Renal Sodium Excretion, Edematous Disorders, and Diuretic Use


Renal Sodium Excretion, Edematous Disorders, and Diuretic Use

Robert W. Schrier

An understanding of body fluid volume regulation, as modulated by renal sodium and water excretion, is critical for the practice of clinical medicine. Knowledge of the intrarenal and extrarenal factors affecting renal sodium excretion is important to comprehend the mechanism of body fluid volume regulation in health and disease because the sodium ion is the primary determinant of extracellular fluid (ECF) volume. In this regard, the edematous disorders—cardiac failure, liver disease, and the nephrotic syndrome—present a particular challenge to our understanding of body fluid volume regulation. In normal humans, if the ECF volume is expanded by the administration of isotonic saline, then the kidney excretes the excess amount of sodium and water in the urine, thus returning the ECF volume to normal. However, in these edematous states, avid renal sodium and water retention persists despite expansion of ECF volume and the presence of total body sodium and water excess. In circumstances where advanced kidney disease is present and renal function and excretory capacity are diminished (e.g., acute or chronic intrinsic renal failure), it is obvious why the decreased glomerular filtration rate (GFR) may be associated with retention of sodium and water to the point of pulmonary and/or peripheral edema. However, it is clear that the integrity of the kidney as the ultimate effector organ of body fluid volume regulation is intact in patients with heart failure or liver disease and some patients with the nephrotic syndrome. Thus, the kidney must be responding to extrarenal signals from the afferent limb of a volume regulatory system in these edematous disorders. The study of these edematous disorders has led to our proposal of a unifying hypothesis of body fluid volume regulation that applies to both health and disease (18). The purpose of this chapter is to review the afferent and efferent mechanisms that determine renal sodium and water handling, particularly in the context of the edematous disorders, and discuss the treatment of edema with diuretic agents.

Sodium Ion as Determinant of ECF Volume

Sodium ions reside primarily in the ECF compartment to which they are extruded from cells by active transport mechanisms. These transport processes result in an intracellular sodium concentration of 10 mEq/L and an ECF sodium concentration of 145 mEq/L. The sodium ion and its major anions, chloride and bicarbonate, constitute >90% of the total solute in the ECF space. Thus, total body sodium and its accompanying anions are the osmotically active solutes that are the major determinants of ECF volume. In turn, the regulation of sodium balance is determined by the relationship among sodium intake, extrarenal sodium loss, and renal sodium excretion. Practically, renal sodium excretion may be considered to be the primary determinant of sodium balance because the kidney is able to excrete virtually sodium-free urine as well as rapidly excrete large sodium loads in response to diminished or increased sodium intakes, respectively.

A positive sodium balance is associated with increased amounts of sodium, located predominantly in the ECF compartment. Because cellular membranes are freely permeable to water, the osmotic gradient created by the addition of ECF sodium causes water to move from cells into the ECF compartment, thus expanding ECF volume. In addition, an increase in ECF osmolality stimulates the hypothalamic thirst center and leads to increased fluid intake and also releases arginine vasopressin (AVP) from the posterior pituitary, which decreases renal water excretion by increasing the water permeability of collecting duct epithelium (9). The latter two effects of an increased ECF osmolality result in a positive water balance, and the combined influence of positive sodium and water balances leads to further expansion of ECF volume. If this expansion of ECF is of sufficient magnitude, then an alteration of the Starling forces that govern the transfer of fluid from the vascular compartment to the surrounding interstitial spaces occurs and edema results (10). Conversely, a negative sodium balance results in a depletion of ECF volume. A decrease in ECF volume may result in a parallel decline in plasma volume. Maintenance of ECF volume and plasma volume is necessary for adequate circulation and survival of the organism. Thus, renal sodium and water retention is clearly appropriate in situations of ECF volume depletion. However, in edematous disorders, continued renal sodium and water retention despite total body sodium and water excess defines a paradoxical clinical situation.

It is worth mentioning that the osmolality of ECF is regulated by the AVP–thirst–renal axis (as discussed in depth in Chapter 1). However, the osmolality of the ECF is not a reliable index of ECF volume. ECF volume and its determinant total body sodium are best assessed by physical examination and determination of urinary sodium concentration. For example, a finding of generalized edema suggests an expanded ECF volume and increased total body sodium. Conversely, orthostatic tachycardia and/or hypotension, flat neck veins, and decreased skin turgor suggest depletion of ECF volume and decreased total body sodium. In fact, alterations in the osmolality of the ECF can occur in association with normal, increased, or decreased ECF volume (Chapter 1).

In summary, the control of ECF volume is dependent on the regulation of sodium balance. The kidneys play the pivotal role in the regulation of sodium balance and therefore of ECF volume homeostasis. In certain edema-forming states associated with a normal GFR, the kidney retains sodium and water despite expansion of the ECF volume and total body sodium and water. A knowledge of the afferent (“sensor”) and efferent (“effector”) mechanisms of sodium and water retention associated with the edematous disorders forms the basis of our understanding of body fluid volume regulation.

Afferent Mechanisms Involved in Body Fluid Volume Regulation


If the afferent receptors of body fluid volume regulation primarily sense total blood volume, then the kidneys of edematous patients should increase their excretion of sodium and water since their total blood volumes are increased. However, as mentioned, this does not occur in patients with advanced cardiac failure, liver disease, or the nephrotic syndrome. Thus, there must be some body fluid compartment that is still “underfilled”—even in the presence of expansion of total ECF and blood volumes—and comprises the afferent limb of renal sodium and water retention in patients with edematous disorders. In 1948, Peters coined the enigmatic term effective blood volume as a reference to such an underfilled body fluid compartment (11). Accordingly, extrarenal signals must be initiated by this decrease in effective blood volume, which enhances tubular sodium and water reabsorption by the otherwise normal kidney. In this regard, it is clear that renal sodium and water retention can occur in patients with cardiac failure or cirrhosis and in some patients with the nephrotic syndrome before any diminution in GFR.

Borst and deVries (12) first suggested cardiac output as the primary modulator of renal sodium and water excretion. In this context, the level of cardiac output would constitute the effective blood volume and thus serve as the primary stimulus for renal sodium and water retention in patients with edematous disorders. Although this concept is appealing, substantial renal sodium and water retention may occur in the presence of an increase in cardiac output. For example, a significant elevation in cardiac output may occur in the presence of avid renal sodium and water retention and expansion of ECF volume in association with cirrhosis, pregnancy, arteriovenous (AV) fistulas, and other causes of high-output cardiac failure, such as thyrotoxicosis and beriberi. Consequently, there must exist some other or additional determinant(s) of effective blood volume.


The unifying hypothesis of body fluid volume regulation in health and disease states that the fullness of the arterial vascular compartment or the so-called effective arterial blood volume (EABV) is the primary determinant of renal sodium and water excretion (18). In a 70-kg man, total body water approximates 42 L, of which only 0.7 L (1.7% of total body water) resides in the arterial circulation. From a teleologic viewpoint, it is attractive to propose that the primacy for regulation of renal sodium and water excretion, and body fluid volume homeostasis, is modulated by the smallest body fluid compartment—thus endowing the system with exquisite sensitivity to relatively small changes in body fluid volume. Another advantage of the integrity of the arterial circulation constituting the main afferent sensing compartment for body fluid volume regulation is that perfusion of the vital organs is dependent on the arterial circulation. As a result, total ECF, interstitial fluid, or total intravascular volumes are not primary determinants of renal sodium and water excretion, and the venous component of intravascular volume likewise is excluded as the primary determinant of sodium and water excretion, because all of these body fluid compartments may be expanded while the renal sodium and water retention persists in edematous patients. It is acknowledged, however, that there are experimental and clinical circumstances in which selective rises in right and left atrial pressure stimulate the release of atrial natriuretic peptide (ANP) (13) or suppression of AVP (14), respectively, which may enhance sodium and water excretion. These events, however, must be subservient to the more potent determinants of the arterial circulation because the patient with advanced left or right ventricular dysfunction, or both, exhibits avid sodium and water retention despite markedly elevated atrial and ventricular pressures.


The EABV is a measure of the adequacy of arterial blood volume to “fill” the capacity of the arterial circulation. Normal arterial filling exists when the ratio of cardiac output to systemic vascular resistance maintains venous return and cardiac output at normal levels. Thus, arterial underfilling may be initiated by either a decrease in cardiac output or a fall in systemic arterial resistance (i.e., arterial vasodilatation, which increases the holding capacity of the arterial vascular tree). Arterial underfilling results in unloading of high-pressure baroreceptors with subsequent activation of the three major neurohormonal vasoconstrictor systems—namely, the sympathetic nervous system, the renin–angiotensin–aldosterone system, and the nonosmotic release of AVP—which diminish renal hemodynamics and promote renal sodium and water retention. This hypothesis accounts for the initiation of sodium and water retention in low- and high-output cardiac failure, liver disease, and other states of arterial underfilling (Figs. 2-1 and 2-2).


As mentioned, the afferent volume receptors for such a volume regulatory system must reside in the arterial vascular tree, such as the high-pressure baroreceptors in the carotid sinus, aortic arch, left ventricle, and juxtaglomerular apparatus. Although the low-pressure volume receptors of the thorax (cardiac atria, right ventricle, and pulmonary vessels) must be of some importance to the volume regulatory system (15,16), there is considerable evidence that arterial receptors can predominate over low-pressure receptors in volume control in mammals.

High-Pressure Volume Receptors

In humans, the presence of volume-sensitive receptors in the arterial circulation was first suggested by Epstein et al. (17) based on observations made in patients with traumatic AV fistulas. Closure of traumatic AV fistulas was associated with an immediate increase in renal sodium excretion independent of concomitant changes in either GFR or renal blood flow (RBF) (17). Closure of AV fistulas is associated with a decreased rate of emptying of the arterial blood into the venous circulation, as demonstrated by closure-induced increases in diastolic arterial pressure and decreases in cardiac output (17). Further evidence implicating the relative “fullness” of the arterial vascular tree as being the major sensor in modulating renal sodium excretion can be found in denervation experiments. In these studies, surgical or pharmacologic interruption of sympathetic efferent neural pathways emanating from high-pressure areas inhibited the natriuretic response to volume expansion (1820). Moreover, reduction of pressure or stretch at the carotid sinus, similar to that produced by decreased cardiac output or arterial hypotension, has been shown to activate the sympathetic nervous system and promote renal sodium and water retention (21). High-pressure baroreceptors also appear to be important factors in regulating nonosmotic release of AVP and thus renal water excretion (22). One of the best-defined high-pressure receptors that are known to act in an appropriate manner to maintain constancy of the EABV is the renal afferent arteriolar baroreceptor (i.e., juxtaglomerular apparatus). This baroreceptor is an important factor in the control of renal renin secretion and consequently angiotensin II formation and aldosterone synthesis and release (23). The vasoconstrictor and sodium-retaining effects of angiotensin II and sodium-retaining effect of aldosterone then act to restore the fullness of the arterial circulation.

