angiotensin II, and raises the ICF pH in PCT cells. Nevertheless, NaHCO3 can be retained when there is a significant decrease in its filtered load due to a large fall in the GFR.[49]

Tools

Quantitative Estimate of the Extracellular Fluid Volume

It is critical to know the ECF volume to determine the content of HCO3- in the ECF compartment and thereby why there was a rise in the PHCO3. We rely on the hematocrit for this purpose (see Table 24-3 ) (or a change in the hematocrit with therapy in the patient who has anemia or erythrocytosis).

Balance Data for Na+, K+, and Cl-

These data are essential to describe deficits in electroneutral terms, but they are rarely available in clinical medicine. Nevertheless, they can be inferred if one knows the new ECF volume and the PNa, PCl, and PHCO3. One cannot know the balances for K+ from these calculations, but one can deduce their rough magnitude by comparing the differences in the content of Na+ versus that of Cl- and HCO3- in the ECF compartment.[51]

Consult Met Alk-1: Metabolic Alkalosis without Diuretics or Vomiting

After a forced 6-hour run in the desert in the heat of the day, this elite corps soldier was the only one in his squad who collapsed. Although he perspired profusely during the run, he had free access to water and glucose-containing fluids. He did not vomit and denied the intake of medications. Physical examination revealed a markedly contracted ECF volume. Initial laboratory data are provided in Table 24-11 .


TABLE 24-11   -- Data for Consult Met Alk-1

PNa

mmol/L

140

pH

 

7.47

PK

mmol/L

2.7

PHCO3

mmol/L

37

PCl

mmol/L

90

Arterial PCO2

mm Hg

47

Hematocrit

 

0.50

 

 

 

 

 

 

Questions

What is the basis for the metabolic alkalosis?

What is the therapy for metabolic alkalosis in this patient?

Discussion

What is the basis for metabolic alkalosis?

Quantitative Estimate of the Extracellular Fluid Volume

Although this patient had an obvious degree of contraction of his ECF volume, the physical examination cannot provide a quantitative dimension for this deficit. To distinguish between HCl, KCl, and NaCl deficits, a quantitative analysis of the degree of contraction of the ECF volume is needed—his hematocrit of 0.50 provides this information (see Table 24-3 ). With a hematocrit of 0.50, his ECF volume decreased by one third, from its normal value of 15 L (as he weighed 80 kg) to ∼10 L; accordingly, he lost 5 L of ECF in that period.

Balance Data for Na+, K+, and Cl-

Deficit of HCl: There was no history of vomiting so this is very unlikely basis for the metabolic alkalosis.

Deficit of KCl: The basis of hypokalemia could be a shift of K+ into cells or a loss of KCl in the urine. To lower the PK to 2.7 mmol/L due to a deficit of KCl, especially in this muscular elite solider, the loss of KCl must be very large. Moreover, it is extremely unlikely that this happened over such a short period of time. Furthermore, even if there was a KCl deficit, it is difficult to attribute the rise in the PHCO3 to the formation of new HCO3- due to the renal effects of hypokalemia (high excretion of NH4+) because the time course is too short.

Deficit of NaCl: The decrease in his ECF volume was ∼5 L. One can now calculate how much this degree of ECF volume contraction would raise his PHCO3 (divide the normal content of HCO3- in his ECF compartment [15 L × 25 mmol/L or 375 mmol] by his new ECF volume of 10 L = 37.5 mmol/L). This value is remarkably close to the observed 37 mmol/L. Therefore the major reason for his metabolic alkalosis is the NaCl deficit (a “contraction” form of metabolic alkalosis).

Balance for Na+: Multiplying his PNa (140 mmol/L) before the race times his normal ECF volume (15 L) yields a Na+ content of ∼2100 mmol in this compartment. After the race, his PNa was 140 mmol/L and his ECF volume was 10 L; hence his ECF Na+ content was now 1400 mmol. Accordingly, his deficit of Na+ is ∼700 mmol.

Balance for Cl-: Multiplying his PCl before the race (103 mmol/L) times his normal ECF volume (15 L) yields a Cl- content of ∼1545 mmol in this compartment. After the race, his PCl was 90 mmol/L and his ECF volume was 10 L; hence his ECF Cl- content was now 900 mmol. Accordingly, his deficit of Cl- is ∼645 mmol, a value that is similar to his deficit of Na+.

Balance for K+: It is possible to have a loss of KCl in sweat in patients with cystic fibrosis with concentrations of K+ as high as 20 mmol/L to 30 mmol/L. Nevertheless, because there was little difference between the deficits of Na+and Cl-, there is only a minor total body deficit of K+. Accordingly, the major mechanism for hypokalemia is likely to be a shift of K+ into cells (due to adrenergic surge and the alkalemia).

Interpretation

The deficits of Na+ and Cl- are similar in his ECF compartment. The next issue is to examine possible routes for a large loss of NaCl in such a short time period. Because both diarrhea and polyuria were not present, the only route for a large NaCl loss is via sweat. To have a high electrolyte concentration in sweat (e.g., ∼70 mmol/L for Na+ and Cl-), the likely underlying lesion would be cystic fibrosis. Moreover, he would need to lose >1 L of sweat per hour. The diagnosis of cystic fibrosis was confirmed later by molecular studies.

What Is the Therapy for Metabolic Alkalosis in this Patient?

Knowing that the basis for the metabolic alkalosis is largely a deficit of NaCl, he will need to receive NaCl as his major treatment; the goal will be to replace the deficit that was calculated earlier, or ∼700 mmol. If he had a reverse degree of hyponatremia (drank a large volume of water), he should be given hypertonic saline, especially because he may still have a large volume of water in his stomach. Thus acute hyponatremia should be added to the list of emergencies in this setting

Integration: Clinical Approach; Metabolic Alkalosis

The initial first step is to establish whether the “effective” ECF volume is contracted ( Fig. 24-19 ). Quantitative information can be obtained from the hematocrit or total protein concentration (or both). This permits one to know the relative importance of deficits of NaCl, HCl, and KCl, in the patient.

000531

000519

FIGURE 24-19  Clinical approach to the patient with metabolic alkalosis. Abbreviations: Ca-SR, calcium sensing receptor in thick ascending limb of the loop of Henle; AE, Cl-/HCO3- anion exchanger.

