Extracellular fluid (ECF) volume is determined by the balance between sodium intake and renal excretion of sodium. Under normal circumstances, wide variations in salt intake lead to parallel changes in renal salt excretion, such that ECF volume is maintained within narrow limits. This relative constancy of ECF volume is achieved by a series of afferent sensing systems, central integrative pathways, and renal and extrarenal effector mechanisms acting in concert to modulate sodium excretion by the kidney.
Extracellular fluid (ECF) volume is determined by the balance between sodium intake and renal excretion of sodium. Under normal circumstances, wide variations in salt intake lead to parallel changes in renal salt excretion, such that ECF volume is maintained within narrow limits. This relative constancy of ECF volume is achieved by a series of afferent sensing systems, central integrative pathways, and renal and extrarenal effector mechanisms acting in concert to modulate sodium excretion by the kidney.
In the major edematous states, effector mechanisms responsible for sodium retention behave in a more-or-less nonsuppressible manner, resulting in either subtle or overt expansion of ECF volume. In some instances, an intrinsic abnormality of the kidney leads to primary retention of sodium, resulting in expansion of ECF volume. In other instances, the kidney retains sodium secondarily as a result of an actual or sensed reduction in effective circulatory volume.
Renal sodium wastage can be defined as the inability of the kidney to conserve sodium to such an extent that continued loss of sodium into the urine leads to contraction of intravascular volume and hypotension. Renal sodium wastage occurs in circumstances where renal sodium transport is pharmacologically interrupted (administration of diuretics), where the integrity of renal tubular function is breached (tubulointerstitial renal disease) or when mineralocorticoid activity or tubular responsiveness are diminished or absent.
Sodium Intake and Sodium Balance
Under normal circumstances, renal excretion of sodium is regulated so that balance is maintained between intake and output, and ECF volume is stabilized. A subject maintained on a normal-sodium diet is in balance when body weight is constant, and sodium intake and output are equal. When the diet is abruptly decreased, a transient negative sodium balance ensues. A slight contraction of ECF volume signals activation of sodium-conserving mechanisms, which lead to decreases in urinary sodium excretion. After a few days, sodium balance is achieved and ECF volume and weight are stabilized, albeit at a lower value. If sodium intake is increased to the previous normal values, transient positive sodium balance leads to expansion of ECF volume, thereby suppressing those mechanisms that enhanced sodium reabsorption. A new steady-state is reached when ECF volume has risen sufficiently so that sodium excretion now equals intake ( Figure 38.1 ). In both directions a steady-state is achieved, whereby sodium intake equals output, while ECF volume is expanded during salt loads and shrunk during salt restriction. The kidney behaves as though ECF volume is the major regulatory element modulating sodium excretion.
The major edematous states – congestive heart failure, cirrhosis of the liver, and nephrotic syndrome – depart strikingly from those constraints. These states are characterized by persistent renal salt retention despite progressive expansion of ECF volume. Unrelenting sodium reabsorption is not the result of diminished sodium intake or even in most cases diminished plasma volume, as dietary salt is adequate and total ECF and plasma volumes are expanded. Renal sodium excretion no longer parallels changes in ECF volume; rather, the kidney behaves as if sensing a persistent low-volume stimulus. Some critical component of ECF volume remains underfilled.
Primary and Secondary Edema
A common feature of the major edematous states is persistent renal salt retention despite progressive expansion of both plasma and ECF volume. Two themes have been proposed to explain the persistent salt retention that characterizes the major edematous states: salt retention may be a primary abnormality of the kidney or a secondary response to some disturbance in circulation.
Primary edema (overflow, overfill, nephritic) refers to expansion of ECF volume and subsequent edema formation consequent to a primary defect in renal sodium excretion. Increased ECF volume and expansion of its subcompartments result in manifestations of a well-filled circulation. Hypertension and increased cardiac output are commonly present. The mechanisms normally elicited in response to an underfilled circulation are suppressed (↓ renin–angiotensin–aldosterone,↓antidiuretic hormone (ADH),↓activity of sympathetic nerves,↓circulating catecholamines). Acute post-streptococcal glomerulonephritis and acute or advanced chronic kidney disease are disorders in which edema formation is primary in origin.
Secondary edema (underfill) results from the response of normal kidneys to actual or sensed underfilling of the circulation. In this form of edema, a primary disturbance within the circulation secondarily triggers renal mechanisms for sodium retention. Those systems that normally serve to defend the circulation are activated (rises in renin–angiotensin–aldosterone, ADH, activity of sympathetic nerves, and circulating catecholamines). The renal response in underfill edema is similar to that in normal subjects placed on a low-salt diet, that is, low fractional excretion of sodium, increased filtration fraction, and prerenal azotemia. Despite these similarities, a number of critical features distinguish these two states: (1) sodium balance is positive in underfill edema while salt-restricted normal subjects are in balance; and (2) administration of salt to sodium-restricted normals transiently expands ECF volume, after which sodium excretion equals intake, whereas in underfill edema, ECF volume expands progressively consequent to unyielding salt retention; and features of an underfilled circulation persist in underfill edema, while the circulation is normalized in normals.
The circulatory compartment that signals persistent activation of sodium-conserving mechanisms in secondary edema is not readily identifiable. Cardiac output may be high (arteriovenous shunts) or low (dilated cardiomyopathy). Similarly, plasma volume may be increased (arteriovenous shunts and heart failure) or decreased (some cases of nephrotic syndrome). The body fluid compartment ultimately responsible for signaling a volume-regulatory reflex leading to renal sodium retention is effective arterial blood volume (EABV). EABV identifies that critical component of arterial blood volume, actual or sensed, that regulates sodium reabsorption by the kidney. In both normal circumstances and the major edematous states, the magnitude of EABV is the major determinant of renal salt and water handling.
Concept of Effective Arterial Blood volume
In order to explain adequately persistent sodium retention in underfill edema, two cardinal features must exist. First, there must be a persistent low-volume stimulus sensed by the kidney that is then translated into persistent, indeed often unrelenting, retention of sodium despite adequate salt intake and overexpansion of ECF volume. Second, there must be a disturbance in those forces that partition retained fluid into the various subcompartments of the ECF space, resulting in an inability to terminate the low-volume stimulus. The first feature can be ascribed to a shrunken EABV, a feature common to all major edematous states. The second feature can be attributed to a disruption in Starling forces, which normally dictate the distribution of fluid within the extracellular compartment. A disturbance in the circulation exists such that retained fluid is unable to restore EABV but rather is sequestered, resulting in edema formation.
Fluctuations in EABV are modulated by two key determinants: (1) filling of the arterial tree (normally determined by venous return and cardiac output); and (2) peripheral resistance (a factor influenced by compliance of the vasculature and degree of arteriolar runoff). A reduction in EABV can be the result of decreased arterial blood volume owing to low cardiac output, as in congestive heart failure. Conversely, EABV can be reduced in the face of increased arterial blood volume when there is excessive peripheral runoff, as seen in arteriovenous shunting and vasodilation. Increased compliance of the arterial vasculature in which arterial blood volume is reduced relative to the holding capacity of the vascular tree results in decreased EABV. For example, administration of salt to a subject with a highly compliant or “slack” circulation (as in pregnancy or cirrhosis) results in a sluggish natriuretic response, in contrast to a high resistance or “tight” circulation (as in primary aldosteronism or accelerated hypertension) in which salt administration causes prompt natriuresis.
Under normal circumstances, EABV is well-correlated with ECF volume. Figure 38.1 depicts the relationship between subcompartments of ECF volume and renal sodium excretion in both normal and edematous states. Under normal circumstances, subcompartments of ECF volume freely communicate in response to changes in dietary sodium, such that expansion or shrinkage of these compartments occurs in concert ( Figure 38.2 , states 1A and 1B). In steady-state conditions, sodium intake and output are in balance; the set point at which balance is attained is dictated by salt intake.
By contrast, major edematous states are characterized by a shrunken EABV, which cannot be filled despite expansion of one or more subcompartments. No longer is EABV well-correlated with total ECF volume and salt intake. Due to a disturbance in the forces that normally partition fluid into the various subcompartments of ECF space, EABV remains contracted even though total ECF volume is greatly expanded. Activation of sodium-conserving mechanisms persists, despite plentiful salt intake. Such derangements in fluid distribution can be categorized as disturbances in Starling forces within the interstitial space, between interstitial space and vascular tree, and disturbances within the circulation. These disturbances are summarized next.
Trapped fluid ( Figure 38.2 , state 2A). In the first type of disturbance, fluid is trapped within a pathologic compartment such that it cannot contribute to effective extracellular volume, that is, volume capable of filling interstitial and vascular spaces. Decrease in effective extracellular volume leads to decreases in total blood volume, arterial blood volume, and EABV, and renal sodium retention is stimulated. Retention of salt and water cannot re-expand effective extracellular volume, as fluid is sequestered into an abnormal fluid compartment behind the “Starling block” within the interstitial space. Such spacing of fluid into inflamed tissue, vesicles and bullae, peritonitis, necrotizing pancreatitis, rhabdomyolysis, and burns functionally behaves as if lost from the body.
Reduced oncotic pressure. A reduction in the circulating level of albumin can lead to a second type of fluid maldistribution. Decreased plasma oncotic pressure allows fluid to translocate from the vascular compartment to the interstitial space. Reductions in total blood volume, arterial blood volume, and EABV lead to sodium retention. The retained salt and water, owing to a “Starling block” across the capillary bed, leaks into the interstitial space.