Figure 2–1 Clinical conditions in which a decrease in cardiac output causes arterial underfilling with resultant neurohumoral activation and renal sodium and water retention. (From Schrier RW. Decreased effective blood volume in edematous disorders: what does this mean? J Am Soc Nephrol. 2007;18(7): 2028–2031, permission conveyed through Copyright Clearance Center, Inc.)

Figure 2–2 Clinical conditions in which systemic arterial vasodilation causes arterial underfilling with resultant neurohumoral activation and renal sodium and water retention. (From Schrier RW. Decreased effective blood volume in edematous disorders: what does this mean? J Am Soc Nephrol. 2007;18(7):2028–2031, permission conveyed through Copyright Clearance Center, Inc.)

Low-Pressure Volume Receptors

Low-pressure sensors also may have an important role to play in body fluid volume regulation because the more compliant venous side of the circulation contains up to 85% of the total blood volume at any given time (Table 2-1). In fact, a variety of maneuvers that decrease thoracic venous return, such as prolonged standing (24), lower-extremity tourniquets (25,26), and positive pressure breathing (27), are associated with diminished renal sodium excretion. Conversely, maneuvers that augment venous filling, such as recumbency (28) and negative pressure breathing (29), are associated with increased renal sodium excretion. Moreover, a direct correlation between renal sodium excretion and left atrial pressure has been demonstrated in dogs, suggesting a role for an atrial receptor as one type of intrathoracic sensor (30). Immersion in water to the neck or so-called head-out water immersion results in a pressure gradient from 80 mm Hg at the foot to 0 mm Hg at water level. This maneuver increases venous return to the heart. In response to head-out water immersion, a profound increase in renal excretion of salt and water occurs independent of major changes in either GFR or renal hemodynamics (31). As first suggested by Gauer et al. (29) and Henry et al. (32), physiologically significant left atrial receptors have been shown to contribute to ECF volume regulation by exerting nonosmotic control over AVP secretion and thus over renal water excretion. In addition, the atria have been demonstrated to be the site for synthesis, storage, and release of vasoactive and natriuretic humoral agents (33,34).

Thus, increased filling of the thoracic vascular and cardiac atria would be expected to signal the kidney to increase urinary sodium excretion in order to return the blood volume to normal. However, in the setting of chronic heart failure, renal sodium and water retention occur despite increased atrial pressure, which loads the low-pressure baroreceptors. Thus, in low-output chronic heart failure, diminished cardiac output must exert the predominant effect via unloading of high-pressure arterial baroreceptors. Chronic studies in animals employing experimental tricuspid insufficiency (35) further support this hypothesis. The increase in right atrial pressure was associated with avid renal sodium retention in these animal models. However, a concomitant fall in cardiac output likely explains the sodium retention.

Zucker et al. (36) have demonstrated that the inhibition of renal sympathetic nerve activity that is seen during acute left atrial distention is lost during chronic heart failure in dogs. Moreover, a decrease in cardiac preload fails to produce the expected parasympathetic withdrawal and sympathetic activation in humans with heart failure (37). These findings are also consistent with the observation of a strong positive correlation between left atrial pressure and coronary sinus norepinephrine, a marker of cardiac adrenergic activity, in patients with chronic heart failure (38). Taken together, these findings suggest that the normal inhibitory control of sympathetic activation accompanying increased atrial pressures is lost in heart failure patients and somehow may even be converted to a stimulatory signal.

In summary, the afferent or sensor mechanisms for sodium and water excretion may be preferentially located on the arterial side of the circulation where diminished fullness of the arterial vascular tree owing to decreased cardiac output or systemic arterial vasodilation results in unloading of high-pressure receptors and subsequent renal sodium and water retention. Reflexes from low-pressure volume receptors may also be altered so as to influence renal sodium and water handling. In any event, changes in systemic and renal hemodynamics and activation of various neurohormonal systems largely comprise the efferent limb of the volume regulatory system.

Efferent Mechanisms Involved in Body Fluid Volume Regulation


Arterial underfilling secondary to a diminished cardiac output or systemic arterial vasodilation elicits a number of compensatory neurohormonal responses that act to maintain the integrity of the arterial circulation by promoting systemic vasoconstriction as well as expansion of the ECF volume through renal sodium and water retention. As noted, the three major neurohormonal vasoconstrictor systems activated in response to arterial underfilling are the sympathetic nervous system, renin–angiotensin–aldosterone system, and nonosmotic release of AVP. Baroreceptor activation of the sympathetic nervous system appears to be the primary integrator of the hormonal vasoconstrictor systems involved in the volume control system because the nonosmotic release of AVP involves sympathetic stimulation of the supraoptic and paraventricular nuclei in the hypothalamus (39), and activation of the renin–angiotensin–aldosterone system involves renal β-adrenergic stimulation (40). Thus, in low-output cardiac failure, diminished integrity of the arterial circulation as determined by decreased cardiac output causes unloading of arterial baroreceptors in the carotid sinus and aortic arch. Systemic arterial vasodilation produces unloading of these arterial baroreceptors in the setting of high-output cardiac failure, cirrhosis, and other states of arterial underfilling. This baroreceptor inactivation results in diminution of the tonic inhibitory effect of afferent vagal and glossopharyngeal pathways to the central nervous system (CNS) and initiates an increase in sympathetic efferent adrenergic tone with subsequent activation of the renin–­angiotensin–aldosterone system. Various counterregulatory, vasodilatory hormones may also be activated in heart failure, including natriuretic peptides and vasodilating renal prostaglandins. Activation of these various neurohormonal vasoconstrictor and vasodilator systems substantially determines renal sodium and water handling in the edematous disorders and comprises a major part of the efferent limb of body fluid volume regulation. The pathogenesis of sodium and water retention associated with cardiac failure, liver disease, and the nephrotic syndrome are now reviewed in the context of the unifying arterial underfilling hypothesis of body fluid volume regulation.


Sodium and water retention and resultant edema formation are cardinal features of chronic cardiac failure. In fact, the inability to excrete a sodium load has been used as an index of the presence of heart failure (41), and a defect in water excretion is encountered routinely in such patients (42). Two theories have been proposed to explain the renal response to cardiac failure. The “backward” theory of heart failure, proposed in 1832, suggests that increased venous hydrostatic pressure owing to increased ventricular filling pressures causes edema by promoting transudation of fluid from the intravascular to the interstitial compartment, resulting in edema formation (43). The reduced intravascular volume then signals the kidneys to retain sodium and water, further exacerbating the venous hypertension and formation of edema. The alternative “forward” theory of cardiac failure suggests that a primary decrease in cardiac output activates afferent and efferent pathways and results in renal sodium retention (44). As pointed out by Smith (26), these theories are not mutually exclusive and both are operant in the pathophysiology of heart failure because both central venous hypertension and arterial underfilling are implicated in the afferent limb of body fluid volume regulation. Nevertheless, the dominant signal for sodium and water retention in cardiac failure appears to occur in the arterial circulation. Decreased cardiac output is the cause of the arterial underfilling in the case of low-output heart failure, whereas systemic arterial vasodilation initiates the afferent limb of sodium and water retention in high-output cardiac failure (Fig. 2-1).

Renal Hemodynamics in Cardiac Failure

Glomerular Filtration Rate

Many early investigators believed that the cause of sodium retention in heart failure was a decrease in GFR; however, studies failed to confirm such a correlation. In fact, GFR is often normal in early heart failure and may even be elevated in states of high-output cardiac failure. It is acknowledged, however, that the contribution of GFR to sodium balance is difficult to evaluate because very minute changes in GFR could lead to substantial changes in sodium excretion if absolute sodium reabsorption remained unchanged. Nevertheless, although GFR may be diminished in patients with advanced heart failure, an increase in tubular sodium reabsorption undoubtedly is an important cause of sodium and water retention in cardiac failure.

Renal Blood Flow

Heart failure is commonly associated with an increase in renal vascular resistance and a decrease in RBF. In general, RBF decreases in proportion to the decrease in cardiac output. Some investigators have also shown a redistribution of RBF from outer cortical nephrons to juxtamedullary nephrons in experimental heart failure (45). It was proposed that deeper nephrons with longer loops of Henle reabsorb sodium more avidly; thus, the redistribution of blood flow to these nephrons with heart failure would result in renal sodium retention. However, other investigators have not been able to demonstrate such a redistribution of blood flow in other models of cardiac failure (46). Thus, the role of redistribution of RBF in the sodium retention of cardiac failure remains uncertain.

Filtration Fraction

Filtration fraction is often increased in heart failure because RBF falls as cardiac output decreases and GFR is preserved. An increase in filtration fraction results in increased protein concentration and oncotic pressure in the efferent arterioles and peritubular capillaries that surround the proximal tubules. Such an increase in peritubular oncotic pressure has been proposed to increase sodium and water reabsorption in the proximal tubule. These changes in renal hemodynamics and filtration fraction, which favor proximal tubular sodium reabsorption, are primarily a consequence of constriction of the efferent arterioles within the kidney. These renal hemodynamic changes are mediated mainly by activation of neurohormonal vasoconstrictor systems because both activation of renal nerves and increased circulating norepinephrine and angiotensin II have been implicated in efferent arteriolar vasoconstriction (47,48). In addition, decreased activity of such substances as vasodilating renal prostaglandins also may play a role in renal vasoconstriction (49).