000519

 

In the patient with a low ECF volume, the aim is to rule out first the two common causes of metabolic alkalosis: vomiting and diuretics. Although this may be evident from the clinical history, some patients may deny them; hence, measuring the urine electrolytes is particularly helpful (see Table 24-5 ). A very low UCl is expected when there is a deficit of HCl and/or NaCl. Nevertheless, the recent intake of diuretics will cause a higher excretion of Na+ and Cl-. Another group of causes for Na+ and Cl- wasting are patients with disorders of Bartter or Gitelman syndromes that lead to inhibition of reabsorption of Na+ and Cl-. The clue here is that these electrolytes will be present in every urine sample and urine tests for diuretics will be negative. A clinical picture that mimics Bartter syndrome may result from binding of cations to the calcium-sensing receptor in the thick ascending limb of the loop of Henle (high PCa, drugs like gentamicin or cisplatinum, and perhaps cationic proteins).

In the patient with a normal or high ECF volume, the basis of the metabolic alkalosis and a deficit of K+ is a group of disorders of high primary mineralocorticoid activity. Many, but not all of these patients, will have hypertension.

If the cause of the metabolic alkalosis were an exogenous input of alkali in a patient with a marked reduction in GFR, the ECF volume may be expanded and the PK will not be low; it is easy to recognize this subgroup.

METABOLIC ACIDOSIS

Metabolic acidosis is a process that leads to a fall in the PHCO3 and the plasma pH. Similar to metabolic alkalosis, metabolic acidosis represents a diagnostic category with many different causes. The risks for the patient and the treatment to be prescribed depend on the underlying disorder that caused the metabolic acidosis, the ill effects due to the H+ load, and possible dangers associated with the anions accompanying these H+.

Our goal is to provide a bedside approach when the patient first seeks medical attention.[52] The initial decisions are to determine if an emergency is present and to anticipate and prevent threats that may develop during therapy (Fig. 24-20 ). As in previous sections, we shall highlight the concepts that provide the underpinning for our approach. After each is defined, the laboratory tests that are needed to better define the problem will be outlined. Illustrative consults are used to emphasize the concepts and the utility of these tools.

000537

000519

FIGURE 24-20  Initial steps in the patient with metabolic acidosis. The first step is to determine threats that are present prior to therapy and anticipate those that may develop during therapy (left side of the Flow Chart). The second step is to assess buffering by the BBS in both the ECF and ICF compartments (right side of the Flow Chart).

000519

 

Concept M Ac-1

The PHCO3 is the ratio of the content of HCO3- in the ECF compartment to the ECF volume.

There are two ways to lower the PHCO3: decrease the content of HCO3- in, or raise the volume of the ECF compartment (see following equation).

[HCO3-]ECF = Content of HCO3- in the ECF compartment/ECF volume

It is important to distinguish between acidemia (lower plasma pH) and acidosis. Metabolic acidosis is a process that leads to a lower PHCO3 and pH in the ECF compartment; however, acidemia (i.e., a low PHCO3 and pH) might not be present if there is a loss of both NaCl and NaHCO3 (i.e., a decrease in the in the ECF volume, a fall in its content of HCO3-, but not in its concentration) (e.g., the patient with cholera and immense losses of diarrhea fluid[21] and certain patients with DKA[53]). To make a diagnosis of metabolic acidosis in this setting, a quantitative estimate of the ECF volume is needed to assess the content of HCO3- in this compartment.

Tools

Quantitative Assessment of the Extracellular Fluid Volume

To provide quantitative information about the ECF volume, use the hematocrit or the concentration of total proteins in plasma. The assumption that must be made is that the patient did not have a preexisting anemia or a low plasma protein concentration.

Concept M Ac-2

H+ must by removed by HCO3- to avoid its binding to intracellular proteins ( Fig. 24-21 ). Because the vast majority of the bicarbonate buffer system (BBS) is located in the interstitial fluid and in the ICF compartment in skeletal muscle, it is essential that the majority of H+ removal take place by the BBS in this location ( Fig. 24-22 ).

000536

000519

FIGURE 24-21  The BBS in cells; an emphasis on integrative physiology. The large oval represents a skeletal muscle cell and its capillary blood supply is represented by the curved cylindrical structure to its right. Because we are highlighting the BBS, the intracellular proteins are de-emphasized in this figure. The normal state is shown on the left and buffering of a H+ load is shown on the right. Notice that a lower arterial PCO2 resulting from hyperventilation favors the diffusion of CO2 from cells into the capillary blood and the venous PCO2 is virtually equal to the PCO2 in cells. Hence new H+ will be forced to react with HCO3- in cells because the PCO2 in cells declines. Of even greater importance, the concentration of H+ in cells did not rise appreciably and hence very few H+ bind to proteins in cells.

000519

 

000541

000519

FIGURE 24-22  Buffering of H+ by the BBS in vital organs in a patient with a contracted ECF volume. Skeletal muscle cells are shown to the left and brain cells to the right. When the ECF volume is low (below the horizontal dashed line), the PCO2 in the venous blood draining muscle is high. This causes a high PCO2 in muscle cells, which prevents H+ from being buffered by its intracellular HCO3-. As a result, the concentration of H+ in plasma rises and more H+ will be buffered in brain cells. The latter have their usual BBS because the PCO2 in venous blood draining the brain will change minimally with all but a very severe degree of ECF volume contraction as the cerebral blood flow rate is autoregulated. Because brain is 1/20 the size of muscle, it has far less HCO3- and hence H+ will bind to proteins in brain cells. In contrast, when intravenous saline is administered, blood flow to skeletal muscle rises, its venous PCO2 falls, and more H+ are removed by its BBS. As a result, the concentration of H+ in the ECF compartment falls and H+ are released from proteins in brain cells.

000519

 

If buffering by the BBS were compromised in skeletal muscle, many more H+ would be forced to bind to proteins in cells in vital organs (e.g., brain cells). If this occurred, these proteins would have a more positive charge (H·PTN+), a change in their shape, and ultimately a diminution in their essential functions[54] ( Fig. 24-23 ).