Vascular disturbances ( Figure 38.2 , states 2B and 2C). A third type of fluid distributory disturbance results from abnormalities within the circulation, and can be of two types. The prototypical example of the first type is congestive heart failure. A failing ventricle results in decreased cardiac output and high diastolic intraventricular pressures. Venous return is impeded, with consequent reductions in arterial blood volume and EABV. Sodium retention is stimulated, but arterial blood volume and EABV remain contracted due to a circulatory block across the heart. In consequence, venous volume expands and leads to transudation of fluid into the interstitial space. The second type of circulatory abnormality that leads to fluid maldistribution is exemplified by arteriovenous shunting (e.g., Paget’s disease, beriberi, thyrotoxicosis, anemia, cirrhosis). Widespread shunting through multiple small arteriovenous communications results in increased venous return, thereby augmenting cardiac output and arterial filling. However, arterial runoff and vasodilation lead to underperfusion of some critical area in the microcirculation. The circulatory block lies between the arterial blood volume and EABV.
What distinguishes secondary edematous states from the normal circumstance is an inability to expand EABV owing to Starling or circulatory blocks within the extracellular space. Normally, the system of volume regulation behaves as an open system, such that fluctuations in one compartment are quickly translated into parallel changes in other compartments; total ECF volume and EABV are closely related. In contrast, volume regulation in underfill edema can be regarded as clamped; EABV remains shrunken despite expansion of the subcompartments of the extracellular space. EABV becomes dissociated from total ECF volume; salt retention becomes unrelenting and salt administration cannot re-expand the contracted EABV.
The reader is referred to the fourth edition of this book, in which a comprehensive discussion is provided on the afferent and efferent mechanisms involved in the regulation of extracellular fluid volume under normal circumstances. An overview of renal sodium handling in each segment of the nephron was also provided in that discussion. In this edition, the chapter will focus exclusively on the pathophysiology of the major edematous and salt-wasting states.
Congestive Heart Failure
The fundamental abnormality underlying congestive heart failure is an inability of the heart to maintain its function as a pump. As a result, a series of complex compensatory reflexes are initiated that serve to defend the circulation. The renal response to a failing myocardium is retention of salt and water resulting in expansion of ECF volume. If myocardial dysfunction is mild, expansion of ECF volume leads to increased left ventricular end-diastolic volume, which raises cardiac output according to the dictates of the Frank–Starling principle. In this state of compensated congestive heart failure, salt intake and output come into balance, but at the expense of an expanded ECF volume. Further deterioration in ventricular function leads to further renal retention of salt and water. There is progressive expansion of ECF volume and features of a congested circulation become manifest: peripheral edema, engorged neck veins, and pulmonary edema. Despite massive overexpansion of ECF volume, the kidneys behave as though they were responding to a low-volume stimulus. In subsequent sections, a detailed analysis of the afferent and efferent regulatory limbs in congestive heart failure will be provided.
Afferent Sensing Mechanisms in Congestive Heart Failure
Low and High Pressure Baroreceptors
A characteristic feature in many forms of congestive heart failure is increased stretch and transmural pressure within the cardiac atria. These alterations would normally provide afferent signals that suppress sympathetic outflow and decrease the release of renin and ADH, and ultimately result in a diuretic and natriuretic response. In congestive heart failure, this afferent signaling mechanism is markedly perturbed. Despite the presence of venous congestion and elevated cardiac filling pressures, sympathetic nervous activity and serum concentrations of renin and ADH are increased and urinary salt excretion is blunted. Both clinical and experimental studies are consistent with a decrease in sensitivity of the pressure-sensitive receptors in the cardiac atria.
Increased renal sympathetic nerve activity in cardiac failure has also been attributed to impaired arterial baroreceptor function. High pressure baroreceptors in the carotid sinus and aortic arch normally exert a tonic inhibitory effect on central nervous system sympathetic outflow. Although the precise mechanism for the sympathoexcitation is not known, a sustained reduction in arterial pressure is unlikely to be the sole explanation, since arterial pressure is usually normal in congestive heart failure. Rather, sympathetic tone becomes insensitive to manipulations that normally suppress or enhance its activity. For example, infusion of nitroprusside increases both the heart rate and the circulating norepinephrine levels in normal subjects, whereas equivalent hypotensive doses in subjects with congestive heart failure elicit a blunted response. Similarly, patients with heart failure show less bradycardia when arterial pressure is raised by infusion of phenylephrine. Such alterations in baroreflex function may result from abnormalities peripherally or alterations in central autonomic regulatory centers.
Several observations suggest angiotensin II may contribute to the depressed baroreflex sensitivity in heart failure. Angiotensin II has been shown to upwardly reset the arterial baroreflex control of heart rate in the rabbit, independent of a change in arterial pressure. In the rat, increased levels of endogenous angiotensin II produced by changes in dietary salt intake tonically increase the basal level of renal sympathetic nerve activity, and upwardly reset the arterial baroreflex control of renal sympathetic nerve activity. In experimental models, administration of an angiotensin II receptor blocker can reverse these changes and improve the sensitivity of the arterial baroreflex mechanism. Interestingly, captopril administered to patients with congestive heart failure restores the normal hemodynamic response to postural tilt and infusion of vasoconstrictive agents.
A reduction in cardiac output has been suggested as the afferent signal that leads to Na retention in heart failure. When cardiac output is reduced by constriction of the abdominal or thoracic vena cava, urinary sodium excretion is typically decreased. Restoring cardiac output to normal by autologous blood transfusion ameliorates renal salt retention, despite persistently elevated venous and hepatic pressures. By contrast, rats with small-to-moderate myocardial infarctions have normal capacities to increase cardiac output in response to volume loads, and yet renal sodium excretion remains blunted in these animals. Even when cardiac output is increased above normal, as with the creation of an arteriovenous fistula in dogs, clinical findings of ascites and peripheral and pulmonary edema develop. Despite increased cardiac output, levels of renin, aldosterone, and ANP are high. Thus, the signal which initiates renal salt retention in congestive heart failure cannot originate solely from a decrease in cardiac output.
Other afferent sensing mechanisms potentially active in congestive heart failure include intrahepatic baroreceptors and mechanoreceptors within the kidney. Chemosensitive receptors that respond to changing levels of metabolic breakdown products may participate in sensing of ECF volume. One such sensing mechanism may relate to the cardiac sympathetic afferent reflex. The reflex begins with sympathetic afferent fibers that respond to changes in cardiac pressure and dimension or substances that may accumulate in ischemia or heart failure. The reflex is excitatory in nature, such that activation of the afferent fibers leads to increased central sympathetic outflow. In summary, a contracted EABV serves as the afferent signal that elicits activation of effector mechanisms resulting in sodium retention. As with other edematous disorders, the exact volume compartment that comprises EABV has not been elucidated ( Figure 38.3 ).
Effector Mechanisms in Congestive Heart Failure
Nephron Sites of Renal Sodium Retention
Renal sodium handling in the setting of congestive heart failure is similar to that which occurs in an otherwise normal individual who is volume-depleted. Activation of effector mechanisms lead to alterations in renal hemodynamics and tubular transport mechanisms that culminate in renal salt retention.
Renal hemodynamics in congestive heart failure are characterized by reduced renal plasma flow and a well-preserved glomerular filtration rate, such that filtration fraction is typically increased. In a rat model of myocardial infarction, Hostetter et al. found a positive correlation between the decline in renal plasma flow and the degree to which left ventricular function was impaired. The glomerular filtration rate remained well-preserved as a result of an increased filtration fraction, except in animals with a severely compromised left ventricle. When examined at the level of the single nephron, these hemodynamic changes were found to be the result of a disproportionate increase in efferent arteriolar vasoconstriction and increased glomerular capillary hydraulic pressure. Treatment with an angiotensin-converting enzyme inhibitor caused a decline in filtration fraction and efferent arteriolar resistance, suggesting an important role for angiotensin II in mediating efferent arteriolar constriction.
Changes in glomerular and proximal tubular function in heart failure are similar to those that result from infusion of angiotensin II or norepinephrine. Angiotensin II, catecholamines, and renal nerves are all capable of increasing both the afferent and the efferent arteriolar tone, but predominantly act on the latter. These changes serve to maintain glomerular filtration rate near normal as renal plasma flow declines secondary to impaired cardiac function. As cardiac function progressively declines and the reduction in renal plasma flow becomes severe, the glomerular filtration rate will begin to fall. At this point there is an inadequate rise in filtration fraction, because efferent arteriolar vasoconstriction can no longer offset the intense afferent arteriolar constriction. Higher plasma catecholamines and further increases in sympathetic nerve activity acting to provide circulatory stability result in greater constriction of the afferent arteriole, such that glomerular plasma flow and transcapillary hydraulic pressure are reduced. In this setting, the glomerular filtration rate becomes dependent on afferent arteriolar flow.
These observations are similar to what has been observed in human subjects with varying degrees of left ventricular function. As left ventricular function declines, the glomerular filtration rate is initially maintained by an increased filtration fraction. However, in patients with severely depressed left ventricular function, a progressive decline in renal blood flow becomes associated with a fall in glomerular filtration rate due to an inadequate rise in filtration fraction. In this setting, administration of an ACE inhibitor can result in a further lowering of the glomerular filtration rate, even though systemic arterial pressure remains fairly constant.
Both experimental and clinical studies support the proximal nephron as a major site of increased sodium reabsorption in the setting of congestive heart failure. In human subjects, clearance techniques have primarily been employed to demonstrate the contribution of the proximal nephron. For example, infusion of mannitol was shown to increase free water excretion to a greater extent in patients with congestive heart failure as compared to normal controls. Since mannitol inhibits fluid reabsorption proximal to the diluting segment, it was inferred that enhanced free water clearance was reflective of augmented delivery of Na from the proximal tubule to the diluting segment. In dogs with an arteriovenous fistula, there is a failure to escape from the Na-retaining effects of deoxycorticosterone acetate. In addition, these animals do not develop hypokalemia in contrast to normal controls. The absence of hypokalemia in the setting of mineralocorticoid excess is best explained by decreased delivery of Na to the distal nephron due to enhanced proximal Na reabsorption. Alterations in peritubular hydrostatic and oncotic forces, as well as direct effects of various neurohormonal effectors, account for enhanced proximal sodium and water absorption in this setting.