Of note, micropuncture studies in dogs with vena caval constriction and AV fistulas have demonstrated the importance of distal nephron sites of increased sodium reabsorption. Increased filtration fraction primarily affects proximal tubular sodium reabsorption. Thus, although clearance and micropuncture studies in animals with heart failure have demonstrated increased sodium reabsorption in the proximal tubule (50), distal sodium reabsorption also seems to be involved. Furthermore, changes in filtration fraction have been observed in heart failure long before changes in sodium balance occur, questioning the dominance of peritubular factors and proximal reabsorption in the sodium retention of cardiac failure.

The Sympathetic Nervous System in Cardiac Failure

The sympathetic nervous system is unquestionably activated in patients with heart failure. Various studies have demonstrated elevated peripheral venous plasma norepinephrine concentrations in heart failure patients. Using tritiated norepinephrine in patients with advanced heart failure, Davis et al. (51) and Hasking et al. (52) have shown that both increased norepinephrine secretion and decreased norepinephrine clearance contribute to the high venous plasma norepinephrine concentrations seen in these patients, suggesting that increased sympathetic activity is at least partially responsible for the elevated circulating plasma norepinephrine. We have demonstrated that the initial rise in plasma norepinephrine in heart failure is solely caused by increased norepinephrine secretion, providing evidence of increased sympathetic nervous system activity early in the course of cardiac failure (53). Moreover, plasma norepinephrine is increased in patients with asymptomatic left ventricular dysfunction (i.e., before the onset of overt heart failure) (54). Finally, studies employing peroneal nerve microneurography to directly assess sympathetic nerve activity to muscle have confirmed the presence of increased sympathetic activity in heart failure patients (55). Significantly, the degree of activation of the sympathetic nervous system—as assessed by the peripheral venous plasma norepinephrine concentration—has been correlated with poor prognosis in heart failure (56).

Activation of Renal Nerves

Renal nerves also are activated in human heart failure (52). Enhanced renal sympathetic activity may contribute to the avid sodium and water retention in heart failure by promoting renal vasoconstriction, stimulation of the renin–angiotensin–aldosterone system, and direct effects on the proximal tubule epithelium. Indeed, intrarenal adrenergic blockade has been shown to cause a natriuresis in experimental heart failure (57). In addition, in rats, renal nerve stimulation has been demonstrated to produce approximately a 25% reduction in sodium excretion and urine volume (58). The diminished renal sodium excretion that accompanies renal nerve stimulation may be mediated by at least two mechanisms. As already discussed, studies performed in rats have demonstrated that norepinephrine-induced efferent arteriolar constriction alters peritubular hemodynamic forces in favor of increased tubular sodium reabsorption (47). In addition, renal nerves have been shown to exert a direct influence on sodium reabsorption in the proximal convoluted tubule (58).

Bello-Reuss et al. (58) have demonstrated this direct effect of renal nerve activation to enhance proximal tubular sodium reabsorption in whole-kidney and individual nephron studies in rats. In these animals, renal nerve stimulation produced an increase in the tubular fluid–to–plasma inulin concentration ratio in the late proximal tubule, an outcome of increased fractional sodium and water reabsorption in this segment of the nephron. Hence, increased renal nerve activity may promote sodium retention by a mechanism independent of changes in renal hemodynamics. On the other hand, sodium retention persists in dogs with denervated transplanted kidneys and chronic vena caval constriction. Moreover, renal denervation does not prevent ascites in dogs with chronic vena caval constriction (59). Thus, renal nerves probably contribute but do not fully account for the avid sodium retention of heart failure.

The Renin–Angiotensin–Aldosterone System in Cardiac Failure

The renin–angiotensin–aldosterone system also is activated in heart failure, as assessed by plasma renin activity (PRA) (60). Renin acts on angiotensinogen to produce angiotensin I, which is then converted by angiotensin-converting enzyme (ACE) to angiotensin II. In heart failure, the resultant increased plasma concentration of angiotensin II exerts important circulatory effects, including peripheral arterial and venous vascular constriction, renal vasoconstriction, and cardiac inotropism. Activation of angiotensin receptors on the proximal tubule epithelium directly stimulates the Na+/H+ exchanger 3 and thereby increases sodium reabsorption (61). Angiotensin II also acts to promote the secretion of the sodium-retaining hormone aldosterone by the adrenal cortex and in positive-feedback stimulation of the sympathetic nervous system. Activation of this hormonal system may promote sodium retention in the kidney via several mechanisms, as discussed next. Moreover, like adrenergic activation, stimulation of the renin–angiotensin–aldosterone system is associated with an unfavorable prognosis in heart failure (62).

Renal Effects of Increased Angiotensin II and Aldosterone

Angiotensin II may contribute to the sodium and water retention in heart failure through direct and indirect effects on proximal tubular sodium reabsorption and, as mentioned, by stimulating the release of aldosterone from the adrenal gland. Angiotensin II causes preferential renal efferent arteriolar constriction, resulting in decreased RBF and an increased filtration fraction. As with renal nerve stimulation, this results in increased peritubular capillary oncotic pressure and reduced peritubular capillary hydrostatic pressure, which favors the reabsorption of sodium and water in the proximal tubule (48). Moreover, as noted, angiotensin II has been shown to enhance sodium reabsorption in the proximal tubule (63). In a study of the rat proximal tubule, Liu and Cogan (63) demonstrated increased tubular sodium chloride reabsorption during the infusion of angiotensin II, whereas the angiotensin II receptor antagonist, saralasin, decreased proximal tubular sodium chloride reabsorption. Finally, in a report from Abassi et al. (64), the administration of the angiotensin II receptor antagonist, losartan, to decompensated sodium-retaining rats with heart failure secondary to AV fistulas produced a marked natriuresis. Although proximal tubular sodium handling was not examined in this investigation, the observation that losartan restored renal responsiveness to ANP is consistent with a losartan-induced increase in the delivery of sodium to the distal tubular site of ANP action. The role of distal tubular sodium delivery in the renal sodium retention of heart failure is discussed later.

Watkins et al. (65) studied a conscious dog model of heart failure in order to define more precisely the role of the renin–angiotensin–aldosterone axis in cardiac failure. Using either partial constriction of the pulmonary artery or thoracic inferior vena cava (TIVC), these workers acutely produced a low cardiac output state characterized by reduced blood pressure, increased PRA and aldosterone concentrations, and renal sodium retention. As plasma volume and body weight increased over several days, the aforementioned variables all returned toward control levels. During the initial hyperreninemic period, a single injection of an ACE inhibitor significantly lowered blood pressure. Also, chronic administration of the converting enzyme inhibitor prevented a rise in aldosterone and prevented 30% of the sodium retention and subsequent volume expansion. These studies lend support to the hypothesis that aldosterone is an important factor in the pathogenesis of cardiac edema and suggest that angiotensin II plays an important physiologic role in heart failure by supporting blood pressure because of its vasoconstrictor effect and maintaining blood volume secondary to the sodium-retaining effects of angiotensin II and aldosterone. It likewise becomes clear that, depending on the status of cardiac decompensation and plasma volume, the patient with heart failure may have a high or normal PRA and aldosterone level. This may explain some of the controversy that existed regarding the levels of these hormones in patients with heart failure.

A further role for renin–angiotensin–aldosterone system activation in the sodium retention of human heart failure is supported by the finding that urinary sodium excretion inversely correlates with PRA and urinary aldosterone excretion in heart failure patients (66). However, the administration of an ACE inhibitor (ACEI) during heart failure does not consistently increase urinary sodium excretion in spite of a consistent fall in plasma aldosterone concentration (67). The simultaneous fall in blood pressure caused by decreased circulating concentrations of angiotensin II, however, may activate hemodynamic and neurohormonal mechanisms that could obscure the natriuretic response to lowered angiotensin II and aldosterone concentrations. Support for this hypothesis comes from the study performed by Hensen et al. (68). We examined the effect of the specific aldosterone antagonist, spironolactone, on urinary sodium excretion in patients with heart failure who were withdrawn from all medications before study. Avid sodium retention occurred in all patients throughout the period before aldosterone antagonism. During therapy with spironolactone (200 mg b.i.d.), all heart failure patients exhibited a significant increase in urinary sodium excretion and reversal of the positive sodium balance (Fig. 2-3). Moreover, the urinary sodium-to-potassium concentration ratio significantly increased during spironolactone administration, consistent with a decrease in aldosterone action in the distal nephron. Of note, PRA and norepinephrine increased and ANP decreased during the administration of spironolactone. Thus, this investigation demonstrates reversal of the sodium retention of heart failure with the administration of an aldosterone antagonist, despite further activation of various antinatriuretic influences, including stimulation of the renin–angiotensin and sympathetic nervous systems, and supports a role for aldosterone in the renal sodium retention. A prospective trial, Randomized Aldactone Evaluation Study (RALES), has shown improved survival of heart failure patients receiving 25 mg/day of spironolactone (a competitive inhibitor of aldosterone) (69). This effect of spironolactone in the RALES investigation was found to be independent of any change in sodium balance. An effect of spironolactone to block the effect of aldosterone-mediated cardiac fibrosis has been suggested as the mediator of this improved survival response. Natriuretic doses of spironolactone rarely have been used in patients with heart failure. One study was performed in congestive heart failure (CHF) patients receiving low-dose ACEIs who had diuretic resistance. These patients demonstrated a natriuresis with a daily dose of 100 mg of spironolactone (70). The Eplerenone Post-AMI Heart Failure Efficacy and Survival Study (EPHESUS) demonstrated reduced mortality after acute myocardial infarction (71).

In the Acute Decompensated Heart Failure Registry (ADHERE), decompensated CHF patients resistant to oral diuretics were hospitalized and 90% were given intravenous diuretics. Forty-two percent of these patients were discharged with unresolved symptoms, 50% lost ≤5 lb, and 20% actually gained weight. Approximately 25% to 30% of CHF patients become resistant to diuretics and, as discussed later, secondary hyperaldosteronism is an important factor in such diuretic resistance (72).

