Physiologic Principles in the Clinical Evaluation of Electrolyte, Water, and Acid–Base Disorders

The kidneys play a fundamental role in the regulation of water and electrolyte excretion as well as in the maintenance of acid base balance. Water and electrolyte homeostasis is accomplished almost exclusively through changes in the volume and composition of urine. Routes of external loss other than from the kidneys play a limited role in regulation, even though they may display some response to the needs of the organism. Intake of the various electrolytes and of water can vary extremely widely and almost independently of each other. Balance (used as a synonym for external balance) is the difference between intake and excretion by all routes. Internal balance, on the other hand, means the rate of transfer of a substance between the intra- and extracellular fluids, a concept that has proven especially useful with regard to potassium homeostasis. In this chapter, we will limit the discussion to external balance as we review the physiology underlying the clinical approach to the evaluation of water and electrolyte excretion by the kidneys. The objective of this chapter is to summarize the rationale and physiologic principles behind the use of urinary indices and clinical tools available in the evaluation of normal and pathophysiologic states related to sodium, water, potassium, and acid-base disturbances.

Keywords

polyuric syndromes, potassium disorders, metabolic acidosis

The kidneys play a fundamental role in the regulation of water and electrolyte excretion as well as in the maintenance of acid base balance. Water and electrolyte homeostasis is accomplished almost exclusively through changes in the volume and composition of urine. Routes of external loss other than from the kidneys play a limited role in regulation, even though they may display some response to the needs of the organism. Intake of the various electrolytes and of water can vary extremely widely and almost independently of each other. Balance (used as a synonym for external balance) is the difference between intake and excretion by all routes. Internal balance, on the other hand, means the rate of transfer of a substance between the intra- and extracellular fluids, a concept that has proven especially useful with regard to potassium homeostasis. In this chapter, we will limit the discussion to external balance as we review the physiology underlying the clinical approach to the evaluation of water and electrolyte excretion by the kidneys. No attempt will be made to discuss in-depth specific regulatory mechanisms that are outlined in detail in other chapters. The objective of this chapter is to summarize the rationale and physiologic principles behind the use of urinary indices and clinical tools available in the evaluation of normal and pathophysiologic states related to sodium, water, potassium, and acid–base disturbances.

Concept of External Balance and Steady-State

The kidney varies the rate of excretion of various solutes to match the input from dietary intake and endogenous production. “Zero balance” or “in balance” both refer to a state in which input equals output. Commonly the unit of time is a 24-hour day, and the rate of excretion can only be applied with reasonable certainty to substances that are not metabolized. The term “steady-state” is used to mean a dynamic equilibrium, with the implication that input and output are occurring but are equal. Thus, steady-state, when used with regard to body electrolyte homeostasis refers to zero balance. Humans in a steady-state excrete absorbed ions as well as the metabolic products from their diet. Endogenous production of a substance does not preclude the use of the balance concept. Rather, this production is considered to be a component of intake. For example, water is produced continuously by the metabolism of foodstuffs; this production of water contributes an additional approximate 0.3 L/day to the intake of water from other sources.

The term “steady-state” should not be used to refer to a condition in which both input and output are constant but different. In this case, the body composition is changing progressively, and this clearly represents an unsteady-state. The change in the rate of excretion in response to a deficit or a surplus of an electrolyte may occur promptly or require a lag period. To reach a steady-state, a variable amount of time is needed for each ion. For example, a water load is excreted within hours, whereas a potassium-deficient diet requires a few days for the renal conservation of potassium to be completed. A low-sodium diet leads to a transient negative sodium balance where sodium excretion exceeds sodium intake in the initial adaptation period ( Fig. 74.1 ). This negative sodium balance reduces extracellular fluid volume (ECFV), which then activates the sodium-conserving mechanisms of the kidney. A high-sodium diet, by transiently resulting in sodium retention, expands ECFV and thereby elicits a signal that deactivates the mechanisms that promote tubular reabsorption of sodium. In both situations, a new steady-state is ultimately reached in which sodium intake and output is equivalent. To evaluate the composition of urine in patients, one should first consider their diet. A corollary of this is that there are no “normal” values for the composition of urine, only “expected” ones depending on dietary intake and other stimuli. Because dietary intake fluctuates so greatly (for example, normal dietary intake of sodium may vary from 20 mEq/day to 500 mEq/day), urinary net excretion of sodium varies to match sodium intake. Also, the dietary intake of sodium can be extrapolated from the urine composition of sodium, provided that the patient is in a steady-state. For instance, when a subject is in a steady-state, the finding of 500 mEq of sodium in the urine implies that the subject must have consumed about 500 mEq of sodium in the diet. Likewise, the finding of 20 mEq of sodium in the urine reflects dietary intake of only 20 mEq of sodium. This information is very useful for instance when assessing adherence to dietary instructions in subjects with essential hypertension.