Clearance and micropuncture studies are also consistent with enhanced sodium reabsoption in the distal tubule in states of congestive heart failure. The loop of Henle has been identified as a site of enhanced sodium reabsoption in dogs with chronic vena cava obstruction and rats with an arteriovenous fistula.
The renin–angiotensin–aldosterone system is activated when the heart fails as a pump. Components of this system serve to compensate for decreased cardiac output by stabilizing the circulation and expanding ECF volume.
Several mechanisms are activated in the setting of a failing myocardium which serve to increased renin release. Diminished pressure in the afferent arteriole enhances renin release via a baroreceptor mechanism, the sensitivity of which is heightened consequent to augmented baseline sympathetic nerve activity. Enhanced salt and water reabsorption in the proximal tubule and loop of Henle diminishes sodium chloride concentration at the macula densa, providing a stimulatory signal for renin release by way of the tubuloglomerular feedback mechanism. Finally, increased sympathetic nerve activity directly enhances renin release via stimulation of β-adrenergic receptors on the juxtaglomerular cells.
Renin acts on angiotensinogen synthesized in the liver and elsewhere to produce the decapeptide, angiotensin I. Angiotensin I is converted to angiotensin II by the angiotensin-converting enzyme present in the lungs, kidney, and blood vessels throughout the circulation.
Angiotensin II plays a pivotal role in glomerular and proximal tubule function in models of congestive heart failure ( Figure 38.4 ). By selectively increasing efferent arteriolar tone, adjustments in the glomerular and postglomerular circulatory network favor net reabsorption in the proximal tubule. Increased filtration fraction leads to increased peritubular oncotic pressure, and in combination with decreased peritubular hydrostatic pressure net sodium reabsorption is enhanced. Angiotensin II also stimulates salt and water reabsorption through a direct effect on proximal tubular cells. Increased efferent arteriolar resistance increases glomerular capillary hydrostatic pressure, mitigating any fall in GFR that would otherwise occur from decreased renal blood flow. In clinical, as well as experimental, models of heart failure, administration of ACE inhibitors improves renal blood flow and increases urinary sodium excretion, consistent with important angiotensin II-mediated effects on the renal microvasculature.
Angiotensin II also contributes to renal salt and water retention through effects mediated by increased renal sympathetic nerve activity. As previously mentioned, angiotensin II decreases the sensitivity of the baroreflex mechanism such that a higher pressure is required to decrease central sympathetic outflow. In addition, angiotensin II directly stimulates sympathetic outflow at the level of the central nervous system. Chronic administration of an ACE inhibitor to patients with congestive heart failure reduces central sympathetic outflow and improves the sympathoinhibitory response to baroreflex stimulation.
Angiotensin II also influences renal salt and water handling in the distal nephron, primarily through stimulatory effects on aldosterone release in the adrenal gland. Aldosterone acts primarily on the collecting duct to promote tubular reabsorption of sodium. Aldosterone-stimulated sodium reabsorption generates a luminal-negative voltage that secondarily enhances excretion of hydrogen and potassium ions. The magnitude of potassium secretion will depend on the volume and composition of filtrate reaching the collecting duct. In this regard, patients with heart failure rarely manifest hypokalemia and alkalosis, despite oversecretion of mineralocorticoid, unless distal sodium delivery is increased by use of a diuretic. In the absence of diuretic therapy, distal delivery of sodium is low due to enhanced proximal reabsorption mediated by angiotensin II, sympathetic nerves, and peritubular physical factors. Thus, although plasma renin and aldosterone levels are frequently elevated in heart failure, there is conflicting data as to the importance of aldosterone in mediating renal salt retention.
Conflicting data regarding the importance of the renin–angiotensin–aldosterone system in the generation of cardiac edema is best resolved when analyzed with respect to severity of heart failure. The initial response to constriction of the pulmonary artery or thoracic inferior vena cava in dogs is a reduction in blood pressure, and increases in renin and aldosterone levels. During this acute phase there is avid renal sodium retention, and stability of blood pressure is critically dependent on circulating angiotensin II. Over several days plasma volume and body weight increase, while renin, aldosterone, and sodium balance return to control values. During the acute phase, administration of a converting enzyme inhibitor results in hypotension, while no effect on blood pressure is observed during this chronic phase. If plasma renin and aldosterone fail to decrease due to severe impairment of cardiac output, then converting enzyme inhibitor-induced hypotension persists.
A similar pattern is seen in dogs with an arteriovenous fistula. In the early phase of this high-output cardiac failure model, significant elevations in renin and aldosterone levels occur, and renal sodium retention is marked. Several days later, after development of peripheral edema and ascites, renin and aldosterone levels return to baseline and daily sodium excretion begins to match dietary intake.
A similar relationship between the renin–angiotensin–aldosterone system and stage and severity of congestive heart failure exists in humans. This relationship may explain why renal function improves in some patients treated with ACE inhibitors, whereas renal function deteriorates in others. In subjects whose renal function worsens after administration of the drug, there is a greater fall in mean right atrial pressure, left ventricular filling pressure, mean arterial pressure, and systemic vascular resistance as compared to subjects with stable renal function. In addition, plasma renin activity increases to a greater extent. These changes suggest a more contracted EABV and greater dependency of systemic vascular resistance on circulating angiotensin II in patients with ACE inhibitor-induced renal dysfunction.
In summary, during severe decompensated left ventricular failure, decreased EABV elicits release of renin with consequent activation of angiotensin II and aldosterone. Acutely, increased circulating levels of angiotensin II serve to maintain systemic blood pressure and augment renal sodium reabsorption. Salt retention is the result of hemodynamic and direct effects of angiotensin II at the level of the proximal tubule, and enhanced sodium reabsorption in the distal nephron primarily mediated by increased aldosterone. As ECF volume expands, renin, angiotensin II, and aldosterone become suppressed, although not necessarily to normal levels. Maintenance of systemic blood pressure is more dependent on volume rather than angiotensin II. Sodium balance is now achieved, but at the expense of increased steady-state ECF volume. ACE inhibitor therapy is not associated with deleterious effects on renal function at this stage of the disease. Should further deterioration in cardiac function ensue, persistent activation of the renin–angiotensin–aldosterone system may result, such that systemic blood pressure remains dependent on circulating angiotensin II despite expansion of ECF volume. In this setting, ACE inhibitor therapy can precipitate hypotension and significant reductions in the glomerular filtration rate. One has to consider this sequential change in renin to volume dependency of mean arterial blood pressure in attempting to predict net renal and hemodynamic effects of converting enzyme inhibition.
Sympathetic Nervous System
The sympathetic nervous system is activated in congestive heart failure. Plasma norepinephrine levels are elevated, and concentrations correlate with the degree of left ventricular dysfunction. Direct nerve recordings demonstrate a direct relationship between central sympathetic nerve outflow and left ventricular filling pressures.
Increased sympathetic tone influences renal reabsorption of salt and water by indirect, as well as direct, mechanisms ( Figure 38.5 ). Glomerular hemodynamics are affected similarly to that produced by angiotensin II. In addition, sympathetic nerves directly stimulate tubular reabsorption of salt and water in both the proximal and the distal nephron.
Increased sympathetic nerve activity stimulates renin release. The subsequent formation of angiotensin II provides a positive feedback loop leading to further increases in sympathetic nerve activity. Angiotensin II sensitizes tissues to the actions of catecholamines, and acts synergistically with renal nerves in modulating renal blood flow.
Increased circulating levels of AVP is a characteristic finding in patients with congestive heart failure. The nonosmotic release of AVP plays an important role in the development of hyponatremia, which in turn is a well-defined predictor of mortality in heart failure patients. In experimental heart failure, there is upregulation of the mRNA for vasopressin in the hypothalamus. In addition, there is increased expression of the mRNA and the protein for the aquaporin-2 water channel. In a rat model of heart failure, selective antagonism of the V-2 receptor is associated with a significant improvement in free water clearance. Administration of a V-2 receptor antagonist to patients with congestive heart failure is associated with a significant reduction in body weight and improvement in dyspnea, but has not been shown to reduce mortality.
Increased production of prostaglandins plays an important role in maintaining circulatory homeostasis in congestive heart failure. In response to decreases in cardiac output, neurohumoral vasoconstrictor forces (i.e., the renin–angiotensin–aldosterone system, the neurosympathoadrenal axis) participate in the maintenance of systemic arterial pressure, and result in increased total peripheral vascular resistance. These same vasoconstrictors stimulate the renal production of vasodilatory prostaglandins, such that the rise in renal vascular resistance is less than that seen in the periphery. Vasodilatory prostaglandins function in a counter-regulatory role, attenuating the fall in renal blood flow and glomerular filtration rate that would otherwise occur if vasoconstrictor forces were left unopposed.
Renal prostaglandins also serve to moderate salt and water retention that would otherwise occur in the setting of unopposed activation of effector mechanisms such as angiotensin II, aldosterone, renal sympathetic nerves, and ADH. The importance of prostaglandins in modulating renal hemodynamics, sodium excretion, and circulatory homeostasis progressively increases in proportion to the severity of the heart failure ( Figure 38.6 ).
Circulating atrial natriuretic peptide (ANP) and brain natriuretic peptide (BNP) are circulating hormones which are primarily synthesized in the cardiac atria and ventricles respectively. The synthesis and release of these peptides provide a mechanism whereby cardiac atria and ventricles serve both an afferent and efferent function in the control of ECF volume. Levels of ANP are elevated and correlate with the severity of disease in humans and experimental animals with heart failure.