Nevertheless, natriuretic doses of mineralocorticoid antagonists may not be part of the therapeutic armamentarium for heart failure, primarily because of the fear of hyperkalemia (73). Many of these CHF patients are receiving ACEIs or angiotensin receptor blockers (ARBs) and/or β-blockers, which predispose to hyperkalemia. Whether low-potassium diet, sodium polystyrene sulfonate (Kayexalate), and potassium-losing diuretics may avoid the occurrence of hyperkalemia during use of natriuretic doses of mineralocorticoid antagonists has not been studied. Given the challenge of treating cardiac patients with acute decompensation, as noted in the ADHERE registry, inhibiting the secondary hyperaldosteronism in sodium-retaining CHF patients who are diuretic resistant needs to be undertaken. Isotonic removal of sodium in CHF patients with ultrafiltration is another therapeutic approach. Fluid removal in CHF patients with ultrafiltration or diuretics can improve cardiac and renal function in addition to treating pulmonary congestion and edema. The mechanisms are shown in Figure 2-4 (74).

Figure 2–3 Diuretics, digoxin, and ACEI were withdrawn 4 days before admission to General Clinical Research Center. The subjects were placed on a constant daily diet of 100 mEq sodium and 60 mEq potassium. (A) Upper panel demonstrates the positive cumulative sodium balance in the six patients (four ischemic heart disease, one idiopathic cardiomyopathy, and one aortic valvular disease). (B) Middle panel demonstrates in the same patients the significant negative cumulative sodium balance during 200 mg b.i.d. spironolactone (P < 0.01). (C) Lower panel demonstrates the increase in urine Na+:K+ concentration ratio during spironolactone in all six patients (P < 0.05), a finding compatible with aldosterone antagonism. Mean plasma potassium increased from 3.86 ± 0.2 to 4.1 ± 0.2 mEq/L during spironolactone treatment (P < 0.05). Mean systolic blood pressure (112 ± 7 mm Hg vs. 110 ± mm Hg, NS) and creatinine clearance (87 ± 7 mL/min vs. 87.2 ± 8 mL/min, NS) did not change with spironolactone treatment. Plasma hANP decreased significantly with spironolactone (147 ± 58 mg/L vs. 83 ± 30 mg/L, P < 0.05). Fluid intake was not restricted and a mean of 2 kg weight loss occurred. (From Bansal S, Lindenfeld J, Schrier RW. Sodium retention in heart failure and cirrhosis: potential role of natriuretic doses of mineralocorticoid antagonist? Circ Heart Fail. 2009;2(4):370–376, with permission from Wolters Kluwer Health, Inc.)

Figure 2–4 Mechanisms in congestive heart failure whereby negative sodium and water balance by loop diuretics or ultrafiltration therapy may improve myocardial and renal function. (Reprinted from Schrier RW. Role of diminished renal function in cardiovascular mortality: marker or pathogenetic factor? J Am Coll Cardiol. 2006;47(1):1–8, with permission from Elsevier.)

The Nonosmotic Release of Arginine Vasopressin in Cardiac Failure

Plasma AVP is often elevated in patients with CHF and correlates in general with the clinical and hemodynamic severity of disease and the serum sodium level. Using a sensitive radioimmunoassay for AVP, Szatalowicz et al. (75) initially showed that plasma AVP was detectable in 30 of 37 patients with cardiac failure and hyponatremia. It was concluded that the nonosmotic AVP release in these patients was the result of baroreceptor stimulation secondary to diminished cardiac output because these patients had sufficient hyponatremia and hypo-osmolality, which would normally suppress maximally the osmotic release of AVP. Riegger et al. (76) also have reported that several patients with heart failure had inappropriately high plasma AVP levels. Cardiac output increased and plasma AVP levels normalized when two of these patients were treated with hemofiltration to remove excess body fluid. Other studies have also incriminated AVP in hyponatremic CHF patients (77,78). Taken together, these observations demonstrate enhanced nonosmotic AVP release in response to a decrease in cardiac output (i.e., arterial underfilling).

Renal Effects of Arginine Vasopressin

AVP, via stimulation of its renal or V2 receptor subtype, enhances water reabsorption in the distal nephron, namely, the cortical and medullary collecting ducts. The evidence supporting a role for AVP in the water retention of heart failure comes from studies using selective peptide and nonpeptide antagonists of the V2 receptor of AVP in several animal models of cardiac failure. For example, Ishikawa et al. (79) have assessed the antidiuretic effect of plasma AVP in a low-output model of cardiac failure secondary to vena caval constriction in rats. Plasma AVP concentrations were increased in these animals, and an antagonist of the antidiuretic effect of AVP reversed the defect in water excretion. An orally active nonpeptide V2 receptor AVP antagonist, OPC-31260, was originally described in 1992 (80). The intravenous administration of OPC-31260 during a dose-ranging study in normal human subjects was shown to increase urine output to a similar extent as 20 mg of furosemide given intravenously (81). Virtually simultaneous publications by Xu et al. (82) from our laboratory and Nielsen et al. (83) demonstrated the upregulation of aquaporin 2 (AQP2) water channels in coronary-ligated rats with CHF. The latter group also demonstrated that AQP1 and AQP3 were not upregulated in this CHF model and that increased trafficking of the AQP2 to the apical membrane occurred. Our group further showed that a V2 vasopressin antagonist reversed the upregulation of the AQP2 protein in the renal cortex and medulla of the CHF rats (81). This effect of the nonosmotic release of AVP to cause water retention in cardiac failure recently has been associated with increased transcription of messenger RNA (mRNA) for the AVP preprohormone in the rat hypothalamus (84).

In a study by Bichet et al. (85), the effect of the ACEI captopril and the α1-adrenergic blocker prazosin to reverse the abnormality in water retention in patients with class III and IV heart failure was examined. The resultant cardiac afterload reduction and increased cardiac output with either agent were associated with improved water excretion and significant suppression of AVP in response to an acute water load. A role of angiotensin II in modulating the effect of AVP in heart failure was unlikely because captopril and prazosin had divergent effects on the renin–angiotensin system; yet their effects to suppress plasma AVP and improve water excretion were comparable. In this regard, it is important to note that in this study by Bichet et al. (85), the average decrease in mean arterial pressure was 5 mm Hg, a decrement that is less than the 7% to 10% necessary to activate the nonosmotic release of AVP (86). Thus, these results are compatible with the suggestion that a decrease in stroke volume and cardiac output, rather than a fall in mean arterial pressure, may sometimes be the primary stimulus for the nonosmotic release of AVP in low-output cardiac failure. The association of improved cardiac output and water excretion during afterload reduction is compatible with unloading of high-pressure baroreceptors leading to increased AVP release.

The most recent advance relative to the nonosmotic release of AVP in CHF is the FDA approval of vasopressin receptor antagonists for clinical use in the United States. Conivaptan, a combined V1 and V2 receptor antagonist, has been approved for treatment of hyponatremia in cardiac failure. This antagonist can be used inhospital by intravenous administration for 4 days. The potential effect of the combined V1 and V2 antagonist properties in heart failure is shown in Figure 2-5 (74). Recently, the first orally active V2 receptor antagonist, tolvaptan, has been approved for use in cardiac failure, cirrhosis, and the syndrome of inappropriate antidiuretic hormone (SIADH) (87). In association with the increase in plasma sodium concentration in hyponatremic CHF patients, the self-reported SF12 demonstrated a significant improvement in mental status in these patients. There are other V2 receptor antagonists in phase 3 trials. Taken together, these agents are known as aquaretics to emphasize that the resultant increase in solute-free water excretion occurs in the absence of a change in electrolyte excretion. This is the major difference with diuretics that increase urinary sodium chloride and other electrolyte excretion. These aquaretic agents can correct plasma sodium concentration in the absence of fluid restriction. In chronic hyponatremia, the correction of plasma sodium concentration with an aquaretic should not exceed 8 mmol over 8 hours or 10 to 12 mmol over 24 hours in order to avoid osmotic demyelination.

Altered renal hemodynamics may contribute to water retention in heart failure in addition to persistent AVP secretion. Decreased RBF and increased filtration fraction would be expected to increase proximal reabsorption of sodium and water, thereby diminishing fluid delivery to distal diluting segments. Increasing distal fluid delivery by administration of furosemide has improved the diluting ability of patients with heart failure (88).

In summary, activation of the sympathetic nervous system, the renin–angiotensin–aldosterone system, and the nonosmotic release of AVP by exerting direct (tubular) and indirect (hemodynamic) effects on the kidneys are implicated in the renal sodium and water retention of heart failure. These neuroendocrine mechanisms appear to be activated in response to arterial underfilling and suppressed by maneuvers that restore the integrity of the arterial circulation toward normal. In addition, the effects of these neurohormonal vasoconstrictor systems may be counterbalanced by endogenous vasodilatory and natriuretic hormones.

Natriuretic Peptides in Cardiac Failure

The natriuretic peptides, including ANP and brain natriuretic peptide (BNP), circulate at increased concentrations in patients with heart failure (89,90). These peptide hormones possess natriuretic, vasorelaxant, and renin-, aldosterone-, and sympatho-inhibiting properties (91). Both ANP and BNP appear to be released primarily from the heart in response to increased atrial or ventricular end-diastolic or transmural pressures. We demonstrated that increased ANP production rather than decreased metabolic clearance was the major factor contributing to the elevated plasma ANP concentrations in a study of ANP kinetics in patients with cardiac failure (92). This finding is consistent with the observed increase in expression of both ANP and BNP mRNA in the cardiac ventricles of humans and animals with heart failure (93,94). BNP has been shown to reduce pulmonary capillary wedge pressure (PCWP) and increase cardiac index in acute CHF (95). In a coronary ligation model of heart failure in rats, the infusion of a monoclonal antibody shown to specifically block endogenous ANP in vivo caused a significant rise in right atrial pressure, left ventricular end-diastolic pressure, and systemic vascular resistance (96). Thus, natriuretic peptides appear to attenuate to some degree the arterial and venous vasoconstriction of heart failure.

Figure 2–5 Pathways whereby vasopressin stimulation of V2 and V1a receptors can contribute to events that worsen cardiac function. (Reprinted from Schrier RW. Role of diminished renal function in cardiovascular mortality: marker or pathogenetic factor? J Am Coll Cardiol. 2006;47(1):1–8, with permission from Elsevier.)