000539

000519

FIGURE 24-23  Failure of the BBS in a patient with metabolic acidosis. When faced with a H+ load, failure of buffering in cells occurs when H+ cannot be removed by reacting with HCO3-. Failure of the BBS to remove a H+ load occurs when the tissue PCO2 cannot fall because the arterial PCO2 is not low enough (40 versus 24 mm Hg in this patient with metabolic acidosis, left of the dashed line) and/or the blood flow rate is low relative to the rate of CO2 production (right of the dashed line). This rise in tissue PCO2 increases the concentration of H+ in cells, pushing H+ to bind to intracellular proteins (H·PTN+).

000519

 

Tools to Assess the Removal of H+ by the Bicarbonate Buffer System

Arterial PCO2

To remove H+ by the BBS in the ECF compartment, the arterial PCO2 must be low as “expected” for the degree of acidemia. Nevertheless, although the arterial PCO2 sets a lower limit on the possible value for the PCO2 in cells, it is not a reliable indicator of the actual value of the intracellular PCO2.

H+ + HCO3- ➙ CO2 + H2O (“pulled” by a low [CO2])

Venous PCO2

To remove H+ by the BBS in the ICF compartment, the PCO2 must be low in cells (see Fig. 24-21 ). The PCO2 in cells is almost equal to the PCO2 in capillaries draining individual organs. Because CO2 is not added after the capillary, the venous PCO2 will reflect the capillary PCO2. There is, however, one caveat when using the venous PCO2 to reflect PCO2 in cells. If an appreciable quantity of blood shunts from the arterial to the venous circulation without passing through capillaries, the venous PCO2 will not reflect the PCO2 in cells.

The venous PCO2 may be considerably higher than the arterial PCO2 when a larger quantity of CO2 is produced and/or there is a reduced blood flow to individual organs.[54] Venous PCO2 measured in the brachial vein reflects the PCO2 in skeletal muscle, the site where most of the buffering of H+ should occur by the BBS. If this PCO2 is high, there is failure of the BBS in muscle, which indicates that more of the H+ load was buffered in vital organs (e.g., the brain). At the usual blood flow rate, the brachial venous PCO2 is ∼46 mm Hg when the arterial PCO2 is 40 mm Hg, but much higher venous PCO2 values were seen in patients with DKA and a very contracted ECF volume.[53] This tool can also be used during therapy; the high venous PCO2 will fall appreciably when sufficient saline has been infused.

Consult M Ac-1: Hyperglycemia without Obvious Ketoacidosis

A 16-year-old, 50 kg, female had several past admissions for DKA because she failed to take her insulin on a regular basis. Her present illness began gradually. In response to thirst, she drank predominantly large volumes of fruit juice and she voided frequently. On physical examination, her ECF volume was obviously contracted, but there was no odor of acetone on her breath and she was not breathing deeply or rapidly. She was easily roused and answered questions appropriately, but she seemed to be somewhat less alert than usual. The remainder of the physical examination was unremarkable. The pH and PCO2 are from an arterial blood sample whereas the remainder of the data prior to therapy is from blood drawn from an antecubital vein; the venous PCO2 was 69 mm Hg ( Table 24-12 ).


TABLE 24-12   -- Data for Consult MAC-1

Glucose

mg/dL

900

Glucose

mmol/L

50

Na+

mmol/L

120

K+

mmol/L

5.5

Cl-

mmol/L

80

HCO3-

mmol/L

24

pH

 

7.40

PCO2

mm Hg

40

Albumin

g/dL

5.1

Anion gap

mEq/L

16

Creatinine

mg/dL

2.0

Hematocrit

 

0.50

Creatinine

μmol/L

230

 

 

 

 

 

 

Questions

Does this patient have metabolic acidosis?

Is her ability to buffer H+ via BBS in skeletal muscle cells compromised?

Discussion

Does This Patient Have Metabolic Acidosis?

Her arterial pH, PHCO3, and PCO2 were in the normal range. However, the modestly elevated value for her anion gap in plasma (PAnion gap) suggests that added acids were present and that she indeed had metabolic acidosis (see Fig. 24-20 ). Because her ECF volume is contracted, the quantity of HCO3- in her ECF compartment must be calculated. Using the hematocrit of 0.50 for this purpose, her plasma volume was 2 L instead of the normal value of 3 L—hence her ECF volume was reduced by ∼33% (6.7 L instead of 10 L). This marked reduction of ECF volume was due to the urinary loss of NaCl during the osmotic diuresis. This low ECF volume and normal PHCO3 indicate that she had a significant deficit of HCO3- in her ECF compartment (6.7 L × 24 mmol/L = 160 mmol HCO3-) ( Table 24-13 ).


TABLE 24-13   -- Balance of HCO3- in Her Extracellular Fluid Compartment

 

ECF volume (L)

HCO3-

Ketoacid anions

 

 

(mmol/L)

mmol

(mmol/L)

mmol

 

10

24

240

0

0

 

6.7

24

160

1

7

Balance

-3.3

 

-80

 

+7

 

For details, see text.

ECF, extracellular fluid.

 

 

 

Upon reflection, the modestly elevated value for her PAnion gap was due in large part to a high concentration of albumin in plasma (PAlbumin), which represents the contracted ECF volume, and not the addition of new acids (confirmed later because the concentrations of ketoacid anions, L-lactate, and D-lactate anions in plasma were not appreciably elevated). Thus she does have metabolic acidosis due to a deficit of NaHCO3. This deficit represents an indirect loss of NaHCO3 caused by the excretion of ketoacid anions along with Na+ in the urine because the rate of excretion of NH4+ is not high early in the course of DKA ( Fig. 24-24 ).

000546

000519

FIGURE 24-24  Ketoacidosis without high PKETOACIDS early in DKA. In the early stage of DKA, metabolic acidosis develops because ketoacids are produced (step 1), their H+ titrate some of the HCO3- in the ECF compartment (larger rectangle), and the resultant CO2 is exhaled (step 2). Because some of the filtered ketoacid anions (A-) are not reabsorbed, they will be excreted, but they are not excreted with an equal quantity of NH4+ because the renal production of NH4+ has not yet increased. Hence these ketoacid anions are excreted largely with Na+ derived from the ECF compartment (step 3). Overall, there is a net loss of NaHCO3 from the ECF compartment.

000519

 

Is Her Ability to Buffer H+ Via Bicarbonate Buffer System in Skeletal Muscle Cells Compromised?