Figure 74.1, In the normal subject, a low-salt diet results in a reduction of extracellular fluid volume (ECFV) (middle panel) from a normal value of 15 to about 13 L. At that point, intake (upper panel) and urinary excretion (lower panel) are equal. The subject is in balance. (Horizontal dotted lines represent the basal levels of ECFV (middle panel) and dietary sodium during severe sodium restriction (lower panel).) In the edematous patient, a low-salt diet results in a reduction in ECFV from 25 to 20 L from an initial value of 25 L. Nevertheless, ECFV is still above the normal range and natriuresis continues (lower panel). The patient is in negative sodium balance but is not salt wasting; that is, ECFV remains over expanded. In renal salt wastage, a low-salt diet results in a 5-liter reduction in ECFV with persistent negative sodium balance. This constitutes salt wastage.

Salt balance is achieved through the integrated action of several factors that regulate sodium reabsorption by the renal tubules. Under normal circumstances, sodium reabsorption by the kidney is determined by the status of the effective arterial blood volume (EABV). When the EABV decreases, the kidney reabsorbs sodium and chloride, thereby protecting the EABV from contracting further. When the EABV increases, tubular reabsorption decreases, resulting in increased renal salt excretion. Thus, in healthy individuals, sodium balance is directly related to effective arterial blood volume.

When a healthy individual is placed on a low-salt diet, renal NaCl excretion exceeds dietary intake during the initial three to five days ( Figure 74.1 ). The time required to achieve low urine Na ⁺ and Cl ⁻ varies very little from individual to individual. Age, however, influences this time period, with older individuals taking a longer time to achieve sodium balance. The essential feature of renal salt wasting is ongoing natriuresis, despite a contraction in ECFV to below the lower limit of normal in a normal individual on a salt-free diet. This is clinically evident by an exaggerated decline in blood pressure, sometimes to overtly hypotensive levels or at least manifested by postural hypotension. This fall in blood pressure when coupled with a rise in heart rate upon standing up is a good clinical sign of volume depletion.

In subjects with sodium-retaining states, urinary sodium excretion is usually low but at times it may exceed 200 mEq under certain pathological conditions. For instance, an edematous patient on rigid salt restriction may excrete more than 200 mEq of sodium, but this does not necessarily constitute salt wasting. In this instance, the kidney may be excreting sodium above intake because the mobilization of edema and release of sodium into the circulation.

Now let us consider the situation of a subject who develops acute diarrhea, becomes volume-depleted and maintains an intake of sodium in the normal/high range (250 mEq/day). This subject will avidly reabsorb sodium, reflecting a decrease in EABV despite the high-salt intake and the urinary sodium will be much less than the intake. For instance, urinary sodium could be as low as 10 to 20 mEq a day. This subject is not in a steady-state because of volume depletion from the diarrhea, which provides a potent stimulus for sodium retention. In this case, the extent of sodium excretion does not reflect sodium intake. Accordingly, one should examine the composition of the urine in light of the clinical condition of the subject and the prevailing stimuli.

The concept of the steady-state and appropriate renal response also applies to the excretion of water and solutes other than sodium. For instance, a urinary excretion of 60 mmol of potassium per day may be appropriate for subjects consuming a diet containing 60 mEq of K ⁺ , but would be actually clear evidence potassium wasting in a hypokalemic patient on such diet. Here the hypokalemic stimulus, in someone with normal kidney function, leads to renal retention of potassium as a result of both enhanced reabsorption and decreased potassium secretion.

Urinary Constituents as Clinical Indices: General Considerations

Urine collected over 24 hours to determine the daily solute excretion is the conventional approach to evaluate external balance of a given solute that is primarily excreted by the kidneys. To the extent that ions such as Na ⁺ and K ⁺ are largely excreted by the kidneys one can estimate dietary intake in the steady-state and the kidney response to physiologic and pathophysiologic conditions by measuring the urinary concentrations and calculating the excretion rates when urine volume per unit time is known. In most clinical situations, a random “spot” urine collection is sufficient for the initial evaluation of a patient’s condition. Aside from being more convenient and rapid, the spot urine has the advantage that the stimulus is known at the time the urine is collected, and therefore the expected renal response can be predicted. For instance, prior to vigorous volume replacement it is useful to know how the kidneys are responding in terms of avid sodium retention or not.

Since errors are frequent when trying to collect urine accurately, urine creatinine should always be part of the measurement as a way to verify the completeness of the 24-hour collection. The urine creatinine assessment takes advantage of the fact that the kidneys excrete creatinine at a relatively fixed rate depending on the person’s muscle mass, a gauge of the daily creatinine input. While excretion of creatinine for an individual is relatively constant over a 24-hour period, provided that the serum creatinine is stable, it might vary depending upon protein intake. Muscular individuals and exercise, both increase creatinine excretion. For clinical purposes, it is generally accepted that a 24-hour urine collection is complete when the total urine creatinine is 20–25 mg per kg body mass for men and 15–20 mg per kg body mass for women.