Renal Effects of the Natriuretic Peptides

In normal humans, ANP and BNP increase GFR and urinary sodium excretion with no change or only a slight fall in RBF (97). These changes in renal hemodynamics are likely mediated by afferent arteriolar vasodilation with constriction of the efferent arterioles. However, in addition to increasing GFR and filtered sodium load as a mechanism of their natriuretic effect, ANP and BNP are specific inhibitors of sodium reabsorption in the collecting tubule (98). An important role for endogenous ANP in the renal sodium balance of heart failure has been demonstrated by Lee et al. (99). Similar decreases in cardiac output were induced in two groups of dogs by constriction of the TIVC or acute rapid ventricular pacing. Sodium retention paralleled the activation of the renin–angiotensin–aldosterone system in the TIVC constriction group. Atrial pressures and plasma ANP were not increased in this group of dogs. In comparison, the ventricular pacing group did not experience sodium retention or activation of the renin–angiotensin–aldosterone system. This group had similar reductions in cardiac output and arterial pressure as the TIVC constriction group but, unlike the TIVC constriction group, had increased atrial pressures and circulating endogenous ANP levels. In a third group, exogenous ANP was administered to TIVC constriction dogs to increase plasma ANP levels to those observed in the pacing model. The ANP infusion prevented the sodium retention and activation of the renin–angiotensin–aldosterone system.

Unfortunately, the administration of synthetic ANP to patients with low-output heart failure results in a much smaller increase in renal sodium excretion and less significant changes in renal hemodynamics compared with normal subjects (100). Like ANP, the natriuretic effect of BNP is blunted in rats with high-output heart failure produced by AV fistulas (101). In a trial of BNP, 127 patients with a PCWP of 18 mm Hg or higher and a cardiac index of 2.7 L/min/m2 of body surface area or less were randomly assigned to double-blind treatment with placebo or BNP (nesiritide) infused at a rate of 0.015 or 0.030 μg/kg of body weight per minute for 6 hours (95). BNP significantly decreased PCWP and resulted in improvements in global clinical status in most patients (i.e., reduced dyspnea and fatigue). The most common side effect was dose-related hypotension, which was usually asymptomatic. Therefore, intravenous nesiritide may be useful for the short-term treatment of patients hospitalized with decompensated CHF (95). A recent retrospective report, however, demonstrated an increase in serum creatinine and mortality in heart failure patients receiving BNP (102).

The possible mechanism of the relative renal resistance to natriuretic peptides in heart failure are the following:

1. Downregulation of renal ANP receptors

2. Secretion of inactive immunoreactive ANP

3. Enhanced renal neutral endopeptidase activity limiting the delivery of ANP to receptor sites

4. Hyperaldosteronism causing increased sodium reabsorption in the distal renal tubule

5. Diminished delivery of sodium to the distal renal tubule site of ANP action

A strong positive correlation between plasma ANP and urinary cGMP (the second messenger for the natriuretic effect of ANP and BNP in vivo) has been shown in sodium-retaining patients with heart failure (103). This observation supports the active biologic responsiveness of renal ANP receptors in heart failure and thus suggests that diminished distal tubular sodium delivery may explain the natriuretic peptide resistance observed in patients with cardiac failure. In cirrhosis, another edematous disorder associated with renal ANP resistance, increased distal tubular sodium delivery with mannitol has been shown to reverse the ANP resistance (104). Moreover, in heart failure, the administration of an angiotensin II receptor antagonist or furosemide, which is expected to increase distal tubular sodium delivery, also improves the renal response to ANP (64,105). Finally, studies in rats with experimental heart failure have demonstrated that renal denervation reverses the ANP resistance (106), an effect likely mediated by increased distal tubular sodium delivery. In Figure 2-6, the proposed role of diminished distal tubular sodium delivery in natriuretic peptide resistance and impaired aldosterone escape in states of arterial underfilling is shown.

Renal Prostaglandins in Cardiac Failure

Renal prostaglandins do not regulate renal sodium excretion or renal hemodynamics to any significant degree in normal subjects and intact animals. However, prostaglandin activity is increased in patients with heart failure and has been shown to correlate with the severity of disease as assessed by the degree of hyponatremia (107). Moreover, it has been well documented that the administration of a cyclooxygenase inhibitor in heart failure patients may result in acute reversible renal failure, an effect proposed to result from inhibition of renal prostaglandins (108). An investigation in patients with moderate heart failure and a normal sodium intake demonstrated that the administration of acetylsalicylic acid in doses that decrease the synthesis of renal prostaglandin E2 results in a significant reduction in urinary sodium excretion (109). These observations support a role for prostaglandins in attenuating the renal vasoconstriction and sodium retention in patients with heart failure.


Two earlier theories have attempted to explain the pathogenesis of sodium and water retention in cirrhosis (110,111). The classic “underfill hypothesis” suggested that ascites formation secondary to portal hypertension leads to decreased plasma volume, which secondarily increases renal sodium and water retention (110). However, results of animal studies have shown that sodium and water retention precedes ascites formation in cirrhotic animals, thus contradicting the hypothesis (111). Moreover, plasma volume is increased, not decreased, in cirrhosis. An alternative hypothesis was therefore proposed in which primary renal sodium and water retention occurs secondary to a hepatorenal reflex. This would lead to plasma volume expansion of both the venous and arterial compartments and cause overflow ascites (111). This “overfill hypothesis” of ascites formation in cirrhotic patients, however, did not explain the progressive stimulation of the neurohumoral profile as cirrhotic patients progress from compensated to decompensated with ascites to hepatorenal syndrome. Against this background, we have suggested a primary role for systemic arterial vasodilation for the initiation of renal sodium and water retention in cirrhosis (Fig. 2-2) (112,113). This theory encompasses the entire range of cirrhosis from compensated to decompensated to hepatorenal syndrome and explains the progressive increases in both plasma volume and neurohormonal activation that occur as cirrhosis worsens.

Figure 2–6 Proposed mechanism of natriuretic peptide resistance and impaired aldosterone escape in states of arterial underfilling. GFR, glomerular filtration rate. (Reprinted from Schrier RW. Water and sodium retention in edematous disorders: role of vasopressin and aldosterone. Am J Med. 2006;119(7 Suppl 1):S47–53, with permission from Elsevier.)

Systemic Arterial Vasodilation Hypothesis

Splanchnic arterial vasodilation occurs early in cirrhosis, and the resultant arterial underfilling stimulates sodium and water retention with plasma volume expansion before ascites formation. The normal plasma hormone concentrations in compensated cirrhotic patients are relatively increased for the degree of sodium and water retention and plasma volume expansion. The mediators of the early splanchnic vasodilation in cirrhosis may include the opening of existing shunts, activation of vasodilating hormones, and ultimately the development of collaterals. Vasodilation may occur at other sites including the skin, muscle, and lung as cirrhosis progresses. However, although the presence of splanchnic arterial vasodilation is well documented in experimental and human cirrhosis, the development of arterial vasodilation involving other vascular territories is less certain.

Increased synthesis and release of the potent vasodilator nitric oxide, perhaps owing to increased circulating levels of endotoxin in cirrhosis, have been proposed to account for the arterial vasodilation and hyperdynamic circulation seen in cirrhotic patients (114117). Although nitric oxide activity is difficult to assess in vivo, indirect evidence supports this hypothesis. For example, urinary cGMP, the second messenger of nitric oxide, is increased in patients with cirrhosis before the development of ascites and in some patients before an increase in circulating ANP concentrations (117). Markedly increased cGMP concentrations in aortic tissue from rats have been demonstrated in experimental cirrhosis (116). In these animals, aortic cGMP concentrations correlated inversely with arterial pressure (r = −0.54, P < 0.0001). Significantly, the chronic administration of the nitric oxide synthesis (NOS) inhibitor NG-nitro-L-arginine-methyl-ester (L-NAME, 10 mg/kg/day for 7 days) induced a marked reduction in aortic cGMP concentration and an increase in arterial blood pressure in cirrhotic rats to similar levels obtained in L-NAME–treated control animals. This indicated that the high aortic cGMP content and decreased arterial blood pressure in cirrhotic rats were due to an increased NOS (116). Normalization of vascular nitric oxide production by chronic NOS inhibition corrects the systemic hemodynamic abnormalities in cirrhotic rats with ascites (117). Furthermore, chronic L-NAME treatment in drinking water for 10 days normalized mean arterial pressure, cardiac output, and systemic vascular resistance in these cirrhotic animals (118). The neurohumoral response in cirrhotic rats was also normalized as PRA, aldosterone, and AVP returned to control levels after 7 days of NOS inhibition (119). These hemodynamic and neurohumoral alterations during NOS inhibition were associated with profound reversal of the sodium and water retention in these cirrhotic rats. Moreover, Guarner et al. (115) have demonstrated elevated serum nitrite and nitrate levels—a crude index of in vivo nitric oxide generation—in 51 cirrhotic patients. Of note, in these patients, the elevated serum nitrite and nitrate levels significantly correlated with plasma endotoxin levels and decreased in response to a reduction in plasma endotoxin concentration following the administration of the antibiotic colistin (115). In addition, an enhanced sensitivity to mediators of endothelium-­dependent vasodilation has been demonstrated in human cirrhosis (120). Taken together, these observations are compatible with the presence of nitric oxide–induced arterial vasodilation in cirrhosis. Endogenous opioids may also contribute to the peripheral vasodilation and renal sodium and water retention in cirrhosis, as the administration of opioid antagonists (e.g., naloxone or naltrexone) increased sodium and water excretion after water loading in cirrhotic subjects (121). Other factors that have been proposed to mediate the splanchnic vasodilation in cirrhosis include vasodilating prostaglandins, glucagon, calcitonin gene–related peptide, platelet-activating factor, substance P, and vasoactive intestinal peptide; however, definitive proof is lacking for these potential medications. As with cardiac failure, pretreatment hyponatremia and high plasma concentrations of renin, norepinephrine, and aldosterone portend a poor prognosis in the cirrhotic patient. The highest plasma concentrations of these hormones and the lowest blood pressures occur as the decompensated cirrhotic patient with ascites progresses toward the hepatorenal syndrome.