The natriuretic and vasodilatory properties of ANP suggest that this peptide plays an important counter-regulatory role in congestive heart failure. However, attempts to use ANP therapeutically in congestive heart failure have produced disappointing results. ANP infused in patients with heart failure causes only a minimal change in fractional sodium excretion and urine flow rates, as compared to the robust response in normal controls. The mechanism of renal nonresponsiveness in heart failure is not entirely clear. A downregulation of receptors due to sustained exposure to high levels of ANP or altered intrarenal hemodynamics are possibilities. Decreased delivery of sodium to the distal nephron where ANP normally exerts its natriuretic effect is also a likely cause of resistance. While ANP levels are uniformly elevated in congestive heart failure, potentially beneficial natriuretic properties are overwhelmed by more powerful antinatriuretic effector mechanisms.
Similar to ANP, plasma levels of BNP are elevated in congestive heart failure, and in proportion to the severity of systolic and diastolic dysfunction. Infusion of BNP is associated with a significant reduction in pulmonary capillary wedge pressure, pulmonary artery pressure, right atrial pressure, and mean arterial pressure, as well as an increase in cardiac index. These hemodynamic benefits are accompanied by significant increases in urinary volume and sodium excretion, in some but not all studies. Infusion of BNP can be associated with hypotension, particularly when given with other vasodilators. Measurement of plasma BNP levels is often utilized as a diagnostic tool to differentiate between cardiac versus other causes of pulmonary congestion. In addition, BNP levels can be used as a prognostic indicator and a marker reflecting the degree of cardiac dysfunction.
Endothelin and Nitric Oxide
Circulating levels of endothelin are increased in congestive heart failure, and correlate positively with the degree of myocardial dysfunction. Studies in which endothelin antagonists have been administered suggest this substance may play a role in the pathophysiology of cardiac failure. In a randomized, double-blind study of human subjects with heart failure, infusion of an ETa and ETb receptor blocker (bosentan) was associated with a reduction in right atrial pressure, pulmonary artery pressure, pulmonary capillary wedge pressure, and mean arterial pressure. In a dog model of heart failure, ETa blockade alone lead to a reduction in cardiac filling pressures and increased cardiac output. These hemodynamic changes were associated with an increase in GFR and renal plasma flow, as well as increased urinary sodium excretion. By contrast, administration of an ETb receptor blocker caused an increase in cardiac filling pressures and a decrease in cardiac output, suggesting endogenous endothelins adversely effect cardiac hemodynamics and cause fluid retention, primarily through ETa receptors.
Nitric oxide production is increased in congestive heart failure. Increased release of nitric oxide from resistance vessels may partly antagonize neurohumoral vasoconstrictor forces. Inhibiting nitric oxide production in heart failure patients causes a significant increase in pulmonary and systemic vascular resistance, as well as a decline in cardiac output. In the renal vasculature, nitric oxide production is also increased; however, the renal vasodilatory response to nitric oxide is impaired. Administration of an angiotensin receptor antagonist restores nitric oxide-mediated renal vasodilation, suggesting angiotensin II plays a contributory role in this defect.
Renal sodium excretion is normally regulated so that extracellular fluid (ECF) volume is maintained within normal limits. Any maneuver that increases ECF volume will lead to a prompt and sustained natriuresis until the volume returns to normal. In patients with cirrhosis, this homeostatic mechanism becomes deranged such that large increases in ECF volume are accompanied by continued renal salt retention, resulting in edema and ascites formation.
Presinusoidal Versus Postsinusoidal Obstruction and Ascites Formation
In patients with cirrhosis, the kidneys are normal but are signaled to retain salt in an unrelenting manner. The critical event in the generation of this signal is development of hepatic venous outflow obstruction. In the normal state, the portal circulation is characterized by high flow, low pressure, and low resistance. The imposition of a resistance into this high-flow vasculature will uniformly raise portal pressure, but development of ascites is critically dependent on location of the resistance. Conditions associated with presinusoidal vascular obstruction, such as portal vein thrombosis and schistosomiasis, raise portal pressure but are not generally associated with ascites. By contrast, hepatic diseases such as Laennec’s cirrhosis and Budd Chiari syndrome cause early postsinusoidal vascular obstruction, and are associated with marked degrees of salt retention, anasarca, and ascites. Thus, during the development of the cirrhotic process, ascites will accumulate primarily when the pathologic process is associated with hepatic venous outflow obstruction and sinusoidal hypertension.
This distinction between presinusoidal and postsinusoidal obstruction can best be explained by comparing the characteristics of fluid exchange in capillaries of the splanchnic bed versus those in the hepatic sinusoids. The intestinal capillaries are similar to those in the peripheral tissues, in that they have continuous membranes with small pores such that a barrier exists, preventing plasma proteins from moving into the interstitial space. An increase in capillary hydrostatic pressure will cause the movement of a protein-poor fluid to enter the interstitial compartment and decrease the interstitial protein concentration. Interstitial protein concentration is further reduced by an acceleration in lymph flow that is stimulated by the fluid movement. As a result, the interstitial oncotic pressure falls, and the plasma oncotic pressure remains unchanged. The net oncotic force therefore rises and offsets the increase in hydrostatic force, providing a buffer against excessive fluid filtration. The fall in intestinal lymph protein concentration is maximal at relatively low pressures, and is much greater than that observed from the cirrhotic liver. Thus, the increase in net oncotic force associated with dilution of the interstitial protein and accelerated lymph flow contribute to the protection against ascites in patients whose only abnormality is portal hypertension.
The situation across the liver sinusoids is quite different. Hepatic sinusoids, unlike capillaries elsewhere in the body, are extremely permeable to protein. As a result, colloid osmotic pressure exerts little influence on movement of fluid. Rather, direction of fluid movement is determined almost entirely by changes in sinusoidal hydraulic pressure. Thus, efflux of protein-rich filtrate into the space of Disse is critically dependent on hepatic venous pressures. Obstruction to hepatic venous outflow will lead to large increments in the formation of hepatic lymph and flow through the thoracic duct. Unlike the intestinal capillaries, there is little or no restriction in the movement of protein into the interstitium, such that the protein concentration of hepatic lymph will quickly approach that of plasma. As a result, no significant oncotic gradient develops between plasma and the interstitium at high sinusoidal pressures and flow.
When sinusoidal pressure increases to such a degree that hepatic lymph formation exceeds the capacity of the thoracic duct to return fluid to the circulation, interstitial fluid weeps off the liver into the peritoneal space and forms ascites. Lymph formation in the setting of cirrhosis can be more than 20-fold greater than that which occurs under normal circumstances. Whereas in normal humans 1–1.5 liters/day of lymph are returned to the circulation, subjects with cirrhosis, even without ascites, may have lymph flow through the thoracic duct as high as 15–20 liters/day. The predominance of hepatically-produced lymph to overall lymph production is illustrated by studies in experimental animals with cirrhosis. Barrowman and Granger found a 29-fold increase in hepatic lymph flow, while only a three-fold increase was noted in the splanchnic lymphatics. Eleven of 19 animals had normal flows of intestinal lymph, while all the cirrhotic animals had increased flows in liver lymph.
Conditions associated with the rapid onset of postsinusoidal obstruction, such as acute right-sided congestive heart failure and Budd–Chiari syndrome, initially give rise to ascitic fluid that has a high protein concentration that may even approach that of plasma. This high protein concentration is reflective of the liver being the predominant source of the ascitic fluid. However, over time the protein content of ascites in these conditions begins to decrease. Witte et al. measured the total protein in ascitic, pleural, and peripheral edema fluid in acute and chronic heart failure patients. In the setting of acute heart failure, the mean concentration of protein in ascitic fluid was approximately 5 g/dl. By contrast, the protein concentration in ascitic fluid of chronic congestive heart failure patients was 2.7 g/dl. A lower protein concentration is also typical of conditions such as Laennec’s cirrhosis, in which postsinusoidal obstruction develops slowly.
Two phenomena contribute to this change in ascitic fluid protein concentration. If the hepatic sinusoids are subject to an increased hydrostatic pressure for a long period of time, they begin to assume the anatomic and functional characteristics of capillaries found elsewhere in the body, a process referred to as capillarization. This change leads to a decrease in albumin permeability, such that oncotic forces begin to play some role in hepatic lymph formation. At the same time, hypoalbuminemia develops secondary to decreased hepatic synthesis, as well as dilution secondary to ECF volume expansion. As a result, the protein content of hepatic lymph, although still high, falls to approximately 50–55% of plasma values.
The second factor contributing to the lower ascitic protein concentration is the superimposition of portal hypertension. Early in the development of portal hypertension when plasma protein concentration is normal only minimal amounts of ascitic fluid are derived from the splanchnic bed, due to the buffering effect of increased net oncotic force opposing fluid filtration. Extremely high hydrostatic pressures are required to produce significant amounts of ascitic fluid in the setting of normal plasma protein concentrations. By contrast, less and less hydrostatic pressure is required for the formation of ascitic fluid, as the plasma albumin concentration decreases and the net osmotic force declines. In this setting, the splanchnic bed begins to make a greater contribution to the generation of ascites, and the fluid is characterized by a low protein concentration.