Urinary Sodium and Chloride as Indices of Volume Status

The medical history and physical examination is always the first step in assessing arterial blood volume at the bedside. The clinical evaluation should be supplemented by the measurement of urine sodium and chloride in spot urine. The kidney has the extraordinary ability to elaborate virtually salt-free urine in the setting of dietary salt deprivation or when extra-renal losses of sodium are sufficient to decrease the EABV. Under normal circumstances, the kidney excretes sodium and chloride in the urine in an amount equivalent to the dietary intake. As discussed earlier, the kidney modulates sodium excretion and reabsorption in accordance with the status of the EABV. When the EABV decreases, the kidney reabsorbs sodium and chloride, thereby protecting the EABV from shrinking further. On the other hand, when the EABV expands, tubular reabsorption of sodium and chloride decreases, resulting in greater urinary excretion of these ions.

In analyzing renal sodium excretion, the existing volume status must always be taken into account, as noted in the earlier discussion regarding renal NaCl excretion in individuals in various clinical settings who were placed on low-salt diets. Further urinary chloride should be concurrently measured with urinary sodium. In volume-depleted states, the simultaneous measurement of urine sodium and chloride provides complementary information regarding the underlying cause, particularly when acid–base status is also altered such as in vomiting or diarrhea.

In states of volume depletion, the urine concentrations of sodium and chloride are both usually below 15 mEq/L, and in volume-replete conditions, they are well above 20 mEq/L (usually above 40 mEq/l). As summarized in Table 74.1 , four patterns of urinary sodium and chloride can be encountered.

(1)
Low levels of urinary sodium and chloride. This reflects low intake of salt or extrarenal salt losses. This can also be seen after cessation of diuretics when the volume deficit still persists.
(2)
Low urinary chloride with normal/high urinary sodium. In a setting of low EABV, this pattern suggests that another anion is present in the urine, usually HCO ₃ ⁻ , as in the setting of recent vomiting (see following paragraphs). This can also occur in the syndrome of congenital chloridorrhea, which is a syndrome of severe diarrhea, hypokalemia, and often metabolic alkalosis that is caused by a defect in the ileal and colonic chloride–bicarbonate exchange transporter that normally mediates active chloride absorption. In this syndrome, there is appropriate renal response to volume depletion with virtually absent urine chloride, but urinary sodium is elevated because of the high filtered load of bicarbonate which obligates sodium excretion.
(3)
Low urinary sodium with normal/high urinary chloride. In a patient with low EABV, as seen in the setting of diarrhea or laxative abuse, this pattern suggests that another cation is being excreted, most often ammonium and rarely potassium.
(4)
High excretion of urinary sodium and chloride. In a setting of low EABV, this pattern suggests ongoing diuretic use (both thiazide and loop diuretic) or an intrinsic tubular defect as in Bartter’s or Gitelman’s syndromes. This pattern is also consistent with salt wastage of any cause with post-obstructive diuresis being a classic example.

Table 74.1

Patterns of Urinary Sodium and Chloride in Various Conditions

a. Low urinary sodium and chloride	Salt restriction
	Extrarenal salt losses
	Previous use of diuretics
	Vomiting (maintenance phase)
b. Low urinary chloride but “normal/high” sodium	Vomiting with metabolic alkalosis (generation phase)
b. Low urinary chloride but “normal/high” sodium	Congenital chloridorrhea
c. Low urinary sodium but “normal/high” chloride	Diarrhea with volume depletion but metabolic acidosis
d. High urinary sodium and chloride	Loop and thiazide Diuretics
	Bartter’s and Gitelman’s syndromes
	Salt-wasting as seen in postobstructive diuresis and some types of chronic interstitial nephritis

Urinary chloride concentration under physiologic conditions, like urinary sodium, reflects the chloride intake. While urinary sodium and chloride change in the same direction under most circumstances, they may occasionally dissociate ( Table 74.1 ). The most notable situation is patients with vomiting and metabolic alkalosis with volume contraction. In this setting, urinary chloride concentration is low, reflecting the salt-retaining state induced by volume depletion. When the patient is actively vomiting, the increase in plasma bicarbonate and thus filtered bicarbonate might exceed the reabsorptive capacity of the tubules, resulting in bicarbonate excretion. Therefore, bicarbonate is rejected in the proximal tubules and this obligates the excretion of sodium resulting in a sodium bicarbonate diuresis. In this situation, chloride is retained appropriately, reflecting volume contraction. Thus, in a patient with active vomiting, the urine sodium may be high, while chloride concentration is invariably low ( Table 74.1 ).