Nephron Sites of Sodium Retention in Cirrhosis

There is indirect evidence for both enhanced proximal and distal tubular reabsorption in human cirrhotic subjects. The following findings support enhanced proximal tubular reabsorption in hepatic cirrhosis: (a) maneuvers that expand plasma volume and increase distal nephron delivery of fluid (i.e., head-out neck immersion and infusion of saline or mannitol) result in increased renal sodium excretion and solute-free water formation independent of changes in GFR (122); (b) in some water-loaded cirrhotic patients with ascites and minimal urine osmolalities, urine flow rates (an index of distal delivery of tubular fluid under these circumstances) are lower than in normal subjects (123); and (c) enhanced proximal reabsorption of tubular fluid has been found in micropuncture studies with chronic bile duct ligation (124).

Evidence for enhanced distal nephron sodium reabsorption is based on the following observations: (a) water-loaded patients with sodium retention and cirrhosis with minimal urine osmolalities often have urine flow rates comparable to normal controls (125); (b) water-loaded cirrhotic patients with minimal urine osmolalities have increased calculated distal fractional sodium reabsorption after receiving hypotonic saline infusions (125); (c) acetazolamide, a diuretic acting at the proximal tubule, produces a significant natriuresis in cirrhotic subjects only when there is concomitant distal nephron blockade of sodium reabsorption with ethacrynic acid (126); and (d) micropuncture studies in the dimethylnitrosamine and bile duct ligation models of cirrhosis demonstrate enhanced distal nephron sodium reabsorption (127,128).

In summary, clinical and experimental studies suggest that both proximal and distal nephron sites participate in enhanced renal tubular sodium reabsorption in cirrhosis. As in cardiac failure, neurohormonal activation appears to play a major role in the sodium and water retention of cirrhosis. The mechanisms responsible for enhanced sodium and water reabsorption in cirrhosis are no doubt multifactorial. A decrease in GFR may not be observed in some sodium-retaining cirrhotic patients, suggesting that sodium retention can occur independently of a decrease in GFR. An increase in renal vascular resistance and filtration fraction often is seen in decompensated cirrhosis. Thus, peritubular physical forces (decreased hydrostatic pressure and increased oncotic pressure) may act to enhance proximal tubular sodium reabsorption in advanced cirrhosis.

The Sympathetic Nervous System in Cirrhosis

Elevated plasma levels of norepinephrine have been observed in cirrhotic patients with ascites. Plasma norepinephrine levels correlate positively with plasma AVP concentrations and PRA and negatively with urinary sodium excretion (122). Moreover, norepinephrine spillover rates in cirrhotic patients have been shown to be increased compared with normal controls, whereas norepinephrine clearance rates were comparable between the two groups (129). Floras et al. (130), using the technique of peroneal nerve microneurography to directly measure sympathetic nerve activity to muscle, also have demonstrated adrenergic activation in cirrhotic patients. Finally, Ring-Larsen et al. (131) have demonstrated normal hepatic norepinephrine clearances and increased renal norepinephrine release in cirrhotic patients. Taken together, these findings are compatible with the presence of systemic and renal adrenergic activation in cirrhosis.

These findings indicate that increased activity of the sympathetic nervous system and renal nerves may result in enhanced renal sodium reabsorption in cirrhosis. As mentioned, renal adrenergic stimulation has been shown to increase proximal tubular sodium reabsorption. In addition, a negative correlation between plasma norepinephrine and urinary sodium excretion has been shown in cirrhotic patients (132). Ring-Larsen et al. (133) have demonstrated an inverse correlation between plasma norepinephrine and RBF. Moreover, in the report from Floras et al. (130), muscle sympathetic nerve activity was inversely correlated with urinary sodium excretion.

Role of Aldosterone in Cirrhosis

In a study by Gregory et al. (134), 16 out of 21 cirrhotic patients exhibited disappearance of ascites with spironolactone treatment over 3 to 4 weeks; sometimes, a loop diuretic furosemide was added. Thus, the near-uniform natriuretic response to spironolactone in cirrhotic patients, when given in adequate doses (100–400 mg/day), suggests that the increased levels of aldosterone contribute to the increased distal sodium reabsorption. Since exogenous aldosterone administration does not cause edema in normal subjects and absence of edema is a hallmark of primary hyperaldosteronism, the major problem in cirrhotic patients appears to be related to a failure to escape from the sodium-retaining effect of aldosterone as occurs in normal subjects. Aldosterone escape in normal subjects is associated with increased sodium delivery to the distal collecting duct site of aldosterone action. In cirrhosis, the increased neurohumoral activation, particularly angiotensin and α-adrenergic stimulation, enhances proximal tubular sodium reabsorption and diminishes distal sodium delivery. This sequence of events appears to be the main cause for the failure of cirrhotic patients to escape from the sodium-retaining effect of aldosterone. Because of the elevated endogenous plasma level of aldosterone, mineralocorticoid antagonists, such as spironolactone, need to be given in higher doses. Diuretic resistance in cirrhosis is therefore defined as absence of a natriuresis with daily spironolactone doses of 400 mg and furosemide of 160 mg. On this background, mineralocorticoid antagonists are established as the diuretic of first choice in cirrhotic patients, followed by a loop diuretic if necessary.

The Nonosmotic Release of Vasopressin in Cirrhosis

Hyponatremia with impaired ability to excrete a water load occurs in a substantial number of patients with cirrhosis of the liver, thereby demonstrating an impairment in urinary dilution in these patients (135,136). Decompensated cirrhotic patients with ascites and/or edema have an abnormal response to water administration, whereas cirrhotic patients without ascites or edema usually excrete water normally (136). There are two potential mechanisms for this inability to excrete solute-free water in decompensated cirrhotic patients with ascites: (a) a derangement in renal hemodynamics with decreased fluid delivery to the distal nephron; and (b) an extrarenal mechanism involving nonosmotic AVP release. Volume expansion maneuvers that improve distal fluid delivery of ascitic fluid (137,138), as well as head-out water immersion (122), improve urinary dilution and water excretion in cirrhosis. These maneuvers also increase central blood volume and could improve water excretion by suppressing baroreceptor-mediated nonosmotic AVP release.

Studies of patients with cirrhosis also implicate the nonosmotic release of AVP as a major factor responsible for water retention in cirrhosis. Bichet et al. (138) studied 26 cirrhotic patients who received a standard water load (20 mL/kg). Patients could be separated into two groups on the basis of their ability to excrete this water load: those able to excrete >80% of the water load in 5 hours (“excretors”) and those unable to excrete a water load normally (“nonexcretors”). Nonexcretors had lower serum sodium concentrations and higher plasma AVP levels after the water load. These nonexcretors also were found to have higher pulse rates, lower plasma albumin concentrations, higher PRA and aldosterone concentrations, and higher plasma norepinephrine levels than normonatremic cirrhotic patients with normal water excretion (139). A greater increase in systemic arterial vasodilatation in the nonexcretors is supported by these studies. Thus, arterial underfilling may provide the nonosmotic stimulus for AVP release in hyponatremic cirrhotic patients. Enhancement of central blood volume by water immersion to the neck suppressed AVP release and improved, but did not normalize, water excretion in subsequent experiments (122). However, a comparable suppression of AVP with head-out water immersion and norepinephrine infusion normalized water excretion in decompensated cirrhotic patients (140). The increment in water excretion with this combined maneuver, which increases renal perfusion pressure, would be expected to increase distal fluid delivery.

Studies performed in rats made cirrhotic by exposure to carbon tetrachloride and phenobarbital supported AVP hypersecretion as the predominant mechanism of the impairment of water excretion because the administration of a V2 vasopressin antagonist normalized water excretion in 9 of the 10 rats studied (141). Moreover, using the orally active nonpeptide V2 receptor AVP antagonist OPC-31260, Tsuboi et al. (142) normalized the defect in solute-free water excretion in this animal model of cirrhosis. Additional experimental data supporting a primary role for AVP in the impaired water excretion in cirrhosis were reported by Fujita et al. (143). These investigators examined the effect of experimental cirrhosis on expression of the mRNA for the AVP-dependent collecting duct water channel, AQP2, in rats. Binding of AVP to the V2 receptor initiates a chain of intracellular signaling events that leads to the insertion of AQP2 water channels into the apical membrane of collecting duct cells, thus rendering these cells permeable to water. In the cirrhotic rats studied by Fujita et al. (143), AQP2 mRNA was markedly increased as compared with control animals. Moreover, an oral water load (30 mL/kg) did not reduce AQP2 mRNA expression, but the blockade of AVP action by the V2 receptor AVP antagonist OPC-31260 significantly diminished its expression in the cirrhotic animals. Nonpeptide V2 receptor antagonists have been shown to increase plasma sodium concentration in hyponatremic cirrhotic patients and improve urinary dilution (144,145).

Natriuretic Peptides in Cirrhosis

As with other edematous states associated with arterial underfilling, the neurohumoral responses to the systemic arterial vasodilation of cirrhosis are associated with factors that diminish distal sodium delivery. The impaired aldosterone escape (146) and resistance to ANP (147) that occur in cirrhosis, therefore, are most likely mediated by diminished distal sodium delivery to the collecting duct site of these hormonal actions. As with experimental cardiac failure, renal denervation has been shown to reverse the resistance to ANP in experimental cirrhosis (148). This finding supports a role of diminished distal sodium delivery in the ANP resistance. Moreover, Skorecki et al. (147) have demonstrated a normal increase in urinary cGMP but no natriuresis in some cirrhotic patients infused with ANP. Since cGMP is the secondary messenger of ANP, this finding supports the biologic responsiveness of renal ANP receptors in these patients. An increased distal sodium delivery with mannitol (as assessed by lithium clearance) has been shown to reverse resistance to exogenous ANP.

Renal Prostaglandins in Cirrhosis

Zambraski and Dunn (149) demonstrated that prostaglandins with vasodilator properties are necessary to maintain RBF and GFR in dogs with cirrhosis secondary to bile duct ligation. Similar conclusions about the importance of prostaglandins have been obtained in cirrhotic humans. Inhibition of prostaglandin synthesis in decompensated cirrhotic patients decreases RBF, GFR, sodium excretion, and solute-free water excretion and impairs the natriuretic response to diuretic agents (150,151). Infusion of prostaglandin has been shown to reverse the diminutions in RBF and GFR observed after prostaglandin inhibition in cirrhotic patients (151). Moreover, inhibition of prostaglandin synthesis may cause a syndrome that mimics the hepatorenal syndrome (150). Vasodilating renal prostaglandins may also play an important counterregulatory role in early or well-compensated cirrhosis (152).