The development of portal hypertension is also associated with changes in the splanchnic circulation that secondarily lead to increased lymph production in the splanchnic bed. The importance of the splanchnic lymphatic pool in the generation of ascites is reflected by the fact that in most instances ascitic fluid is transudative and characterized by a protein concentration of approximately 2.5 g/dl. Classically, portal hypertension was considered to be the sole result of increased resistance to portal venous flow. However, studies in experimental models suggest that increased portal venous flow resulting from generalized splanchnic arteriolar vasodilation also plays a role in the genesis of increased portal pressure. This vasodilation leads to changes in the splanchnic microcirculation that may predispose to increased filtration of fluid. For example, an acute elevation of venous pressure in the intestine normally elicits a myogenic response that leads to a reduction in blood flow. This decrease in flow is thought to serve a protective role against the development of bowel edema. However, in chronic portal hypertension this myogenic response is no longer present. In this setting, arteriolar resistance is reduced, such that capillary pressure and filtration are increased. The loss of this autoregulatory mechanism may account for the greater increase in intestinal capillary pressure and lymph flow seen under conditions of chronic portal hypertension when compared to acute increases in portal pressure of the same magnitude. The potential causes of splanchnic arteriolar vasodilation are discussed below.
The importance of portal hypertension in the pathogenesis of ascites is highlighted by several observations. First, patients with ascites have significantly higher portal pressures as compared to those without ascites. Although the threshold for ascites development is not clearly defined, it is unusual for ascites to develop with a pressure below 12 mmHg. Gines found that only 4 of 99 cirrhotic patients with ascites had a portal pressure <12 mmHg, as estimated by hepatic venous wedged pressure. Second, portal pressure correlates inversely with urinary sodium excretion. Third, maneuvers designed to reduce portal pressure are known to have a favorable effect on the development of ascites. For example, surgical portosystemic shunts used in the treatment of variceal bleeding reduce portal pressure, and are associated with a lower probability of developing ascites during follow-up. Both side-to-side and end-to-side portocaval anastomosis have been shown effective in the management of refractory ascites in cirrhosis. Recent studies also suggest that reducing portal pressure with a transjugular intrahepatic portasystemic shunt has a beneficial effect on ascites.
In summary, ascites develops when the production of lymph from either or both the hepatic sinusoids and the splanchnic circulation exceeds the transport capacity of the lymphatics. In this setting, fluid will begin to weep from the surface of the liver and the splanchnic capillary bed, and accumulate as ascites. The final protein concentration measured in the peritoneal fluid is determined by the sum of the two contributing pools of fluid; one relatively high in protein originating in the liver and the other, a low protein filtered across splanchnic capillaries. Hepatic venous outflow obstruction leading to increased sinusoidal pressure and portal hypertension are the major determinants of whether lymph production will be of a magnitude sufficient for ascitic fluid to accumulate. Increased sinusoidal pressure is also related to the subsequent development of renal salt retention. The mechanism by which sinusoidal hypertension signals the kidney to retain sodium is discussed in the following section.
Afferent Limb of Sodium Retention: Overfill Versus Underfill Mechanisms
Classical Underfill Mechanism for Renal Salt Retention
The mechanism by which hepatic venous outflow obstruction leads to sufficiently high sinusoidal pressures for ascites formation is controversial. The classical (underfill) theory predicts that the degree of hepatic venous outflow obstruction is sufficient in the presence of normal splanchnic perfusion to perturb the balance between rates of hepatic lymph formation and thoracic duct flow, thereby resulting in the formation of ascites. Both increased sinusoidal and portal venous pressures, in conjunction with hypoalbuminemia, cause formation of ascites in the presence of normal splanchnic perfusion. The formation of ascites, however, occurs at the expense of decreased intravascular volume. In consequence, a low venous filling pressure and a low cardiac output activate baroreceptor mechanisms, resulting in renal salt retention. According to this formulation, development of ascites is the primary event that leads to an underfilled circulation and subsequent renal salt retention.
The failure of measured hemodynamic parameters to satisfy predictions of the classical theory has raised questions regarding its validity. As originally conceived, it was predicted that extrasplanchnic plasma volume would be decreased, and that cardiac output would be low. When measured, however, these values have rarely been low. In fact, measurements have indicated that total plasma volume is usually elevated in cirrhotic patients. Similarly, cardiac output is rarely low, but tends to vary from normal to very high. In addition, studies performed in animal models of cirrhosis have found that sodium retention precedes the formation of ascites, suggesting that salt retention is a cause and not a consequence of ascites formation.
Overfill Mechanism for Renal Salt Retention
The incompatibility of measured hemodynamic parameters and timing of renal salt retention with the classical theory of ascites has led others to propose the overflow theory. Once again, hepatic disease with venous outflow obstruction is viewed as a prerequisite for development of increased sinusoidal and portal pressures. In contrast to the classical theory, however, normal splanchnic perfusion fails to raise sinusoidal pressure sufficiently to cause ascites formation. Rather, venous outflow obstruction signals renal sodium retention independent of diminished intravascular volume. Salt retention, in turn, increases plasma volume, cardiac output, and splanchnic perfusion, thus raising sinusoidal and portal pressures sufficiently to culminate in translocation of fluid into the interstitial space and eventually the peritoneum. The combination of portal hypertension and increased arterial volume would lead to overflow ascites formation. This hypothesis is supported by the positive correlation between plasma volume and hepatic venous pressure, and the persistence of increased plasma volume after portacaval anastomosis. Moreover, patients with ascites have significantly higher portal pressure than patients without ascites, and portal pressure correlates inversely with urinary sodium excretion.
Additional evidence linking hepatic venous outflow obstruction directly to renal sodium retention comes from studies performed in dogs fed the potent hepatotoxin dimethylnitrosamine. The pathophysiologic disturbances and histologic changes that develop over a 6–8-week period are similar in nature to those seen in Laennec’s cirrhosis. In this model, sodium retention and increases in plasma volume precede formation of ascites by about 10 days. In order to exclude the possibility that the increase in plasma volume was solely due to an increased splanchnic plasma volume, repeat measurements were obtained after ligation of the superior and inferior mesenteric arteries, the celiac axis, and portal vein. In this way, any contribution of the splanchnic circulation could be excluded. These studies clearly showed that extrasplanchnic plasma volume was elevated at a time when dogs were in positive sodium balance. To further prove that extrasplanchnic plasma volume was increased, end-to-side portacaval shunts were placed prior to inducing cirrhosis. This maneuver was designed to prevent any increase in splanchnic plasma volume. In these studies, evidence of salt retention preceded the formation of ascites, and was accompanied by a parallel increase in plasma volume.
In another series of studies using this same model, hemodynamic parameters were monitored during control, precirrhotic, and postcirrhotic sodium balance periods. Sodium retention was found to precede any detectable change in cardiac output or peripheral vascular resistance. Once ascites developed, plasma volume increased further, and this was associated with increased cardiac output and a fall in peripheral vascular resistance. It was concluded that initiation of sodium retention and plasma volume expansion was not dependent on alterations in systemic hemodynamics. This conclusion has been corroborated in the canine model of hepatic cirrhosis induced by bile duct ligation, as well as in rats made cirrhotic with carbon tetrachloride inhalation and oral phenobarbital.
The pathway by which primary renal sodium retention would be linked to venous outflow obstruction in the overfill theory is not clear. Convincing evidence does exist for the presence of an intrahepatic sensory network composed of osmoreceptors, ionic receptors, and baroreceptors. Studies in which hepatic venous pressure is raised have demonstrated increases in hepatic afferent nerve activity. Furthermore, a neural reflex pathway linking hepatic venous congestion and augmented sympathetic nerve activity has been identified. In addition, acute constriction of the portal vein in dogs results in renal sodium and water retention in the innervated unilateral kidney, while these effects are abolished in the contralateral denervated kidney. In addition to a neural mechanism, there may also be a hormonal system by which the liver and kidney can communicate. Hepatically-produced cAMP has been proposed to be a component of such a system. Circulating cAMP is known to inhibit proximal salt and water absorption, as well as to contribute to the regulation of glomerular filtration rate. According to this hypothesis, decreased circulating cAMP levels as a result of liver disease could secondarily lead to renal salt retention and impaired renal function.
In summary, the overfill hypothesis is supported by a number of observations that indicate sodium retention precedes development of ascites in the absence of hemodynamic factors known to lead to salt retention. Moreover, high cardiac output coupled with increased plasma volume argues strongly for increased arterial blood volume, a finding seemingly incompatible with the underfill theory. Against such an analysis, however, is that mechanisms that sense arterial volume physiologically may be more sensitive than methods used to measure it. It should be noted that while statistically insignificant, there was a fall in blood pressure at the time of positive sodium balance in the dimethylnitrosamine model. This decrease may have been of sufficient magnitude to signal renal salt retention. Since cardiac output was unchanged, total peripheral resistance may have decreased. Similarly, patients with hepatic cirrhosis and ascites behave as if they are effectively volume-depleted. Despite an increase in cardiac output and plasma volume, arterial pressure is typically low. This fall in systemic blood pressure is consistent with an underfilled arterial vascular compartment. Thus, the distinction between classical and overflow theories better rests on the measurement of effective arterial blood volume (EABV).
Use of EABV to Distinguish Underfill and Overfill Mechanisms of Renal Salt Retention
The classical (underfill) theory predicts that EABV is low in patients with ascites, and is the afferent mechanism signaling renal salt retention. The overflow theory predicts that EABV is expanded due to primary salt retention. While EABV cannot be measured directly, assessing the level of activation of neurohumoral effectors known to be regulated by EABV can be considered a measure of it. In this regard, levels of renin, aldosterone, ADH, and norepinephrine can serve as markers reflective of the magnitude of the EABV.
When renin and aldosterone values have been measured in patients with cirrhosis, values have varied from low to high. It is important, however, to consider these levels in the context of whether ascites is present or not. In the absence of ascites, subjects are in sodium balance, and renin and aldosterone levels are normal. In the presence of ascites, mean renin and aldosterone levels are elevated, but individual values are often still normal. This observation seems in conflict with the classical theory, as all patients with ascites who are in positive sodium balance should have decreased EABV and high aldosterone levels. However, not all patients with ascites are retaining sodium. In fact, some patients are in balance such that sodium intake equals output. Thus, in examining the mechanism of sodium retention in cirrhosis with ascites, neurohumoral effectors such as renin, aldosterone, ADH, and sympathetic nerve activity should be considered with respect to the rate of sodium excretion. When examined in this fashion, a significant inverse relationship is found between urinary sodium excretion and neurohumoral markers, suggesting the presence of a contracted EABV.