Because her brachial venous PCO2 was 69 mm Hg, buffering of H+ by her BBS in the interstitial fluid and in muscle cells was compromised. Hence there would be more H+ binding to intracellular protein in vital organs (e.g., brain and heart); we call this a “tissue form” of respiratory acidosis.[21] One of the results of this process is that there is a net addition of HCO3- to the ECF compartment (see Fig. 24-17 ).

The venous PCO2 should fall once tissue perfusion improves. As a clinical guide, enough saline should be given to lower venous PCO2 to a value less than 10 mm Hg higher than the arterial PCO2.

Concept M Ac-3

When added acids are the cause of metabolic acidosis, one can detect the addition of H+ by finding a fall in the PHCO3 along with the appearance of new anions. These new anions may remain in the body or be excreted (e.g., in the urine or diarrhea fluid).

Tools

Detect New Anions in Plasma

The accumulation of new anions in plasma can be detected from the calculation of the PAnion gap ( Fig. 24-25 ). When using this calculation, one must adjust the PAnion gap for the concentration of the major unmeasured anion in plasma, albumin (PAlbumin). As a rough estimate, the baseline value for the PAnion gap rises (or falls) by 3 mEq/L to 4 mEq/L for every 10 g/L or 1 g/dL rise (or fall) in the PAlbumin.

000544

000519

FIGURE 24-25  Assessment of the anion gap in plasma. The normal values are shown in the left portion of the figure; the PAnion gap is the shaded area between the cation (left) and anion (right) columns. When L-lactic acid is added (middle portion of the figure), the PHCO3 will fall, and the HCO3- will be replaced with L-lactate anions such that the rise in the PAnion gap equals the fall in the PHCO3. A loss of NaHCO3 is depicted in the right portion of the figure. Note that the PHCO3 fell and the PCl was elevated, but no new anions were added.

000519

 

Another approach to detect new anions in plasma was recommended by Stewart[55]; it is called the strong ion difference (SID). This approach is rather complex and offers only a minor advantage over PAnion gap in that it includes a correction for the net negative charge on PAlbumin. It suffers from the same limitations as the PAnion gap in that it relies only on concentrations in plasma and it does not include information from the venous PCO2 or the hematocrit. There are, however, additional weaknesses when using the SID approach in that it leads to misunderstandings concerning interpretations of acid-base physiology.[56]

Detect New Anions in the Urine

These anions can be detected with the urine net charge using a value for the concentration of NH4+ in the urine (UNH4) that is estimated from the urine osmolal gap (Uosm gap) as discussed in the next section (see following equation). The nature of these new anions may sometimes be deduced by comparing their filtered load to their excretion rate. For example, when a very large proportion of new anions are excreted, suspect that this anion was secreted in the proximal convoluted tubule (PCT)[57] (see “Consult M Ac-2”) or freely filtered and poorly reabsorbed by the PCT (e.g., reabsorption of ketoacid anions is inhibited by acetosalycilate anions). On the other hand, a very low excretion of new anions suggests that they were avidly reabsorbed in the PCT (e.g., L-lactate).

New urine anions = (UNa + UK + UNH4) - UCl

Concept M Ac-4

The identification of new anions helps to predict important dangers for your patient. Examples include anions such as citrate that chelate ionized calcium in plasma,[58] anions that are excreted at a high rate and cause a very high rate of excretion of Na+ and K+ (see “Consult M Ac-2”), and those that suggest that toxins were produced during the metabo-lism of unusual alcohols (e.g., methanol, ethylene glycol metabolism).

Tools

Detect Toxic Alcohols

The presence of alcohols in plasma can be detected by finding a large increase in plasma osmolal gap (Posm gap) (see following equation). This occurs because the compound is uncharged, has a low molecular weight, and because large quantities have been ingested.

Posm gap = Measured Posm - (2 PNa + PGlucose + PUrea), all in mmol/L terms

Concept M Ac-5

The expected renal response to chronic metabolic acidosis is a high rate of excretion of NH4+.

In a patient with chronic metabolic acidosis, the expected rate of excretion of NH4+ should be >200 mmol/day.[59] We stress the term “chronic” because there is a lag period of a few days before high rates of excretion of NH4+ can be achieved.

Tools: Detect Urinary Ammonium

Urine Osmolal Gap

Because the Uosm gap detects all NH4+ salts in the urine, it provides the best indirect estimate of the UNH4, and hence, we no longer use the urine net charge (or urine anion gap) for this purpose (see following equation).[60] The premise of the test is that NH4+ is detected by its contribution to the Uosm ( Fig. 24-26 ).

000551

000519

FIGURE 24-26  Use of the urine osmolal gap to reflect the concentration of NH4+ in the urine. Because most urine anions are monovalent, the UNH4 is equal to one half the Uosmolal gap.

000519

 

Uosm gap = Measured Uosm - calculated Uosm Calculated Uosm = 2 (UNa + UK) + UUrea + UGlucose, all in mmol/L terms Urine NH4+ = Uosm gap/2

We use the UNH4/UCreatinine ratio in a spot urine sample to assess the rate of excretion of NH4+. The rationale is that the rate of excretion of creatinine is relatively constant over the 24-hour period in complete, timed urine collections.[27] In a patient with chronic metabolic acidosis the expected renal response is UNH4/UCreatinine ratio >150 mmol/g creatinine (>15 in mmol/mmol terms).

Consult M Ac-2: Metabolic Acidosis Due to Glue Sniffing

A 28-year-old male sniffs glue on a regular basis. He developed profound weakness over the course of 3 days. On physical examination, his ECF volume was obviously contracted. His pH and PCO2 were from arterial blood and the other data were from venous blood; his venous PCO2 was 70 mm Hg, PGlucose was 3.5 mmol/L (63 mg/dL), and his PAlbumin was 4.5 g/dL (45 g/L) ( Table 24-14 ).


TABLE 24-14   -- Data for Consult MAC-2

 

 

Blood

Urine

 

 

Blood

Urine

pH

 

7.30

6.0

PCO2

mm Hg

30

HCO3-

mmol/L

15

<5

K+

mmol/L

2.3

20

Na+

mmol/L

120

50

Cl-

mmol/L

90

0

Creatinine

mg/dL

1.7

3.0

Urea

mmol/L

2.5

50

Osmolality

mOsm/L

245

500

Hematocrit

 

0.50

 

 

 

Questions

What is the basis for the metabolic acidosis?