In summary, numerous afferent and efferent mechanisms are involved in the abnormal sodium and water excretion seen in patients with liver disease. These mechanisms appear to be initiated by arterial underfilling caused by primary systemic arterial vasodilation. The sympathetic nervous system, renin–angiotensin–aldosterone axis, and the nonosmotic release of AVP are the major effector components of this increased sodium and water reabsorption, which may also be modulated by the release of natriuretic peptides and renal prostaglandins.


Two views of the pathogenesis of edema formation in the nephrotic syndrome are illustrated in Figure 2-7. According to the “underfill” theory, urinary loss of albumin occurs as a consequence of an increase in glomerular capillary permeability and results in hypoalbuminemia. This decline in serum albumin lowers intravascular colloid oncotic pressure, thereby increasing transudation of plasma from the intravascular to the interstitial space. It is this decrease in plasma volume that causes arterial underfilling and serves as the stimulus for renal sodium and water retention. Ultimately, the decrease in intravascular colloid oncotic pressure and the increase in interstitial hydrostatic pressure secondary to edema formation come into balance, and the edematous state stabilizes. Thus, the diminution in total plasma volume is the critical afferent stimulus in inducing renal sodium and water retention and should be observed in the initiating phase of formation. Several lines of evidence support this traditional underfill theory of edema formation in the nephrotic syndrome (153): (a) plasma volume may be modestly decreased in some nephrotic patients in the absence of diuretic therapy; (b) systemic arterial hypotension and diminished cardiac output, correctable by plasma volume expansion, have been observed in some patients with nephrotic syndrome; (c) some nephrotic patients have humoral markers of arterial underfilling such as elevated plasma levels of PRA, aldosterone, and catecholamines; and (d) head-out water immersion and intravascular infusion of albumin, maneuvers that increase plasma volume, may result in substantial increases in GFR and in fractional excretion of sodium chloride and water in these patients.

Figure 2–7 Proposed underfilled and overfilled mechanisms of sodium and water retention in the nephrotic syndrome. (Adapted from Bansal S, Lindenfeld J, Schrier RW. Sodium retention in cardiac failure and cirrhosis: potential role of natriuretic doses of mineralocorticoid antagonist? Circ Heart Fail. 2009;2(4):370–376.)

Usberti et al. (154) have described two groups of nephrotic syndrome patients distinguished on the basis of their plasma albumin concentrations. Patients in the first group had a plasma albumin concentration of <1.7 g/dL associated with low blood volumes and plasma ANP levels, elevated plasma angiotensin II concentrations, and increased proximal tubular reabsorption of sodium (determined by lithium clearance). In contrast, the second-group patients had a plasma albumin concentration >1.7 g/dL and exhibited normal blood volumes and plasma hormone concentrations. In all patients, blood volume was positively correlated with the plasma albumin concentration, and PRA was inversely correlated with both blood volume and plasma albumin concentration. Of note, GFR was not different between the first- and second-group patients (100 ± 25 mL/min vs. 101 ± 22 mL/min, P = NS), whereas urinary sodium excretion was substantially lower in the first-group patients (4.88 ± 5.53 mEq/4 h vs. 29.9 ± 9.3 mEq/4 h, P < 0.001). Moreover, the acute expansion of blood volume in the first-group patients normalized PRA, plasma angiotensin II and aldosterone concentrations, fractional sodium excretion, and lithium clearance, whereas circulating ANP concentrations increased. Taken together, these observations support the traditional underfill view of the pathogenesis of edema formation in the nephrotic syndrome.

To further explore the state of arterial filling in patients with the nephrotic syndrome, sympathetic nervous system activity was evaluated in six edematous patients with the nephrotic syndrome of various parenchymal etiologies and in six normal control subjects (155). As mentioned, increased adrenergic activity occurs in states of arterial underfilling and may be the earliest sign. Sympathetic nervous system activity was assessed by determining plasma norepinephrine secretion and clearance rates using a whole-body steady-state radionuclide tracer method. Patients were withdrawn from all medications 7 days before study. Mean creatinine clearances and serum creatinine concentrations were normal in both the nephrotic syndrome patients and controls. However, the nephrotic syndrome patients exhibited significant hypoalbuminemia (2.0 ± 0.4 g/dL vs. 3.8 ± 0.1 g/dL, P < 0.01). The supine plasma norepinephrine levels were elevated in the patients with the nephrotic syndrome as compared with controls. More significantly, the secretion rate of norepinephrine was significantly increased in nephrotic patients, whereas the clearance rate of norepinephrine was similar in the two groups (Fig. 2-8). PRA and plasma aldosterone, AVP, and ANP concentrations were not different in nephrotic syndrome patients compared with controls. These observations indicate that the sympathetic nervous system is activated in patients with the nephrotic syndrome before a significant fall in GFR or a marked activation of either the renin–angiotensin–aldosterone system or the nonosmotic release of AVP. These data also support the presence of arterial underfilling in the nephrotic syndrome.

Several investigators, however, have challenged this traditional underfill model based on the following observations: (a) several studies of plasma and/or blood volume in edematous nephrotic patients have reported either normal or elevated values (156); (b) hypertension and low PRA, two indices suggesting volume expansion, have been reported in some patients with nephrotic syndrome (157); (c) hypoalbuminemia in animal studies as well as in patients with analbuminemia do not necessarily lead to edema formation (158); and (d) a low filtration fraction is often observed in patients with the nephrotic syndrome (159), in contrast to the increased filtration fraction usually associated with states of arterial underfilling. Thus, the “overfill” hypothesis has been proposed to account for nephrotic edema formation in some patients. According to this view, the renal retention of sodium and water occurs as a primary intrarenal phenomenon independent of systemic factors. In this setting, renal sodium and water retention produces an expanded plasma volume, and the overfilled plasma volume then leaks into the interstitium and induces edema formation. The hypoalbuminemia and decreased plasma oncotic pressure serve to enhance the formation of edema.

A possible explanation for the variable volume and humoral results obtained in patients with the nephrotic syndrome is that the afferent stimulus may not be attributed to a single mechanism. Specifically, patients with the nephrotic syndrome are heterogeneous with regard to type of renal lesion, GFR, presence of underlying systemic disease, degree of hypoalbuminemia, and diuretic usage. In rat studies, aminonucleoside-induced nephrosis was characterized by a decreased plasma volume, as well as a well-maintained GFR, and edema could be prevented by adrenalectomy. In contrast, nephrotic syndrome induced by nephrotoxic serum was characterized by increased plasma volume and a very low GFR, and edema occurred independently of the adrenal glands. In this regard, the studies of Meltzer et al. (157) also are of note. In 1979, these investigators characterized a group of patients with the nephrotic syndrome associated with volume depletion and stimulation of the renin–angiotensin–aldosterone system and described a second group with low or normal PRA and aldosterone concentrations and hypervolemia. The “hypovolemic” group was characterized by minimal change disease and well-preserved GFRs. These patients fit nicely into the traditional underfill schema depicted in the left panel in Figure 2-7. The “hypervolemic” patients were characterized by having chronic glomerulopathy and reduced GFR (mean, 53 mL/min) in addition to suppressed plasma concentrations of renin and aldosterone, findings consistent with intrarenal mechanisms contributing to the renal sodium and water retention and thus the overfill theory.

Figure 2–8 Plasma norepinephrine secretion and clearance rates in patients with the nephrotic syndrome and normal glomerular filtration rates and in normal control subjects. The findings of increased norepinephrine secretion and normal norepinephrine clearance in the nephrotic syndrome patients are consistent with early activation of the sympathetic nervous system in the nephrotic syndrome. (From Rahman SN, Abraham WT, Van Putten VJ, et al. Increased norepinephrine secretion in patients with the nephrotic syndrome and normal glomerular filtration rates: evidence for primary sympathetic activation. Am J Nephrol. 1993;13:266, with permission of S. Karger AG [Basel].)

Nephron Sites of Sodium Retention in the Nephrotic Syndrome

The nephron site of enhanced renal sodium retention in the nephrotic syndrome has been studied predominantly in animal models of glomerulonephritis. Bernard et al. (160) used micropuncture and clearance methodology to study the nephron site of increased sodium reabsorption in saline-loaded rats with autologous immune complex nephritis. These rats developed heavy proteinuria, hypoalbuminemia, and hypercholesterolemia. Etiopathologic examination of kidneys from these animals revealed slight thickening of basement membranes, uniform finely granular deposits of IgG and complement distributed along the basement membranes of all glomeruli, and electron-dense subepithelial deposits. These findings are similar to those observed in human idiopathic membranous nephropathy. Arterial blood pressure, hematocrit, GFR, and renal plasma flow were comparable in control and experimental animals. Proximal tubular sodium reabsorption was decreased in nephrotic rats as compared with controls (35% vs. 44%, P < 0.05). Absolute sodium reabsorption along the loop of Henle and in the distal convoluted tubule was comparable in nephrotic and control animals. Despite comparable sodium delivery to sites beyond the late distal convoluted tubule, the fractional excretion of sodium was significantly lower in nephrotic (2.2%) than in control (4.0%) animals. From these results, the authors conclude that nephron sites beyond the late distal convoluted tubule are primarily responsible for the enhanced sodium reabsorption seen in this nephrotic model. Alternatively, it remains possible that enhanced sodium reabsorption by deep nephrons not accessible to micropuncture also could contribute to the diminished sodium excretion.

Different results were reported by Kuroda et al. (161) using a rat nephrotoxic serum model of nephrotic syndrome. Proteinuria, hypoalbuminemia, and hypercholesterolemia also occurred in these studies. Histologic examination of the kidneys revealed mild glomerular hypercellularity, widely dilated proximal tubules, diffuse glomerular linear immunofluorescence, and electron-dense subepithelial deposits. In contrast to the study by Bernard et al. (160), these animals were actively retaining sodium. In micropuncture studies, single-nephron GFR was decreased, and the percentage of filtered water reabsorbed before late proximal and distal tubular convolutions was increased in the nephrotic rats.