In the maintenance phase of metabolic alkalosis, however, the ECFV depletion persists even if vomiting has ceased, leading to secondary hyperaldosteronism and sodium retention. Therefore urinary sodium, as well as urinary chloride concentration, can be low after protracted vomiting or nasogastric suction. Similarly, diuretic-induced metabolic alkalosis results in increased urinary concentrations of both sodium and chloride during the generation phase of the metabolic alkalosis when the nephron is under the effects of the diuretic ( Table 74.1 ). When the diuretic is discontinued, the metabolic alkalosis can be maintained by preexisting volume depletion, decreased glomerular filtration rate (GFR), and secondary hyperaldosteronism, and urinary sodium is usually low. The urinary concentration of chloride, likewise, is appropriately low (<15 mEq/L), reflecting the persistence of volume depletion and indeed, the administration of saline to restore the volume status to normal results in correction of the metabolic alkalosis.

Under conditions of metabolic acidosis other than the type resulting from renal tubular acidosis, the presence of large amounts of ammonium in the urine obligates the loss of an anion (chloride) resulting in a higher than expected urinary concentration of chloride. The urinary concentration of sodium, however, in this situation will be appropriately low reflecting the intravascular volume depletion. This pattern is typically seen in metabolic acidosis caused by diarrhea.

In patients with metabolic alkalosis due to Bartter’s or Gitelman’s syndromes, urinary chloride and sodium concentrations are elevated due to a tubular defect in reabsorption. When these patients become volume-depleted, renal conservation of salt is clearly suboptimal as manifested by relatively high level of urine sodium and chloride.

Spot Urine vs Timed Urine Collections

The 24 hour urine collection remains the proper way to evaluate solute excretion. One can use urine creatinine in a spot urine sample to calculate the urinary sodium/creatinine ratio as a way to estimate 24-hour urinary sodium excretion much in the same way that the urinary albumin/creatinine ratio is used to estimate 24-hour albumin excretion rate. The good linear relationship between the urine sodium/urine creatinine ratio has moreover been used to develop formulas to generate values of 24-hour sodium excretion based on the spot sodium/creatinine ratio. Kawasaki et al. proposed the formula:

\begin{matrix} 24 ‐ Hour Urinary Sodium (mEq / day) & = 16.3 \times \sqrt X_{Na} \\ {Where X}_{Na} = ({Spot U}_{Na} {/ Spot U}_{Cr}) \times predicted 24 h urinary Cr excretion \\ Predicted 24 h urinary Cr excretion (Male) & = - 12.63 \times Age + 15.12 \times Weight + 7.39 \times Height - 79.90 (mg / day) \\ Predicted 24 h urinary Cr excretion (Female) & = - 4.72 \times Age + 8.58 \times Weight + 5.09 \times Height - 74.50 (mg / day) \end{matrix}

$\begin{matrix} 24 ‐ Hour Urinary Sodium (mEq / day) & = 16.3 \times \sqrt X_{Na} \\ {Where X}_{Na} = ({Spot U}_{Na} {/ Spot U}_{Cr}) \times predicted 24 h urinary Cr excretion \\ Predicted 24 h urinary Cr excretion (Male) & = - 12.63 \times Age + 15.12 \times Weight + 7.39 \times Height - 79.90 (mg / day) \\ Predicted 24 h urinary Cr excretion (Female) & = - 4.72 \times Age + 8.58 \times Weight + 5.09 \times Height - 74.50 (mg / day) \end{matrix}$

Linear correlation between formula-estimated and actual 24-hour urinary sodium was 0.53. Others have reported a similar correlation (r=0.55).

Interestingly, a strong linear correlation between spot urine Sodium/Creatinine ratio and 24-hour Sodium excretion has been reported when evening spot samples were used but not when morning or random urine samples were used (see Table 74.2 and Figure 74.2 ). The linear correlation was even better when the evening spot urine Sodium/Creatinine ratio was adjusted for 24-hour creatinine excretion.

Table 74.2

Correlations between Spot Urine Sodium/Creatinine Ratio and 24-Hour Sodium Excretion

Source: Mann SJ, Gerber LM et al. Estimation of 24-hour sodium excretion from spot urine samples. J Clin Hypertens. 2010;12(3):174–80.

Correlation of 24-Hours Sodium Excretion with:	Spot Sodium/Creatinine Ratio: (correlation coefficients)	Predicted 24-Hours Sodium Excretion: (correlation coefficients)
Random Urine Sample	−0.01	0.17
Morning Urine Sample	0.14	0.31
Evening Urine Sample	0.67	0.86

Figure 74.2, Correlation between actual 24-hour sodium excretion and sodium excretion predicted from random sample (panel A), morning (AM) sample (panel B), and evening (PM) sample (panel C).