What dangers are implied from the high rate of excretion of anions in the urine?

Discussion

What Is the Basis for the Metabolic Acidosis?

Metabolic acidosis is present because of the low pH and PHCO3. He also has a very low content of HCO3- in his ECF compartment because his PHCO3 is low and his ECF volume is contracted (hematocrit 0.50).

Detect new anions in plasma: Because there was no increase in the PAnion gap corrected for the PAlbumin, this suggested that the metabolic acidosis was not due to a gain of acids.

Detect new anions in the urine: For this, the UNH4 must be estimated using the Uosm gap. The measured Uosm was 500 mOsm/L and the calculated Uosm was 190 mOsm/L (2 × (UNa 50 + UK 20) + Uurea 50 mmol/L). Hence the Uosmgap was 310 mOsm/L and the UNH4 was half of this value or 155 mmol/L (confirmed later by direct measurement). Therefore the patient had a high UNH4 and many unmeasured anions in the urine. Hence we deduced that there was overproduction of acids while the kidney excreted their accompanying anions. The high rate of excretion of the anions suggested that they were not only filtered but also secreted by PCT—this is the fate of hippurate anions.[61]Hippuric acid is the end product of the metabolism of toluene, a constituent of glue—hence the basis of the metabolic acidosis is overproduction of hippuric acid together with excretion of hippurate anions in the urine at a rate that exceeded the rate of excretion of NH4+.[57]

What Dangers Are Implied from the High Rate of Excretion of Anions in the Urine?

Hippurate is secreted by the PCT and it is excreted very rapidly. When the excretion of hippurate exceeds that of NH4+, this obligates the excretion of Na+ and K+, which may cause ECF volume contraction and hypokalemia (seeFig. 24-23 ). The presence of ECF volume contraction lowers the rate of excretion of NH4+ if it also leads to a low GFR.[62]

The higher rate of excretion of K+ can lead to a cardiac arrhythmia and hypoventilation if it causes respiratory muscle weakness. Hypoventilation caused by muscle weakness, along with the low ECF volume, can lead to hypokalemia and, as a result, a high venous PCO2 with diminished buffering of H+ by the BBS in skeletal muscles and hence buffering of more H+ in vital organs (e.g., the brain and the heart) (see Fig. 24-22 ).

Concept M Ac-6

A low rate of excretion of NH4+ could be due to a decreased medullary NH3 or a decreased net H+ secretion in distal nephron.

A low rate of ammoniagenesis could be due to due to an alkaline PCT cell (e.g., hyperkalemia, genetic, or acquired disorders compromising proximal H+ secretion) or decreased availability of ADP reflecting less work performed in PCT cells because of a low filtered load of Na+ due to a reduced GFR.[62] Another cause for a low rate of excretion of NH4+ is a low net secretion of H+ in the distal nephron. This could be due to a H+-ATPase defect (e.g., autoimmune and hypergammaglobulinemic disorders including Sjögren syndrome), back-leak of H+ (e.g., drugs like amphotericin B), or disorders associated with the distal secretion of HCO3- (e.g., in certain patients with Southeast Asian Ovalocytosis (SAO)[63]).

Tools: Detect Why the Rate of Excretion of NH4+ is Low

Urine pH

The urine pH is not helpful to imply that the rate of excretion of NH4+ is low.[64] For example, at a urine pH of 6.0, the UNH4 can be 20 mmol/L or 200 mmol/L ( Fig. 24-27 ). However, the basis for the low rate of excretion NH4+may be deduced from the urine pH. A urine pH that is ∼5 suggests that the basis for a low rate of excretion of NH4+ is due primarily to a decreased availability of NH3 in the medullary interstitial compartment. On the other hand, a urine pH that is >7 suggests that NH4+ excretion is low because there is a defect in H+ secretion and/or that there was a high rate of excretion of HCO3- in the distal nephron ( Fig. 24-28 ).

000549

000519

FIGURE 24-27  Failure of the urine PH to detect UNH4. As shown on the left, the urine pH is low during acute metabolic acidosis due to enhanced distal H+ secretion. The rate of excretion of NH4+ is only modestly high at this time due to the lag period before the rate of renal production of NH4+ is augmented. In contrast, during chronic metabolic acidosis shown on the right, the rate of renal production of NH4+ is very high and the availability of NH3 in the medullary interstitial compartment provides more NH3 in the lumen of the MCD than H+ secretion in this nephron segment. Therefore note the much higher NH4+ excretion rate at a urine pH of 6.

000519

 

000554

000519

FIGURE 24-28  Basis of metabolic acidosis. One must use a definition of metabolic acidosis based not only the PHCO3 but also on the content of HCO3- in the ECF compartment.

000519

 

Assess Distal H+ Secretion

H+ secretion in the distal nephron can be evaluated using the PCO2 in alkaline urine (UPCO2) ( Fig. 24-29 ). A UPCO2 that is ∼70 mm Hg in a second-voided alkaline urine implies that H+ secretion in the distal nephron is likely to be normal whereas much lower UPCO2 values suggest that distal H+ secretion is impaired.[65] In patients with low net distal H+ secretion, the UPCO2 can be high if there is a lesion causing a back-leak of H+ from the lumen of the collecting ducts (e.g., use of amphotericin B[66]) or distal secretion of HCO3- as in some patients with SAO who also have a second mutation in their HCO3-/Cl- anion exchanger that leads to mis-targeting of the exchanger to the luminal membrane of the α-intercalated cells. [54] [63] A caveat with this test is that the UPCO2 is also influenced by the renal concentrating ability.[67]

000542

000519

FIGURE 24-29  The basis for an increased PCO2 in alkaline urine. When NaHCO3 is given, there is a large delivery of HCO3 to the distal nephron, which makes HCO3- virtually the only H+ acceptor in its lumen. Because there is no luminal carbonic anhydrase (CA), the H2CO3 formed will be delivered downstream and form CO2 plus water. Thus a urine PCO2 that is appreciably higher than the plasma PCO2 provides evidence for distal H+ secretion. The PCO2 in alkaline urine will be low if there is a lesion involving the H+-ATPase or causes an alkaline cell pH in intercalated cells in the distal nephron. If the lesion is one that causes back-leak of H+ or distal secretion of HCO3-, the urine PCO2 will be high.