One component of the circulation that appears to be contributing to the overall decrease in EABV is the central circulation. Indirect measurements demonstrate that central blood volume is reduced, while noncentral blood volume is expanded. In fact, the size of central and arterial blood volume is inversely correlated with sympathetic nervous system activity, suggesting that unloading of central arterial baroreceptors is responsible for enhanced sympathetic activity. This conclusion is supported by studies using the technique of head-out water immersion (HWI). In this technique, subjects are seated and immersed in a water bath up to their necks. This technique results in redistribution of ECF volume from the interstitial space into the vasculature, with a sustained increase in central blood volume. The central volume expansion is comparable to that induced by infusion of 2 liters of isotonic saline. Such a maneuver would be expected to raise both the EABV and the hepatic sinusoidal pressure. The classical theory would predict that HWI would lead to decreases in renin, aldosterone, ADH, and norepinephrine concentrations in response to expansion of EABV. Since renin levels correlate with wedged hepatic vein pressures, the overfill theory would predict further rises in renin and other hormonal systems consequent to initiation of a sinusoidal pressure-sensitive hepatorenal reflex. When HWI was performed in a heterogeneous group of patients with cirrhosis, the natriuretic response was variable, but suppression of renin and aldosterone levels was uniform. In a more homogenous group of patients characterized by impaired ability to excrete water and sodium, HWI was shown consistently to suppress plasma AVP, renin, aldosterone, and norepinephrine, as well as to increase sodium and water excretion.
Taken together, the multiplicity of data support the presence of decreased EABV in patients with decompensated cirrhosis, and is most consistent with an underfill mechanism of renal salt retention ( Figure 38.7 ). Since blockade of endogenous vasoconstrictor systems in patients with cirrhosis and ascites leads to marked arterial hypotension, activation of these systems function to contribute to the maintenance of arterial pressure. At least one component of the decrease in EABV may be due to an underfilled central circulation. As discussed in the following paragraphs, increased perfusion of arteriovenous communications, systemic vasodilation, and increased perfusion of the splanchnic bed are important factors in the genesis of an underfilled circulation. In addition, these factors play a major role in the hyperdynamic circulation that is typical of patients with chronic liver disease.
Hyperdynamic Circulation in Cirrhosis
The characteristic circulatory changes observed in animal, as well as clinical, studies of cirrhosis consist of increased cardiac output, low mean arterial pressures, and low peripheral vascular resistance. The most attractive explanation for a contracted EABV in the setting of such a hyperdynamic circulation assigns a pivotal role to increased vascular capacitance. An increased vascular holding capacity out of proportion to plasma volume results in an underfilled circulation and decreased EABV. One factor that may account for increased vascular capacitance and a hyperdynamic circulation is the formation of widespread arteriovenous communications. In cirrhotics, arteriovenous fistula formation has been identified in the pulmonary, mesenteric, and upper and lower extremity circulations. In addition, increased blood flow has been measured in muscle and skin of the upper extremity not attributable to increased oxygen consumption, anemia or thiamine deficiency. Postmortem injection demonstrated intense proliferation of small arteries in the splenic vasculature of patients with cirrhosis.
The hemodynamic changes and salt retention that occur with an arteriovenous fistula are reminiscent of what occurs in cirrhotic humans. With an open fistula, peripheral vascular resistance falls, cardiac output increases, and diastolic and mean blood pressures fall. The proportionately greater increase in vascular capacitance over cardiac output results in a contracted EABV. Consequent sodium retention expands ECF volume, raises venous filling pressure, and further increases cardiac output until balance is achieved between cardiac output and lowered peripheral resistance. At this point, sodium intake equals excretion, EABV is normalized, and the patient is in balance.
In cirrhosis, a similar imbalance occurs between plasma volume and vascular capacitance, such that EABV remains contracted and renal sodium retention is stimulated. In contrast to a simple arteriovenous fistula, however, several factors are present in cirrhosis that make sodium balance more difficult to achieve. First, these patients often have impaired cardiovascular function. Diminished venous return consequent to tense ascites or cardiomyopathy from alcohol or malnutrition may limit increases in cardiac output. Furthermore, depression of left ventricular function in response to increased afterload suggests subclinical cardiac disease, despite elevated forward output. Second, retained sodium does not remain in the vascular space and lead to increased venous return. Rather, retained sodium becomes sequestered within the abdomen as ascites. Third, increased vascular permeability may further impair the ability of retained sodium to expand EABV. Peripheral arterial vasodilation in cirrhotic rats is associated with increased vasopermeability to albumin, electrolytes, and water. Examination of interstitial fluid dynamics by means of a subcutaneous plastic capsule reveals substantial increases in interstitial fluid volume early in cirrhosis before the appearance of ascites or peripheral edema. Such capillary leakage impedes filling of the intravascular compartment and prevents replenishment of a contracted EABV.
Primary Arterial Vasodilation
Arteriovenous fistulas and formation or hyperdynamic perfusion of pre-existing capillary beds are changes that develop as cirrhosis progresses. Nevertheless, salt retention occurs early in the cirrhotic process before these anatomic changes are fully-established. Since sodium retention antedates the formation of overt ascites and portosystemic shunting, peripheral arterial vasodilation has been proposed to be a primary event in the initiation of sodium and water retention in cirrhosis. In this manner, a decreased EABV and increased vascular capacitance could still be the signal for renal salt retention, even in the earliest stages of liver injury. The peripheral arterial vasodilation hypothesis is supported by several studies in animal models. In rats with partial ligation of the portal vein, evidence of a reduced systemic vascular resistance precedes the onset of renal salt retention. In addition, a direct correlation has been found between the onset of decreased arterial pressure and renal sodium retention in spontaneously hypertensive rats with experimental cirrhosis. As opposed to the classical underfilling theory, the arterial vascular underfilling would not be the result of a reduction in plasma volume, which in fact is increased, but rather to a disproportionate enlargement of the arterial tree secondary to arterial vasodilation. In the rat with carbon tetrachloride-induced cirrhosis, the fall in peripheral vascular resistance and hyperdynamic circulatory state precede ascites formation, suggesting that generalized vasodilation is indeed an early finding with hepatic injury.
Perhaps the best evidence to date in support of an underfilled circulation due to arterial vasodilation comes from human studies of HWI accompanied by infusion of a vasoconstrictor. HWI is associated with increased perfusion of the central circulation, however, urinary excretion rates of salt and water improve, but do not normalize with this procedure alone. Since systemic vascular resistance falls during HWI, it was proposed that further vasodilation may prevent complete restoration of EABV in subjects already peripherally vasodilated. Infusion of a vasoconstrictor will increase peripheral vascular resistance, but will do little to improve an underfilled central circulation. Predictably, infusion of norepinephrine alone into cirrhotic subjects fails to significantly increase urinary sodium excretion. By contrast, when norepinephrine is infused during HWI so as to increase central perfusion and at the same time attenuate the fall in systemic vascular resistance, sodium excretion increases significantly. In six subjects with decompensated cirrhosis, this combined maneuver was found to increase urinary sodium excretion to an amount that when extrapolated over a 24-hour period was greater than sodium intake. These results are consistent with the hypothesis that arterial vasodilation causes an abnormal distribution of the total blood volume, such that effective central blood volume is reduced.
Splanchnic Arterial Vasodilation
As alluded to earlier, arterial vasodilation is particularly marked in the splanchnic arteriolar bed. Increasing degrees of splanchnic vasodilation contribute to the fall in mean arterial pressure and unloading of baroreceptors in the central circulation. As a result, central afferent sensors signal the activation of neurohumoral effectors, which in turn decrease perfusion of other organs, but in particular in the kidney. The importance of splanchnic vasodilation in the genesis of renal ischemia has been indirectly illustrated by the response to ornipressin, an analog of AVP that is a preferential splanchnic vasoconstrictor. The administration of ornipressin to patients with advanced cirrhosis leads to correction of many of the systemic and renal hemodynamic abnormalities that are present. These include an elevation in mean arterial pressure, reductions in plasma renin activity and norepinephrine concentration, and increases in renal blood flow, glomerular filtration rate, and urinary sodium excretion and volume. Similar benefits have been reported with the combined use of octreotide and midodrine.
Role of Nitric Oxide in Arterial Vasodilation
The underlying cause of arterial vasodilation, particularly in the early stages of cirrhosis, has not been fully-elucidated, but a great deal of attention has been focused on humoral factors. There is an increasing body of experimental and preliminary human evidence suggesting that increased nitric oxide production may be an important factor in this process. In both experimental models and in human subjects with cirrhosis, increased production of nitric oxide can be demonstrated. In the cirrhotic rat, evidence of increased production is already present when the animals begin to retain sodium, and antedates the appearance of ascites. Administration of nitric oxide synthase inhibitor L-NMMA to cirrhotic human subjects improves the vasoconstrictor response to noradrenaline, suggesting that overproduction of nitric oxide is an important mediator of the impaired responsiveness of the vasculature to circulating vasoconstrictors. In addition, this same inhibitor administered in low doses has been shown to correct the hyperdynamic circulation in cirrhotic rats. In a more recent study utilizing this same model, normalization of nitric oxide production was associated with a marked natriuretic and diuretic response, as well as a reduction in the degree of ascites in cirrhotic rats.