It should also be noted that there is a diurnal variation in the excretion of water and electrolytes. A larger amount of the potassium excreted daily occurs close to noon while the rate of excretion of sodium is lowest overnight. Ideally, this diurnal variation should be taken into account when interpreting urine tests, but in practice this is rarely done. There is limited information addressing the possible shortcomings of not taking into account the diurnal variation in the interpretation of electrolyte disorders at the bedside using spot urine samples. As noted above, a recent study suggests that spot urine sodium/creatinine ratio when taken in the evening correlates better with 24-Hour Sodium excretion than when taken in the morning. For practical purposes, however, a spot urine sample for electrolytes can be ordered any time of the day when the purpose is to assess a pathophysiologic situation such as volume depletion or renal failure.

Fractional Excretion of Sodium

The fractional excretion of sodium (FE _Na ) reflects the percent of filtered sodium that is excreted in the urine. The filtered sodium load can be determined from the product of the Glomerular Filtration Rate (GFR) and the Plasma Sodium (P _Na ). Sodium excretion is equal to the product of the urine sodium (U _Na ) and the urine flow rate (V). Since GFR is often not available, the FE _Na can be derived as follows:

{FE}_{Na} (%) = (U_{Na} \times V / GFR \times P_{Na}) \times 100

${FE}_{Na} (%) = (U_{Na} \times V / GFR \times P_{Na}) \times 100$

Since GFR=U _Cr ×V/P _Cr , the equation for FE _Na (%) becomes:

{FE}_{Na} (%) = ((U_{Na} \times V) {/ (U}_{Cr} \times {V / P}_{Cr}) \times P_{Na}) \times 100

${FE}_{Na} (%) = ((U_{Na} \times V) {/ (U}_{Cr} \times {V / P}_{Cr}) \times P_{Na}) \times 100$

Because V cancels out, it follows that:

{FE}_{Na} (%) = (U_{Na} \times P_{Cr}) {/ (U}_{Cr} \times P_{Na}) \times 100

${FE}_{Na} (%) = (U_{Na} \times P_{Cr}) {/ (U}_{Cr} \times P_{Na}) \times 100$

FE _Na is usually used to distinguish prerenal disease from acute tubular necrosis (ATN), the two most common causes of acute renal failure. In prerenal disease, the reabsorption of almost all of the filtered sodium represents an appropriate response to decreased renal perfusion. When GFR decreases due to renal vasoconstriction with intact tubular function, the FENa is low, usually less than 1%. In ATN, there is an associated tubular dysfunction, with an inability to conserve sodium and thus, the FE _Na is high, usually more than 2%. A value between 1 and 2% may be seen with a prerenal disorder or ATN, but typically it is below 1% in prerenal disorders. However, FE _Na can be low (<1%) in patients with ischemic ATN, if they are evaluated early in the course, as they make the transition from prerenal disease to postischemic ATN, or if they remain ischemic with established ATN. A low FE _Na is appropriate in this setting due to continued renal ischemia and can be achieved because tubular function is relatively intact early in the course of ATN. A clinical setting in which ATN is frequently associated with decreased FE _Na is the acute renal failure associated with contrast nephropathy. Occasionally, it is also observed in sepsis, myoglobinuria, obstructive uropathy, acute glomerulonephritis, and renal allograft rejection. In patients with chronic renal failure, renal vasoconstriction may not be associated with a FE _Na less than 1% due to chronic adaptation to an increased single-nephron GFR. Here, there is an adaptive decrease in tubular reabsorption in order to maintain sodium balance, thus making the interpretation of FE _Na difficult. One mathematical consideration also complicates the interpretation of FE _Na . As the renal failure progresses, the urinary creatinine in the denominator decreases, making the FE _Na increase on that basis alone. Thus, a patient can be pre-renal, but if the reduction in GFR is severe enough, the FE _Na can sometimes be pushed well above 1%. The pre-renal state can still be correctly diagnosed, however, by the gold standard test of seeing if the renal failure reverses with volume repletion. Due to the limitations of using FE _Na , the need arises for other indices that could help to clinically distinguish between ATN and pre-renal diseases (see following sections). This is not to say that the other indices are better but provide additional information that may help to assess the overall picture. Table 74.3 provides a summary of the expected values in the differential diagnosis of pre-renal vs. acute renal failure.