000519

 

Tools to Assess Proximal Cell pH

Fractional excretion of HCO3-: In patients with metabolic acidosis associated with a low capacity to reabsorb filtered HCO3- (e.g., disorders with defects in H+ secretion in the PCT called proximal RTA), some would measure the fractional excretion of HCO3- after infusing NaHCO3 to confirm this diagnosis. This is rarely needed in our opinion—often the results are far from clear (e.g., in a patient with an abnormal ECF volume or PK) and, in addition, the test can impose a danger (e.g., in a patient with a low PK). These patients will be detected clinically by failure to correct their metabolic acidosis despite being given large amounts of NaHCO3.

Rate of citrate excretion: The rate of excretion of citrate is a marker of pH in cells of the PCT.[68] The daily rate of excretion of citrate in children and adults consuming their usual diet is ∼400 mg (∼2 mmol/g or 10 mmol of creatinine). Although the rate of excretion of citrate is very low during most forms of metabolic acidosis,[69] a notable exception is in patients with disorders causing an alkaline PCT cell pH.[70]

Consult M Ac-3: Determine the Cause of Hyperchloremic Metabolic Acidosis

A 23-year-old female suffers from Southeast Asian Ovalocytosis and was referred for assessment of hypokalemia. Her physical examination was unremarkable. Her laboratory results in plasma and a spot urine sample are summarized in Table 24-15 . Her urine urea was 220 mmol/L and the urine was glucose-free.


TABLE 24-15   -- Data for Consult Mac-3

 

 

Plasma

Urine

 

 

Plasma

Urine

pH

 

7.35

6.8

PCO2

mm Hg

30

HCO3-

mmol/L

15

10

K+

mmol/L

3.1

35

Na+

mmol/L

140

75

Cl-

mmol/L

113

95

Anion gap

mEq/L

12

5

Creatinine

mg/dL

0.7

60

Osmolality

mOsm/L

290

450

Citrate

mg/dL

Low

 

 

 

Questions

What is the basis of the hyperchloremic metabolic acidosis?

What is the basis for the low rate of excretion of NH4+?

Discussion

What Is the Basis of the Hyperchloremic Metabolic Acidosis?

Urine osmolar gap: The patient had a low UNH4 because her measured Uosm (450 mOsm/kg H2O) was very similar to her calculated Uosm [420 mOsm/kg H2O, i.e., 2 (UNa 75 + UK 35 mmol/L) + UUrea (220 mOsm/kg H2O) + UGlucose (0)]. In addition, her rate of excretion of NH4+ was low because the ratio of UNH4/UCreatinine was very low. Because her GFR was not very low, the diagnosis is renal tubular acidosis (RTA).

What Is the Cause for the Low Rate of Excretion of NH4+ ?

Urine pH: Because the urine pH is 6.8, the basis for the low rate of NH4+ excretion is a low net secretion of H+ in the distal nephron ( Fig. 24-30 ).

000558

000519

FIGURE 24-30  Approach to the patient with HCMA and a low excretion of NH4+. The initial approach to determine the pathophysiology in patients with HCMA and a low rate of excretion of NH4+ is based on the urine pH.

000519

 

Assess distal H+ secretion: After hypokalemia was corrected, the PCO2 in alkaline urine was 70 mm Hg.

Assess proximal cell pH: The rate of excretion of citrate was low. In addition, her PHCO3 remained in the normal range after the infusion of NaHCO3.

Interpretation

The high urine pH and the low rate of excretion of NH4+ suggested that she had a defect in distal H+ secretion. H+ secretion by her PCT seemed to be intact (her PHCO3 remained in the normal range after initial therapy, Ucitrate was low, and her FEHCO3 was <3%). Because her UPCO2 was unexpectedly high and a back leak of H+ type of defect is unlikely, perhaps her mutant Cl-/HCO3- anion exchanger was targeted abnormally to the luminal membrane of α-intercalated cells. The UPCO2 would be high due to distal secretion of HCO3- by alkaline intercalated cells.

Integration: Clinical Approach to the Patient with Metabolic Acidosis

The first step in our clinical approach to the patient with metabolic acidosis is to establish that metabolic acidosis is present by finding either a low pH and PHCO3 or a low content of HCO3- in the ECF compartment. Our next step is to identify threats present before therapy begins and to anticipate and prevent dangers that may develop during the course of the illness or with its therapy (see Fig. 24-20 ). After the emergencies are considered, the arterial and venous PCO2 should be assessed to examine the effectiveness of the BBS. If the venous PCO2 is too high, the BBS cannot prevent H+ from binding to ICF proteins (see Fig. 24-22 ).

The third step in our approach is to determine the basis of metabolic acidosis (see Fig. 24-28 ). The most important step is to hunt for new unmeasured anions in the plasma and the urine. This is followed by an assessment of the renal response to metabolic acidosis. If the rate of excretion of NH4+ is low (low Uosm gap) while the GFR is not very low, a diagnosis of RTA is suspected. To uncover the cause for the low rate of excretion of NH4+, the urine pH is the most important of the laboratory tests in the urine (see Fig. 24-30 ). The UPCO2 is helpful in the subgroup with a high urine pH and measuring the rate of excretion of citrate is useful to assess the pH in PCT cells.

CONCLUSION

Given the tremendous advances in our understanding of the integrative physiology of the topics covered in this chapter and the many new insights gained from molecular and genetic advances, clinicians should now pursue diagnoses right down to the enzyme or transporter that is defective. A physiology-based clinical approach is a crucial component to make more exact diagnoses and design more appropriate therapy for patients with disorders of water, Na+, K+, and/or acid-base homeostasis.

References

1. Kamel KS, Ethier JH, Richardson RMA, et al: Urine electrolytes and osmolality: When and how to use them.  Am J Nephrol  1990; 10:89-102.

2. Halperin ML, Kamel KS: Use of the composition of the urine at the bedside: Emphasis on physiological principles to provide insights into diagnostic and therapeutic issues.   In: Seldin DW, Giebisch G, ed. Physiology and Pathology of Electrolyte Metabolism. The Kidney: Physiology and Pathophysiology,  3rd ed. New York: Raven Press; 2001.

3. Nielsen S, Frokiaer J, Marples D, et al: Aquaporins in the kidney: From molecules to medicine.  Physiol Rev  2002; 82:205-244.