The precise mechanism for increased nitric oxide production in cirrhosis is not known, but may be mediated at least in part via the release of tumor necrosis factor-alpha. In experimental models of hepatic disease, for example, the administration of anti-TNF-alpha antibodies or an inhibitor of nitric oxide synthesis results in increases in splanchnic and total vascular resistance, an elevation in the mean arterial pressure, and a reduction in cardiac output toward or, with nitric oxide inhibition, to normal. Similarly, blocking the signaling events induced by TNF and nitric oxide production, via inhibition of protein tyrosine kinase, ameliorates the hyperdynamic abnormalities in rats with cirrhosis and portal hypertension. Studies in cirrhotic humans with an increased cardiac output and systemic vasodilatation have shown evidence of enhanced nitric oxide production, a finding compatible with the experimental observations. Portosystemic shunts and decreased reticuloendothelial cell function may allow intestinal bacteria and endotoxin to enter the systemic circulation, providing a potential stimulus for tumor necrosis factor-alpha and/or nitric oxide production.
It is not yet known with certainty whether the endothelial (eNOS) or the inducible (iNOS) isoform is primarily responsible for increased production of nitric oxide. The hyperdynamic circulatory state of cirrhosis may impose a shear stress on the vascular endothelium, thus providing a stimulus for the upregulation of eNOS. On the other hand, increased activity of nitric oxide synthase in polymorphonuclear cells and monocytes (cells that primarily contain iNOS) in cirrhotic human subjects suggest the inducible isoform may also play a role in increased production.
In summary, an underfill mechanism appears to explain the bulk of experimental, as well as clinical, findings in established cirrhosis ( Figure 38.7 ). Less certain are mechanisms responsible for sodium retention that precede the development of ascites. The overfill theory invokes the presence of a hepatorenal reflex sensitive to subtle rises in intrahepatic pressure mediating initiation of renal salt retention. However, the finding of decreased peripheral vascular resistance, even at this early stage, suggests diminished arterial filling. Early peripheral arterial vasodilation and later formation of anatomic shunts lead to disproportionate increases in vascular capacitance with subsequent contraction of EABV, thereby signaling renal salt retention. While it is conceivable that both overfill mechanisms and underfill mechanisms may be operative at different stages of disease, the multiplicity of data both clinical and experimental can be assimilated into an underfill theory.
Concept of Balance in Cirrhosis
In the earliest stages of cirrhosis when arterial vasodilation is moderate and the lymphatic system is able to return increased lymph production to the systemic circulation, renal sodium and water retention are sufficient to restore EABV and thereby suppress neurohumoral effectors. Balance is re-established such that sodium intake equals sodium excretion, but at the expense of an increased ECF volume. As liver disease progresses, this sequence of arterial underfilling followed by renal salt retention is repeated. As long as the EABV can be restored to near normal levels the activation of effector mechanisms will be moderated, and balance will be achieved albeit at ever-increasing levels of ECF volume ( Figure 38.8 ). Eventually, lymph production will begin to exceed the drainage capacity of the lymphatic system. At this stage of the disease renal salt retention becomes less efficient at restoring EABV, as retained fluid is sequestered in the peritoneal cavity as ascites. At the same time arterial underfilling is more pronounced, particularly as splanchnic arteriolar vasodilation becomes more prominent. Activation of neurohumoral effectors is magnified, resulting in more intense renal salt retention. Even at this stage of the disease cirrhotic patients with ascites eventually re-establish salt balance. The terminal stages of the cirrhotic process are characterized by extreme arterial underfilling. At this time there is intense and sustained activation of neurohumoral effectors. As a result, renal salt retention is nearly complete as the urine becomes virtually devoid of sodium. The vasoconstrictor input focused in on the kidney is of such a degree that renal failure begins to develop.
Effector Mechanisms in Cirrhosis
Nephron Sites of Renal Sodium Retention
Salt retention and impaired free water clearance are characteristic disturbances in renal function in cirrhotic patients. Evidence is available to support an important role for proximal and distal nephron segments in mediating enhanced sodium reabsorption.
Indirect evidence supporting enhanced proximal salt reabsorption comes from studies in human cirrhotic subjects in which infusion of mannitol or saline improves free water clearance. Increased proximal tubular salt reabsorption leads to decreased delivery of filtrate to the distal diluting segments, thereby impairing free water formation. Presumably, by restoring distal delivery of filtrate, mannitol and saline infusions result in increased solute free water formation. Similar increases in free water clearance occur when central hypervolemia is induced by HWI. The increase in urine sodium is accompanied by increased K excretion, suggesting enhanced distal delivery of sodium. Increased water excretion seen in response to HWI, combined with simultaneous infusion of norepinephrine, is also consistent with baseline enhanced proximal sodium reabsorption in decompensated cirrhotics.
Experimental models of cirrhosis have provided more direct assessment of nephron function. Micropuncture studies in rats made cirrhotic by ligation of the common bile duct demonstrated increases in both the proximal tubule solute reabsorption and filtration fraction. Enhanced proximal reabsorption was attributed to increased peritubular oncotic pressure. In a dog model of cirrhosis, intrarenal administration of the vasodilator acetylcholine was found to ameliorate the blunted natriuretic response to saline infusion. In this model, sodium reabsorption was enhanced in both the proximal and the diluting segments of the nephron.
Clinical and experimental evidence also supports an important role for distal nephron sodium retention in cirrhosis. In cirrhotic patients manifesting a sluggish natriuretic response to HWI, phosphate clearance was found similar to a group who demonstrated an appropriate increase in urinary sodium excretion. Since phosphate clearance was used as a marker of proximal sodium reabsorption, it was concluded that distal sodium reabsorption contributed importantly to renal sodium retention in patients with a sluggish natriuretic response. The results of a prospective, double-blind study comparing the diuretic response of furosemide to spironolactone in cirrhotic patients with ascites suggest that salt absorption in the cortical collecting tubule is enhanced. When administered furosemide, only 11 of 21 patients exhibited a diuresis, while 18 of 19 patients responded to spironolactone. Furthermore, 10 patients who failed to respond to furosemide demonstrated a diuretic response to spironolactone. Furosemide inhibits sodium reabsorption in the loop of Henle, thereby increasing delivery to the collecting duct. All patients treated with furosemide had increases in the rate of potassium excretion, including the 11 patients who failed to increase urinary sodium excretion. These results, combined with the clinical effectiveness of spironolactone in treatment of cirrhotic ascites, suggest enhanced salt absorption in the aldosterone-sensitive cortical collecting tubule.
In summary, clinical and experimental studies suggest an important role for proximal, as well as distal, nephron sites mediating renal salt retention in cirrhosis. The relative contribution of different nephron sites to impaired salt and water excretion may depend on the degree to which systemic hemodynamics are altered. With each stage of advancing liver disease there becomes a greater contraction of the EABV. In the earliest stages of liver disease, enhanced proximal reabsorption limits distal delivery of solute in a manner analogous to a nonedematous subject with intravascular volume depletion. If distal delivery can be normalized at this early stage, distal nephron sites may continue to reabsorb sodium avidly, and therefore appear as the primary site responsible for ECF volume expansion. With severe reductions in EABV, presumably the proximal nephron becomes the dominant site of fluid reabsorption, such that the contribution of the distal nephron becomes much less apparent.
Sympathetic Nervous System
The sympathetic nervous system has been shown to contribute importantly to abnormalities in body fluid homeostasis in cirrhosis. Studies in rats made cirrhotic by ligating the common bile duct suggest that increased renal nerve activity is a major factor in the progressive salt retention that occurs in these animals. In this model, baseline renal nerve activity is increased, and fails to decrease appropriately in response to intravenous saline. Renal denervation significantly improves the impaired ability to excrete an oral or intravenous salt-load. In addition, renal denervation has been shown to normalize the attenuated diuretic and natriuretic response to the intravenous administration of ANP. In chronic metabolic studies, renal denervation also leads to a significant improvement in the positive cumulative sodium balance. The cause of increased renal nerve activity is multifactorial. Impaired aortic and cardiopulmonary baroreceptor regulation, as well as abnormalities in hepatic NaCl-sensitive receptors and their immediate intrahepatic afferent connections, have been implicated as a cause of heightened autonomic activity.
Studies in human cirrhotic subjects are more indirect, but also suggest an important role for the sympathetic nervous system. Levels of norepinephrine in patients with cirrhosis are high, and are inversely correlated with urinary sodium excretion. In addition, direct measurement of peripheral nerve firing rates show evidence of increased central sympathetic activity. Patients characterized by impaired ability to excrete water loads have plasma levels of norepinephrine that correlate positively with levels of ADH, aldosterone, and plasma renin activity.
Decreased EABV leads to baroreceptor-mediated activation of sympathetic nerve activity, with subsequent enhancement of proximal salt reabsorption. The subsequent decrease in sodium delivery to the diluting segment, in addition to nonosmotic release of ADH, contributes to the inability to maximally excrete water-loads. Increased renal nerve activity also contributes to enhanced distal sodium reabsorption through activation of the renin–angiotensin–aldosterone system.
In addition to stimulating renal salt and water retention, activation of the sympathetic nervous system serves as a compensatory response to cirrhosis-induced vasodilation. Increased renal nerve activity contributes to increased renal vascular resistance, and is one of several factors responsible for the progressive decline in renal function which occurs as patients develop the hepatorenal syndrome.
In summary, the sympathetic nervous system is activated under conditions of decompensated cirrhosis. Overactivity of this system is the result of a contracted EABV. In addition, there is impaired regulation of sympathetic outflow due to abnormalities in several afferent sensing mechanisms. Increased renal nerve activity contributes to the cumulative salt retention that accompanies advancing liver disease. In addition, activation of sympathetic outflow plays an important compensatory role in maintaining vascular tone in the setting of decreased vascular resistance.