Table 74.3

Urine Indices Used in the Differential Diagnosis of Prerenal and Ischemic Intrinsic Renal Azotemia

Diagnostic Index	Prerenal Azotemia	Ischemic Intrinsic Azotemia
Fractional excretion of Na ⁺ (%)	<1%	>1%
Urinary Na ⁺ concentration (mEq/L)	<10	>20
Urinary creatinine/plasma creatinine ratio	>40	<20
Urinary urea nitrogen/plasma urea nitrogen ratio	>8	<3
Urine specific gravity	>1.018	<1.012
Urine osmolality (mOsm/kg H ₂ O)	>500	<250
Plasma BUN/creatinine ratio	>20	<10–15
Renal failure index, (UNa/(Ucr/Pcr))	<1	>1
Urine sediment	Hyaline casts	Muddy brown granular casts

Fractional Excretion of Urea

Urine sodium and chloride concentrations are difficult to interpret with concurrent diuretic therapy. The ensuing rise in urinary sodium and chloride likewise will raise the FE _Na and FE _Cl (fractional excretion of chloride) even in patients with pre-renal disease and oliguria. The coexistence of a pre-renal state and natriuretic agents (diuretics) clearly alters the interpretation of FE _Na such that despite volume depletion sodium excretion is increased during ongoing use of diuretics. In volume depletion and pre-renal azotemia due to vomiting or nasogastric suction, the ensuing metabolic alkalosis causes bicarbonaturia which increases urinary sodium and hence FE _Na at high levels. Given the limitations of using FE _Na , the need arose for a better marker for differentiating pre-renal conditions from ATN. This has led to the use of the fractional excretion of urea (FE _Urea ).

The FE _Urea is calculated using the formula used to calculate the fractional excretion of other electrolytes:

{FE}_{Urea} = (U_{ur} \times P_{cr}) {/ (P}_{ur} \times U_{cr}) \times 100

${FE}_{Urea} = (U_{ur} \times P_{cr}) {/ (P}_{ur} \times U_{cr}) \times 100$

The FE _Urea is 50 to 65% in healthy subjects and usually below 35% in pre-renal disease. To distinguish pre-renal states from ATN, the use of FE _Urea was evaluated by Carvounis et al. The study consisted of three groups; the first group had 50 patients with pre-renal disease, the second group had 27 patients with prerenal disease who had been given diuretics, and the third group had 25 patients with established ATN. The FE _Na was characteristically low (<1%) in the pre-renal group but was substantially higher than 1% in the prerenal group on diuretics and in the ATN group. In contrast, the FE _Urea was similar in both the prerenal groups (27.9 and 24.5% in groups 1 and 2, respectively) and significantly different from the ATN group (58.6%, p <0001). While 92% of patients in group 1 had a FE _Na <1%, only 48% of patients in group 2 had a low FE _Na . By contrast, 89% of patients who were prerenal and on diuretic therapy had a FE _Urea below 35%. In the prerenal cases (groups 1 and 2), FE _Urea had the best sensitivity and specificity (90 and 96%, respectively), and the best positive and negative predictive values (99 and 75%, respectively). These data suggest that when FE _Urea of less than 35% is found, 99% of such cases will be prerenal, including prerenal states with diuretic use.

The clearance of urea reflected by the FE _Urea is affected by decreased renal perfusion as seen in prerenal states. In evaluating the relationship of urine volume to urea excretion in humans using diuretic agents, it was found that the percent of filtered urea excreted increased with proximal diuretics (mannitol and acetazolamide) but not with distal agents (mercaptomerin and ethacrynic acid). This is consistent with the notion that urea is not reabsorbed significantly in the distal nephron. As most of the diuretics in clinical use work beyond the proximal tubule, FE _Urea should not be affected by their use. This is unlike FE _Na , which is increased by all forms of diuretics.

There are situations, however, where FE _Urea may not predict the ongoing prerenal status. Such settings should be kept in mind when FE _Urea is used. For instance, in states of osmotic diuresis, and with the administration of acetazolamide and mannitol, the proximal tubular absorption of salt and water is reduced, and thus an increased FE _Urea is to be expected even in the presence of renal hypoperfusion. In other studies, the utility of FE _Urea has been questioned. It is reasonable, however, to use fractional urea excretion in conjunction with other urinary indices such as FE _Na to gain a more complete picture than with either one alone in situations where the clinical picture is not straightforward.

Fractional Excretion of Lithium

Markers of proximal fluid reabsorption can be used to assess proximal tubular sodium reabsorption. Such a marker should be freely filtered at the glomerulus, should be reabsorbed in the same fashion as sodium and water in the proximal tubule, should not be reabsorbed beyond the proximal tubule, should not be secreted, and its plasma levels should not fluctuate significantly in response to hormones and changes in ECFV. Lithium fits this profile with the exception that it is reabsorbed in the loop of Henle as well as in the proximal tubule.

To assess proximal fluid reabsorption, the fractional clearance of endogenous or exogenous lithium can be utilized. The endogenous lithium clearance, despite its clear advantages over the exogenous method, would require very sensitive assays to measure trace amounts of lithium present in the serum and is difficult to use. Thus, it is necessary to give lithium either acutely or for several days prior to measuring lithium clearance or its fractional clearance. The fractional excretion of lithium (FE _Li ) is approximately 20% in healthy controls and below 10% in prerenal disease regardless of diuretic therapy. In ATN, FE _Li is typically higher than 25%.