Two clearance studies have been undertaken in nephrotic patients in an attempt to clarify the nephron site of enhanced sodium reabsorption (162,163). Both of these studies were undertaken in patients with a wide variety of primary renal diseases and GFRs. Usberti et al. (163) measured tubular reabsorption of glucose in 21 patients with glomerulonephritis. Tubular glucose reabsorption was used as a marker for proximal tubular sodium reabsorption. The threshold for glucose reabsorption was reduced in 10 nephrotic patients, suggesting diminished proximal tubular reabsorption. A similar conclusion was reached by Grausz et al. (162) in studies undertaken in five nephrotic patients. Blockade of distal tubular nephron sites of sodium reabsorption with ethacrynic acid and chlorothiazide was used to assess proximal sodium reabsorption in these clearance studies. With this approach, proximal sodium reabsorption was found to be lower in nephrotic patients than in normal and cirrhotic patients. However, the more recent study of Usberti et al. (154) demonstrated increased proximal tubular sodium reabsorption—using the more precise technique of lithium clearance—in nephrotic syndrome patients with low albumin concentrations and blood volumes and elevated PRA.

In summary, it appears from experimental and clinical studies that distal nephron sites are primarily involved in the avid sodium retention of the nephrotic syndrome. However, it is likely that increased proximal tubular sodium reabsorption may also be operative in selected cases, depending on the nature of the underlying renal disease, the blood volume status, and the phase of sodium retention.

Mechanisms of Enhanced Tubular Sodium Reabsorption

Several studies have been undertaken to identify the mechanism underlying the enhanced renal tubular sodium reabsorption in the nephrotic syndrome. Many nephrotic patients with a normal GFR avidly retain sodium, although a reduced GFR frequently is observed in nephrotic patients. Thus, factors in addition to a reduced filtered load of sodium are important in many nephrotic patients. Based on both experimental and clinical studies, however, nephrotic patients with the lowest GFR often demonstrate the greatest degree of sodium retention (157).

Peritubular capillary physical forces (oncotic and hydrostatic pressures) are believed to exert a modulating influence on renal sodium and water reabsorption. This influence is most likely exerted at the level of the proximal convoluted tubule. However, the low filtration fraction, high renal plasma flow, and normal renal vascular resistance frequently observed in nephrotic patients suggest that factors other than peritubular capillary physical forces are responsible for enhanced tubular sodium reabsorption.

The Renin–Angiotensin–Aldosterone System in the Nephrotic Syndrome

A potential role for the renin–angiotensin–aldosterone system in the pathogenesis of nephrotic sodium retention has been studied in detail. Two early experimental studies strongly supported a role for aldosterone in nephrotic edema (164,165). In rats made nephrotic with aminonucleoside, Tobian et al. (164) found an increase in juxtaglomerular cell granularity during sodium retention. Moreover, Kalant and collaborators (165) found that adrenalectomy prevented the sodium retention of aminonucleoside nephrosis. In one study, aminonucleoside was administered into one renal artery. Proteinuria, a reduced GFR, and sodium retention were observed only in the kidney that received aminonucleoside (166).

Several studies have measured components of the renin–angiotensin–aldosterone system in nephrotic humans (153). In general, studies have been carried out in heterogeneous patient populations at a variety of stages during the patients’ illness. A wide range of values varying from very high to very low have been observed. However, PRA values tend to be highest in patients demonstrating characteristics of arterial underfilling and lowest in overfilled patients.

Brown et al. (167) have undertaken clinical studies to examine the physiologic significance of activation of the renin–angiotensin–aldosterone system in sodium-retaining nephrotic patients. Eight of 16 patients had high PRA. Administration of the ACEI captopril to these eight patients did not induce a diuresis despite a significant reduction in plasma aldosterone to normal values. Mean arterial pressure, however, fell in these patients during converting enzyme inhibition. These results suggested that additional factors are responsible for the avid renal sodium retention even in nephrotic patients with high plasma aldosterone. Aldosterone antagonism studies are more definitive, however, because no fall in blood pressure secondary to diminished angiotensin II, as occurs with ACE inhibition, would occur. In this regard, we have demonstrated reversal of the positive sodium balance in patients with the nephrotic syndrome owing to a variety of glomerular diseases treated with the aldosterone antagonist spironolactone (168).

Natriuretic Peptides in the Nephrotic Syndrome

Plasma ANP and BNP concentrations may be elevated in humans and animals with the nephrotic syndrome (169,170); however, the hemodynamic and renal responses to exogenous ANP or BNP have been found to be blunted in experimental nephrosis (170) and in patients with the nephrotic syndrome (169). Perico and Remuzzi (171) have proposed tubular insensitivity to ANP as an initiating factor in the formation of edema in the nephrotic syndrome. According to their hypothesis, renal unresponsiveness to ANP results in distal tubular sodium and water retention with subsequent edema formation. Any ANP cellular resistance may result from a postreceptor, rather than receptor, mechanism, because urinary cGMP responds appropriately to ANP infusion in nephrotic animals (172). Alternatively, this blunted natriuretic response to ANP and BNP may be secondary to neurohormonal activation. The observed sympathetic activation in edematous patients with the nephrotic syndrome supports this possibility (155). Moreover, Koepke and DiBona (106) have shown that renal denervation, which is known to increase distal sodium delivery, reversed the blunted diuretic and natriuretic responses to ANP in a rat model of the nephrotic syndrome.

Other humoral factors (e.g., kinins and prostaglandins) may modulate renal sodium reabsorption in nephrotic patients. Specifically, inhibitors of prostaglandin synthesis have been reported to reduce GFR in patients with the nephrotic syndrome and may precipitate renal insufficiency (173). Thus, prostaglandins may attenuate factors in nephrotic syndrome that decrease GFR and cause sodium retention.

Renal Water Retention in the Nephrotic Syndrome

The nephrotic syndrome is less frequently associated with hyponatremia, in contrast with the two previously described clinical edematous disorders, heart failure and cirrhosis. In fact, serum sodium concentration is usually normal unless it is influenced by vigorous diuretic measures or during an acute water load (174,175). Furthermore, high serum lipid levels may cause pseudohyponatremia in nephrotic patients unless serum sodium concentration is measured by a direct ion-specific electrode. Nevertheless, abnormal water excretion was clearly demonstrated by Gur et al. (176) in six nephrotic children because their solute-free water clearance during water loading was negative as compared with a positive value after remission of their disease. Head-out water immersion induced an increase in solute-free water clearance in patients with the nephrotic syndrome (177,178), and this improvement may have been secondary to the suppression of the nonosmotic release of AVP. Alternatively, water immersion might improve intrarenal hemodynamics, increase the amount of fluid delivered to the distal diluting nephron, and thereby improve water excretion. Plasma AVP concentrations have been found to be elevated in nephrotic subjects (175,179181) and to correlate best with blood volume (181). Water immersion and hyperoncotic albumin infusion reduced plasma levels of AVP and induced a water diuresis in nephrotic patients (179181). Studies by Shapiro et al. (182) have shown a close correlation between decrements in GFR and water excretion during an acute water load in patients with nephrotic syndrome. Therefore, analysis of these studies indicates that the impaired water excretion in nephrotic patients may be related both to intrarenal factors involving a fall in GFR and diminished distal fluid delivery and to extrarenal factors that primarily involve the nonosmotic release of AVP.

In summary, it appears that the effector mechanisms for sodium and water retention in the nephrotic syndrome may involve a fall in GFR and activation of the sympathetic nervous system, the renin–­angiotensin–aldosterone system, and the nonosmotic release of AVP. Enhanced renal tubular sodium and water reabsorption observed in nephrotic patients also may involve a diminution in ANP sensitivity.

Treatment of Edematous Disorders

Given the preceding background discussion of the pathophysiology of sodium and water retention in the edematous disorders, the approach to the treatment of cardiac failure, liver disease, and the nephrotic syndrome is now considered. The general principles for such therapy are described in Table 2-2.


In cardiac failure, cirrhosis, and the nephrotic syndrome, the initiation of sodium and water retention involves the arterial underfilling caused by these diseases. Initial therapeutic attempts should be directed toward treatment of the primary disease. In low-output cardiac failure, the restoration of cardiac output to normal levels abolishes the arterial underfilling and thus the initiating event for renal sodium retention. The use of positive inotropic agents (e.g., digoxin) and afterload-reducing agents (e.g., ACEIs, ARB, arterial vasodilators) to improve cardiac output should be aggressively pursued in heart failure patients. This approach may alleviate the need for inhibiting tubular reabsorption with diuretics, a maneuver that may further decrease cardiac output and worsen the arterial underfilling. In this regard, it should be noted that the clinical practice guidelines for the treatment of heart failure from the Agency for Health Care Policy and Research recommend ACE inhibition as first-line therapy in nonedematous patients with heart failure (183). In the nephrotic syndrome, particularly of the nil disease or lipoid nephrosis variety, administration of corticosteroids may diminish or eliminate the proteinuria and thereby correct the hypoalbuminemia (184). In addition, treatment with ACEIs (185,186) or an ARB (187) has been shown to reduce the urinary protein loss associated with human nephrotic syndrome. In contrast, the administration of albumin solutions is of very little lasting value in the nephrotic syndrome because the concomitant increase in blood volume is associated with increased urinary clearance of albumin, and thus only a transient increase in plasma albumin concentration occurs. In extreme states of hypoalbuminemia, however, an infusion of albumin may be a lifesaving treatment for a hypotensive episode. Albumin solutions also may be of value for patients with cirrhosis, hypoalbuminemia, and edema, particularly when there is evidence of intravascular volume depletion, such as diminished central venous pressure and a fall in orthostatic blood pressure. The administered albumin is excreted less readily in cirrhotic patients because they have no defect in glomerular capillary permeability and frequently have lower levels of GFR. However, a potential complication of such albumin infusions is the resulting increase in portal hypertension with increased bleeding from esophageal varices and the precipitation of hepatic encephalopathy because of the protein load. In some patients with acute alcoholic hepatitis accompanying cirrhosis, corticosteroid therapy may improve liver function in those with elevated bilirubin and prolonged prothrombin times.

Stay updated, free articles. Join our Telegram channel

Dec 22, 2019 | Posted by in NEPHROLOGY | Comments Off on Renal Sodium Excretion, Edematous Disorders, and Diuretic Use

Full access? Get Clinical Tree

Get Clinical Tree app for offline access