4. Halperin ML, Bichet DG, Oh MS: Integrative physiology of basal water permeability in the distal nephron: Implications for the syndrome of inappropriate secretion of antidiuretic hormone.  Clin Nephrol  2001; 56:339-345.

5. Shafiee MA, Charest AF, Cheema-Dhadli S, et al: Defining conditions that lead to the retention of water: The importance of the arterial sodium concentration.  Kidney Int  2005; 67:613-621.

6. Oh MS, Carroll HJ, Roy A, et al: Chronic hyponatremia in the absence of ADH: Possible role of decreased delivery of filtrate.  J Am Soc Nephrol  1997; 8:108A.

7. Cherney DZI, Davids MR, Halperin ML: Acute hyponatraemia and MDMA (“Ecstasy”): Insights from a quantitative and integrative analysis.  Quart J Med  2002; 95:475-483.

8. Laureno R, Karp BI: Pontine and extrapontine myelinolysis following rapid correction of hyponatremia.  Lancet  1988; 1:1439-1441.

9. Kamel KS, Bichet DG, Halperin ML: Studies to clarify the pathophysiology of partial central diabetes insipidus.  Am J Nephrol  2001; 37:1290-1293.

10. Spira A, Gowrishankar M, Halperin ML: Factors contributing to the degree of polyuria in a patient with diabetes mellitus in poor control.  Am J Kidney Dis  1997; 30:829-835.

11. Deetjen P, Baeyer HV, Drexel H: Renal glucose transport.   In: Seldin D, Giebisch G, ed. The Kidney: Physiology and Pathophysiology,  2nd ed. New York: Raven Press; 1992:2873-2888.

12. Davids MR, Edoute Y, Stock S, Halperin ML: Severe degree of hyperglycemia: Novel insights revealed by the use of simple principles of integrative physiology.  Quart J Med  2002; 95:113-124.

13. Carlotti APCP, Bohn D, Halperin ML: Importance of timing of risk factors for cerebral oedema during therapy for diabetic ketoacidosis.  Arch Dis Child  2003; 88:170-173.

14. Gowrishankar M, Lenga I, Cheung RY, et al: Minimum urine flow rate during water deprivation: Importance of the permeability of urea in the inner medulla.  Kidney Int  1998; 53:159-166.

15. Gamble JL, McKhann CF, Butler AM, Tuthill E: An economy of water in renal function referable to urea.  Am J Physiol  1934; 109:139-154.

16. Halperin ML, Rolleston FS: Clinical Detective Stories: A Problem-Based Approach to Clinical Cases in Energy and Acid-Base Metabolism,  1st ed. London, England, Portland Press, 1993.

17. Labow BI, Souba WW: Glutamine.  World J Surg  2000; 24:1503-1513.

18. Carlotti APCP, Bohn D, Mallie J-P, Halperin ML: Tonicity balance and not electrolyte-free water calculations more accurately guide therapy for acute changes in natremia.  Intensive Care Med  2001; 27:921-924.

19. Carlotti APCP, Bohn D, Rutka JT, et al: A method to estimate urinary electrolyte excretion in patients at risk for developing cerebral salt wasting.  J Neurosurg  2001; 95:420-424.

20. Mitch WE, Collier VU, Walser M: Creatinine metabolism in chronic renal failure.  Clin Sci  1980; 58:327-335.

21. Zalunardo N, Lemaire M, Davids MR, Halperin ML: Acidosis in a patient with cholera: A need to redefine concepts.  Quart J Med  2004; 97:681-696.

22. Russell JM: Sodium-potassium-chloride cotransport.  Physiol Rev  2000; 80:211-276.

23. Clausen T: Regulation of active Na+-K+ transport in skeletal muscle.  Physiol Rev  1986; 66:542-580.

24. Solemani M, Burham C: Physiology and molecular aspects of the Na+:HCO3- contransporter in health and disease processes.  Kidney Int  2000; 57:371-384.

25. Huth EJ, Squires RD, Elkinton JR: Experimental potassium depletion in normal human subjects. II. Renal and hormonal factors in the development of extracellular alkalosis during depletion.  J Clin Invest  1959; 38:1149-1165.

26. Talbott JH, Schwab RS: Recent advances in the biochemistry and therapeutics of potassium salts.  N Engl J Med  1940; 222:585-590.

27. Cockcroft DW, Gault MH: Prediction of creatinine clearance from serum creatinine.  Nephron  1976; 16:31-41.

28. Steele A, deVeber H, Quaggin SE, et al: What is responsible for the diurnal variation in potassium excretion?.  Am J Physiol  1994; 36:R554-R560.

29. Halperin ML, Kamel KS: Potassium.  Lancet  1998; 352:135-142.

30. Field MJ, Giebisch GJ: Hormonal control of renal potassium excretion.  Kidney Int  1985; 27:379-387.

31. Ethier JH, Kamel KS, Magner PO, et al: The transtubular potassium concentration in patients with hypokalemia and hyperkalemia.  Am J Kidney Dis  1990; 15(4):309-315.

32. Cheema-Dhadli S, Lin S-H, Chong CK, et al: Requirements for a high rate of potassium excretion in rats consuming a low electrolyte diet.  J Physiol London  2006; 572(2):493-501.

33. Lin S-H, Lin Y-F, Halperin ML: Hypokalemia and paralysis: Clues on admission to help in the differential diagnosis.  Quart J Med  2001; 94:133-139.

34. Alazami M, Lin S-H, Chu C-J, et al: Unusual causes of hypokalaemia and paralysis.  Quart J Med  2006; 99:181-192.

35. Lin S-H, Chu P, Cheng C-J, et al: Early diagnosis of thyrotoxic periodic paralysis: Spot urine calcium to phosphate ratio.  Crit Care Med  2006; 34:2984-2989.

36. Lin SH, Lin YF: Propranolol rapidly reverses paralysis, hypokalemia and hypophosphatemia in thyrotoxic periodic paralysis.  Am J Kidney Dis  2001; 37:620-624.

37. Edwards CRW: Lessons from licorice.  N Engl J Med  1991; 24:1242-1243.

38. Rosa RM, Williams ME, Epstein FH: Extrarenal potassium metabolism.   In: Seldin DW, Giebisch G, ed. The Kidney: Physiology and Pathophysiology,  2nd ed. New York: Raven Press; 1992:2165-2190.