In patients with cirrhosis and ascites, plasma concentrations of aldosterone are frequently elevated. Although aldosterone metabolism is impaired in liver disease, secretion rates are greatly elevated, and are the major cause of elevated levels. The relationship between hyperaldosteronism and sodium retention is not entirely clear. Several studies have provided evidence that argues against an important role of aldosterone in mediating salt retention in cirrhosis. For example, patients treated with an aldosterone synthesis inhibitor do not necessarily exhibit a natriuretic response. In one study, renal salt excretion and changes in plasma renin and aldosterone levels were examined in 11 patients with ascites subjected to 5 days of high-salt intake. In patients with normal suppression of renin and aldosterone, salt retention and weight gain occurred to the same extent as patients who had persistent hypersecretion of renin and aldosterone. In addition, cirrhotic patients in positive sodium balance, as compared to controls with matched sodium excretion, have increased fractional distal sodium reabsorption despite lower plasma aldosterone levels. In 16 cirrhotic patients subjected to HWI, plasma renin activity and plasma aldosterone levels were found to decrease promptly. Despite suppression of the hormones, however, half of the patients manifested a blunted or absent natriuretic response. In another group of cirrhotic patients with ascites and edema, HWI induced a significant natriuresis despite acute administration of desoxycorticosterone, suggesting that enhanced sodium reabsorption can occur independently of increased mineralocorticoid activity.
By contrast, other studies suggest aldosterone is an important factor in the pathogenesis of sodium retention in patients with cirrhosis. For example, adrenalectomy or administration of a competitive inhibitor of aldosterone increases urinary sodium excretion. Patients who fail to manifest a diuretic response to furosemide tend to have higher renin and aldosterone levels and lower urinary sodium concentrations prior to treatment. Inability of furosemide to increase urinary sodium in these patients may result from reabsorption of delivered sodium in the collecting tubule under the influence of aldosterone. Similarly, patients with the highest renin and aldosterone levels are those who fail to diurese in response to HWI.
As with the conflicting data regarding the role of the proximal and distal nephron in salt retention discussed previously, the degree to which systemic hemodynamics and EABV are impaired may explain some of the conflicting data noted above. It is possible that in patients with the greatest contraction of EABV, intense proximal sodium reabsorption limits distal delivery to such an extent that the contribution of aldosterone to increase salt absorption is difficult to detect. By contrast, with less impairment in EABV, distal delivery is better maintained, and the contribution of aldosterone to renal sodium retention is more obvious.
Alternatively, acquired inhibition of 11β-hydroxysteroid dehydrogenase type 2 may be of importance in the salt retention that occurs in some patients with cirrhosis of the liver. Bile acids which can accumulate in the setting of chronic liver disease have been shown to inhibit the activity of 11β-hydroxysteroid dehydrogenase type 2. Such an effect would allow cortisol-mediated stimulation of the mineralocorticoid receptor, and potentially explain aldosterone-independent salt retention in the distal nephron in liver cirrhosis. Studies in the bile duct ligation and carbon tetrachloride models of chronic liver disease are consistent with a component of cortisol-mediated stimulation of the mineralocorticoid receptor. In these models there is decreased activity of 11β-hydroxysteroid dehydrogenase type 2 that is temporally related to increased ENaC abundance in the apical membrane of the cortical collecting duct. These changes are most pronounced in the sodium-retaining stage of disease.
Prostaglandins function in a protective role in decompensated cirrhosis. Similar to other hypovolemic states, prostaglandins act to maintain renal blood flow and GFR by ameliorating pressor effects of angiotensin II and sympathetic nerves. These agents counterbalance the salt retaining effects of these effectors and mitigate the impairment in free water clearance mediated by AVP. Administration of prostaglandin inhibitors can partially correct excessive hyperreninemia and hyperaldosteronism, and restore the pressor response to angiotensin II.
Urinary kallikrein activity is increased in cirrhotic patients with ascites and preserved GFR, while urinary activity decreases in association with impaired renal function. The correlation between renal plasma flow and GFR suggests that the renal kallikrein–kinin system may contribute to maintenance of renal hemodynamics in cirrhosis.
At the level of the renal tubule, bradykinin has been shown to exhibit a natriuretic effect. However, bradykinin also is a potent peripheral vasodilator, and can cause microvascular leakage. In cirrhosis, these later effects could exacerbate an already contracted EABV, and cause further salt retention. MacGilchrist et al. studied the effects of kinin inhibition by systemically infusing aprotinin (a strong inhibitor of tissue kallikrein) into a group of patients with cirrhosis. This infusion was associated with a doubling of urinary sodium excretion, and an increase in renal plasma flow and GFR. This beneficial effect on renal function in the setting of kinin inhibition was attributed to an improvement in systemic hemodynamics as systemic vascular resistance increased. Similarly, administration of a bradykinin receptor antagonist to cirrhotic rats normalized renal sodium retention, and reduced the activity of the renin–angiotensin–aldosterone system. Inhibiting bradykinin-induced microvascular leakage and lessening the degree of vascular underfilling was felt to be the mechanism of the beneficial effect.
The role of ANP in the pathogenesis of edema in hepatic cirrhosis remains undefined. While atrial ANP content was reduced in cirrhotic rats, most data indicate ANP levels are either normal or elevated in cirrhotic humans. Elevated levels are the result of increased cardiac release rather than just impaired clearance. The cause of the high levels is not understood, because atrial pressure is normal and central blood volume is reduced. Stimulating the endogenous release of ANP induces a natriuretic response in some patients with cirrhosis, while other patients are insensitive. However, both groups of patients exhibited an increase in urinary cGMP, suggesting that the kidney is still capable of responding to ANP even in the absence of a natriuretic effect.
Several potential mechanisms may account for ANP resistance in cirrhosis. This resistance could be the result of a defect intrinsic to the kidney or could be the result of altered systemic hemodynamics leading to activation of more potent sodium-retaining mechanisms. With regards to the first possibility, an altered density of glomerular ANP-binding sites has been demonstrated in the bile duct-ligated rat model of cirrhosis. In addition, ANP resistance was found in the isolated perfused kidney taken from sodium avid rats with cirrhosis induced by carbon tetrachloride. In the chronic caval dog model of cirrhosis, intrarenal infusion of bradykinin restored ANP responsiveness to previously resistant animals, suggesting that an intrarenal deficiency of kinins could be a contributing factor.
Other studies have focused on systemic hemodynamics as a cause of ANP resistance. With each stage of advancing liver disease there becomes a greater reduction in EABV. Since ANP resistance tends to occur with more severe and advanced disease, it is possible that ANP resistance is directly related to the impairment in EABV. Decreased EABV is associated with enhanced proximal reabsorption of solute. As a result, ANP resistance may be due to decreased delivery of salt to the site where ANP exerts its natriuretic effect. In support of this possibility, ANP resistance could be restored in cirrhotic rats by infusions of vasopressors so as to normalize arterial pressure, and presumably improve the decrease in EABV. In human cirrhotics, ANP responsiveness can be markedly improved when distal sodium delivery is increased by administration of mannitol.
Circulating brain natriuretic peptide (BNP) levels are also increased in patients with cirrhosis. Infusion of BNP at a dose that elicits an increase in GFR, renal plasma flow, and urinary sodium excretion in normal controls has no effect in cirrhotic humans. The infusion is associated with an increase in urinary cGMP, as well as a fall in plasma aldosterone levels, suggesting that the peptide is capable of interacting with its receptor in these patients. As with ANP, the lack of natriuretic response to BNP may be due to overactivity of other antinatriuretic factors, as well as decreased delivery of sodium to its tubular site of action.
Adrenomedullin is a peptide with vasodilatory properties that is highly expressed in cardiovascular tissues. Increased circulating levels that correlate with severity of disease have been described in patients with cirrhosis. Urodilatin is a natriuretic factor that is exclusively synthesized within the kidney. Unlike other natriuretic factors, levels are not increased in patients with cirrhosis.
Increased circulating levels of endothelin have been reported in cirrhosis. The stimulus and pathophysiologic significance of these levels is not known with certainty. The peptide may play a role in the renal vasoconstriction seen in the hepatorenal syndrome.
Therapeutic Implications for Treatment of Salt Retention in Cirrhosis
Renal salt retention is the most common abnormality of renal function in chronic liver disease. Whenever urinary sodium excretion falls to an amount less than dietary salt intake ECF volume will begin to expand and eventually lead to the development of ascites and peripheral edema. The approach to the treatment of the cirrhotic patient with ascites is to alter sodium balance in such a way that urinary sodium excretion exceeds dietary salt intake.
In the earliest stages of the disease, urinary sodium excretion is plentiful and negative salt balance can be achieved by simply lowering dietary sodium intake. As the disease advances, neurohumoral effectors become more activated, initially resulting in more intense renal salt retention and later in a progressive decline in renal function. Eventually, the filtered load of sodium becomes completely reabsorbed by the tubule, such that the final urine becomes virtually devoid of salt. If some component of the filtered load reaches the collecting duct or beyond spironolactone will be effective in increasing urinary sodium excretion. Once sodium reabsorption is complete proximal to the collecting duct then thiazides, and later loop diuretics, will have to be added to spironolactone in order to increase urinary sodium excretion. Eventually, the filtered load is completely reabsorbed proximal to the thick ascending loop of Henle. At this point the patient is resistant to the effects of diuretics, and requires more invasive procedures such as repetitive large-volume paracentesis in order to remain in salt balance. In the terminal stages of the disease, the glomerular filtration rate falls to such a degree that oliguria, azotemia, and eventually uremia are present, and the patient is clinically diagnosed with hepatorenal syndrome ( Figure 38.9 ). Vasoconstrictive input focused on the kidney is severe. The renal failure is functional in nature, however, since restoration of near normal renal function can be obtained following a liver transplant.