FE _Li is probably a more sensitive index than FE _Na in differentiating prerenal states from ATN, especially in patients receiving concomitant diuretics. The use of lithium, however, has its own disadvantages. Its use is limited by the necessity of administering exogenous lithium, by the expertise needed to detect lithium levels in urine and by the acute changes in tubular electrolyte handling induced by lithium loading. Moreover, concern about the value of fractional Lithium excretion as a marker of proximal tubular fluid delivery has been raised because studies suggesting reabsorption of lithium in the thick ascending limb of the loop of Henle, which could be inhibited by loop diuretics. For clinical purposes, lithium clearance is not used but it offers valuable information in clinical physiology studies aimed at assessing segmental sodium handling throughout the nephron. It should be noted, however, that initial studies showed that lithium was reabsorbed almost entirely in the proximal tubule in a fashion similar to that of sodium, thus making lithium the marker of choice for determining proximal reabsorption. However, further studies uncovered a number of problems with using lithium clearance for assessing proximal reabsorption. First, lithium undergoes significant reabsorption in the loop of Henle, and to a lesser extent in the distal tubule and collecting duct. Amiloride can be used to block the reabsorption of lithium in the distal tubule and collecting duct especially when on a Na restricted diet. A second problem with the use of lithium is that in the presence of mineralocorticoid-induced volume expansion and with the administration of inhibitors of prostaglandin synthesis, lithium reabsorption in Henle’s loop is increased. In the aggregate, lithium clearance is the best available method to assess the delivery of sodium and water out of the proximal tubule, and is reasonably accurate in the steady-state. In states of volume and hormonal perturbations, however, lithium clearance is much less reliable.

Fractional Excretion of Uric acid has also been used, but has been found to be no better or inferior to Fe _Na and FE _Li . A summary of specificity and sensitivity of these various indices in the detection of prerenal vs acute renal failure is provided in Table 74.4 .

Table 74.4

Specificity and Sensitivity of FE _na , FE _urea and FE _UA at the Detection Limits of Pre-Renal (PR) and Acute Tubular Necrosis (ATN), as determined by the Receiver Operating Characteristic Curve Technique

Source: Steinhauslin F, Burnier M, Magnin JL, Munafo A, Buclin T, Diezi J, Biollaz J. Fractional excretion of trace lithium and uric acid in acute renal failure. J Am Soc Nephrol 1994;4:1429–37.

	FE _Na	FE _Urea	FE _UA
Pre-renal Detection Limit (%)	1.3	15	12
Specificity	78	72	78
Sensitivity	75	93	68
ATN Detection Limit (%)	3.5	25	20
Specificity	61	56	33
Sensitivity	96	96	96

Evaluation of Disorders of Water Excretion

The two variables governing water excretion include the magnitude of interstitial osmolality and the vasopressin-regulated, aquaporin-2 (AQP2)–dependent water transport in the collecting ducts. The interstitial osmolality increases whenever there is an increase in either sodium or urea concentration in the medullary interstitium and this provides the driving force for water absorption from the collecting duct for any given level of arginine vasopressin (AVP). For example, increased protein intake results in higher urea levels, which then enhances the ability of the kidney to concentrate the urine.

AVP regulates water permeability by principal cells in the collecting tubule by activating basolateral V2 receptors and this result in the insertion of AQP2 water channels into the apical surface of the cells, which results in an increase in water permeability. Synthesized AVP is released in response to two stimuli: an increase in plasma osmolality and changes in blood volume and pressure. AVP-mediated increases in total AQP2 expression contribute to the long-term regulation of urinary concentration. AVP also increases water reabsorption by the kidney by increasing the expression and activity of the Na ⁺ -K ⁺ -2Cl ⁻ cotransporter and the urea transporter. AVP-mediated regulation of AQP2 activity and expression provides by far the main mechanism for regulation of collecting duct water permeability. AVP binding to the V ₂ receptor activates a G-protein that increases adenylyl cyclase activity and cyclic adenosine monophosphate (cAMP) function. cAMP-activated protein kinase A then phosphorylates AQP2 and targets it to the apical surface of principal cells, which immediately increases water transport.

Measurements of AQP2 protein in the urine correlate with AVP levels, during water deprivation and after AVP administration ( Fig. 74.3 ). The critical role of AQP2 in the regulation of water excretion is now widely accepted, but at the clinical level, there is no need currently for measurements of AQP2 in the urine. For practical purposes, the clinical assessment of normal and abnormal handling of water by the kidneys starts with the measurement of plasma and urine osmolality (U _Osm ).