39. Porte DJ: Sympathetic regulation of insulin secretion.  Arch Intern Med  1969; 123:252-260.

40. Choi MJ, Fernandez PC, Patnaik A, et al: Trimethoprim induced hyperkalemia in a patient with AIDS.  N Engl J Med  1993; 328:703-706.

41. Schreiber MS, Chen C-B, Lessan-Pezeshki M, et al: Antikaliuretic action of trimethoprim is minimized by raising urine pH.  Kidney Int  1996; 49:82-87.

42. Carlisle EJF, Donnelly SM, Ethier J, et al: Modulation of the secretion of potassium by accompanying anions in humans.  Kidney Int  1991; 39:1206-1212.

43. Iolascon A, Stewart GW, Ajetunmobi JF, et al: Familial pseudohyperkalemia maps to the same locus as dehydrated hereditary stomatocytosis (hereditary xerocytosis).  Blood  1999; 93:3120-3123.

44. Don BR, Sebastian A, Cheitlin M, et al: Pseudohyperkalemia caused by fist clenching during phlebotomy.  N Engl J Med  1990; 322(18):1290-1292.

45. Kahle KT, Wilson FH, Lalioti M, et al: WNK kinases: Molecular regulators of integrated epithelial ion transport.  Curr Opin Nephrol Hypertens  2004;557-562.

46. Kamel K, Ethier JH, Quaggin S, et al: Studies to determine the basis for hyperkalemia in recipients of a renal transplant who are treated with cyclosporin.  J Am Soc Nephrol  1992; 2:1279-1284.

47. Scheich A, Donnelly S, Cheema-Dhadli S, et al: Does saline “correct” the abnormal mass balance in metabolic alkalosis associated with chloride-depletion in the rat.  Clin Invest Med  1994; 17:448-460.

48. Cheema-Dhadli S, Lin S-H, Halperin ML: Mechanisms used to dispose of a progressively increasing alkali load in the rat.  Am J Physiol  2002; 282:F1049-F1055.

49. Rubin SI, Sonnenberg B, Zettle R, Halperin ML: Metabolic alkalosis mimicking the acute sequestration of HCl in rats: Bucking the alkaline tide.  Clin Invest Med  1994; 17:515-521.

50. Cogan M: Angiotensin II: A powerful controller of sodium transport in the early proximal tubule.  Hypertension  1990; 15:451-458.

51. Shafiee MA, Napalova O, Charest AF, Halperin ML: When is sodium chloride the physiologically appropriate treatment for patients with metabolic alkalosis.  Indian J Nephrol  2003; 13:45-54.

52. Kamel KS, Halperin ML: An improved approach to the patient with metabolic acidosis: A need for four amendments.  J Nephrol  2006; 19:S76-S85.

53. Napolova O, Urbach S, Davids MR, Halperin ML: How to assess the degree of extracellular fluid volume contraction in a patient with a severe degree of hyperglycemia.  Nephrol Dial Trans  2003; 18:2674-2677.

54. Vasuvattakul S, Warner LC, Halperin ML: Quantitative role of the intracellular bicarbonate buffer system in response to an acute acid load.  Am J Physiol  1992; 262:R305-R309.

55. Stewart PA: Modern quantitative acid-base chemistry.  Can J Physiol Pharmacol  1983; 61:1444-1461.

56. Halperin ML, Goldstein MB, Kamel KS: Fluid, Electrolyte and Acid-Base Physiology; A Problem-Based Approach,  Philadelphia, WB Saunders, 2006.

57. Carlisle EJF, Donnelly SM, Vasuvattakul S, et al: Glue-sniffing and distal renal tubular acidosis: Sticking to the facts.  J Am Soc Nephrol  1991; 1:1019-1027.

58. DeMars C, Hollister K, Tomassoni A, et al: Citric acidosis: A life-threatening cause of metabolic acidosis.  Ann Emerg Med  2001; 38:588-591.

59. Halperin ML: How much “new” bicarbonate is formed in the distal nephron in the process of net acid excretion?.  Kidney Int  1989; 35:1277-1281.

60. Dyck RF, Asthana S, Kalra J, et al: A modification of the urine osmolal gap: An improved method for estimating urine ammonium.  Am J Nephrol  1990; 10:359-362.

61. Smith HW, Finfelstein N, Aliminosa L, et al: The renal clearances of substituted hippuric acid derivatives and other aromatic acids in dogs and man.  J Clin Invest  1945; 24:388-404.

62. Halperin ML, Jungas RL, Pichette C, Goldstein MB: A quantitative analysis of renal ammoniagenesis and energy balance: A theoretical approach.  Can J Physiol Pharmacol  1982; 60:1431-1435.

63. Kaitwatcharachai C, Vasuvattakul S, Yenchitsomanus P, et al: Distal renal tubular acidosis in a patient with southeast Asian ovalocytosis: Possible interpretations of a high urine PCO2.  Am J Kidney Dis  1999; 33:1147-1152.

64. Richardson RMA, Halperin ML: The urine pH: A potentially misleading diagnostic test in patients with hyperchloremic metabolic acidosis.  Am J Kidney Dis  1987; 10:140-143.

65. Halperin ML, Goldstein MB, Haig A, et al: Studies on the pathogenesis of type I (distal) renal tubular acidosis as revealed by the urinary PCO2 tensions.  J Clin Invest  1974; 53:669-677.

66. Roscoe J, Goldstein M, Halperin M, et al: Effect of amphotericin B on urine acidification in rats: Implications for the pathogenesis of distal renal tubular acidosis.  J Lab Clin Med  1977; 89:463.

67. Berliner RW, DuBose TDJ: Carbon dioxide tension of alkaline urine.   In: Seldin DW, Giebisch G, ed. The Kidney: Physiology and Pathophysiology,  2nd ed. New York: Raven Press; 1992:2681-2694.

68. Simpson D: Citrate excretion: A window on renal metabolism.  Am J Physiol  1983; 244:F223-F234.

69. Dedmond RE, Wrong O: The excretion of organic anion in renal tubular acidosis with particular reference to citrate.  Clin Sci  1962; 22:19-32.

70. Halperin ML, Kamel KS, Ethier JH, Magner PO: What is the underlying defect in patients with isolated, proximal renal tubular acidosis?.  Am J Nephrol  1989; 9:265-268.