Figure 74.3, Urinary (u) aquaporin-2 (AQP2) excretion during 24 hours of water deprivation in 15 healthy subjects. Baseline, 24-hour collection before the experiment; 4 hours, 1200; 8 hours, 1600; 12 hours, 2000; 16 hours, 2400; 20 hours, 0400, 24 hours, 0800 the next morning. Results are medians with 25th, 75th, 10th, and 90th percentiles. * p< 01 and ** p< 001, significant deviation from baseline.

Plasma and Urine Osmolality

Plasma osmolality is tightly maintained within normal range (275–290 mOsmol/Kg). A 1% increase in plasma osmolality above 280 mOsm/kg triggers corrective response in the form of stimulation of the thirst center and activation of osmoreceptors resulting in the release of AVP. On the other hand, urine osmolality varies greatly, depending on water intake, and therefore should be interpreted in conjunction with measurements of plasma sodium or osmolality. In healthy adults, the urinary osmolality can vary from a minimum of about 50 to a maximum of about 1200 mOsm/kg H ₂ O. It is to be noted that the concentrating ability tends to fall with age, and the maximum U _Osm in an elderly patient may be only 600 to 700 mOsm/kg H ₂ O.

The urine flow rate is directly proportional to the number of osmoles excreted, and is inversely proportional to the urine osmolality, as expressed below:

Urine flow rate (or volume) = Number of osmoles per day / urine osmolality

$Urine flow rate (or volume) = Number of osmoles per day / urine osmolality$

Several examples of this relationship are shown in Table 75.5 . This table shows the urine volume needed to excrete a given amount of solute, at various levels of urine osmolality. In disorders in which the rate of AVP secretion is very low, and essentially constant, for example, in a patient with central diabetes insipidus (DI), the maximum U _Osm may be only 100 mOsm/kg H ₂ O. In this patient, the daily urine volume would be 12 L if 1200 mOsm of solute are to be excreted, as in a typical Western diet that generates a daily solute load of about 900 to 1200 mOsm. Dietary proteins and NaCl are the precursors of most of the urinary osmoles. In patients on a low-solute diet (0.6 g/kg body weight of protein and 10 mg/kg body weight of sodium), the total solute load may be as low as 300 mOsm. In such cases, only 3 L of urine are required to excrete the solute load, if the U _Osm is 100 mOsm/kg H ₂ O ( Table 74.5 ).

Table 74.5

Relationship Between UOsm and Urine Volume

UOsm	Urine Volume (L) if 1200 mOsm of Solutes are to be Excreted	Urine Volume (L) if 300 mOsm of Solutes are to be Excreted
100	12	3
300	4	1
600	2	0.5
1200	1	0.25

In a healthy subject on a diet that generates 1200 mOsm of solutes, about half of the urine osmoles consists of urea. Because the concentration of urea in the luminal fluid is roughly equal to that in the papillary interstitial component, urea will not cause water movement across the luminal membrane in the inner medulla. This renders urea an ineffective osmole. Thus, some authors believe that it is more correct to say that the urine flow rate is proportional to the number of nonurea osmoles (electrolytes, predominantly sodium and chloride in healthy subjects, and ammonium and β-hydroxybutyric acid during prolonged fasting) and to their concentrations in the medullary interstitial compartment. Accordingly, the effective osmoles in the urine of a healthy subject on a typical Western diet would be half the total solute load, or about 600 mOsm. The formula given above can therefore be modified as follows:

Urine flow rate = Number of “ effective ” urine osmoles per day / “ effective ” urine osmolality

$Urine flow rate = Number of “ effective ” urine osmoles per day / “ effective ” urine osmolality$

An important caveat to the generalization that urea is an ineffective osmole exists when an electrolyte-poor urine is excreted. In this case, the concentration of urea in the urine is more than that in the papillary interstitial compartment. In this setting, urea becomes an effective osmole and the increase in osmolality from the increase in urea reflects renal water conservation. The rationale for this change in permeability for urea is the fact that the chief goal of the urinary concentrating process is to ensure that a decrease in urine output ensues when ADH acts. For this to occur, effective osmoles must be excreted with water. These effective osmoles, usually NaCl, are replaced by urea when the content of electrolytes in urine is low. This latter situation may apply, for instance, to sodium-retaining states, such as congestive heart failure, where sodium and chloride are avidly reabsorbed in the proximal tubule and therefore, the excretion in the urine is low.

A complete analysis of water handling by the kidneys requires the calculation of free water excretion and electrolyte-free water excretion. These calculations offer information about the quantity of water absorbed or excreted by the kidney, information that is not provided by measuring only U _Osm . The calculation of electrolyte-free water excretion further dissects out the effect of free water excretion that is independent of urea, and helps predict the trend of plasma sodium.

You're Reading a Preview

Become a Clinical Tree membership for Full access and enjoy Unlimited articles

Become membership

If you are a member. Log in here