Hyponatremia is a common finding in patients with decompensated cirrhosis due to an abnormal regulation of body fluid homeostasis.1 Although hyponatremia in cirrhosis was described more than 50 years ago,2 its importance in the clinical assessment of patients with cirrhosis was overlooked for many years. Interest in hyponatremia was fostered by studies in the late 1970s and 1980s indicating that hyponatremia is an important prognostic indicator in cirrhosis.3, 4 Recent studies extended these observations and showed that hyponatremia is an important marker of prognosis in both the pretransplant and posttransplant settings.5–9 Moreover, hyponatremia has also gained attention because of the discovery of vaptans, drugs that improve solute-free water excretion by antagonizing the effects of arginine vasopressin (AVP) in the renal tubules, which are currently being evaluated for the management of hyponatremia associated with cardiac failure, the syndrome of inappropriate antidiuretic hormone secretion, and cirrhosis.10 The aim of this article is to review the current concepts of the pathophysiology, clinical relevance, and management of hyponatremia in cirrhosis.
Hyponatremia is a frequent complication of advanced cirrhosis related to an impairment in the renal capacity to eliminate solute-free water that causes a retention of water that is disproportionate to the retention of sodium, thus causing a reduction in serum sodium concentration and hypo-osmolality. The main pathogenic factor responsible for hyponatremia is a nonosmotic hypersecretion of arginine vasopressin (or antidiuretic hormone) from the neurohypophysis related to circulatory dysfunction. Hyponatremia in cirrhosis is associated with increased morbidity and mortality. There is evidence suggesting that hyponatremia may affect brain function and predispose to hepatic encephalopathy. Hyponatremia also represents a risk factor for liver transplantation as it is associated with increased frequency of complications and impaired short-term survival after transplantation. The current standard of care based on fluid restriction is unsatisfactory. Currently, a new family of drugs, known as vaptans, which act by antagonizing specifically the effects of arginine vasopressin on the V2 receptors located in the kidney tubules, is being evaluated for their role in the management of hyponatremia. The short-term treatment with vaptans is associated with a marked increase in renal solute–free water excretion and improvement of hyponatremia. Long-term administration of vaptans seems to be effective in maintaining the improvement of serum sodium concentration, but the available information is still limited. Treatment with vaptans represents a novel approach to improving serum sodium concentration in cirrhosis. (HEPATOLOGY 2008.)
If you can't find a tool you're looking for, please click the link at the top of the page to "Go to old article view". Alternatively, view our Knowledge Base articles for additional help. Your feedback is important to us, so please let us know if you have comments or ideas for improvement.
Definition and Prevalence
Hyponatremia in cirrhosis is currently defined as a reduction in serum sodium below 130 mmol/L.11 Nevertheless, it is important to emphasize that the lower limit of normal of serum sodium concentration is 135 mmol/L, and a significant proportion of patients with cirrhosis have a serum sodium concentration above 130 mmol/L but below the lower limit of normal values. These patients are not considered to have hyponatremia with the current definition but show pathogenic and clinical features similar to those of patients with serum sodium below 130 mmol/L.1, 12 The prevalence of hyponatremia, as defined by serum sodium < 130 mmol/L, is 21.6%. If the cutoff level of 135 mmol/L is used, the prevalence increases up to 49.4%.12
Types of Hyponatremia
Patients with cirrhosis may develop two types of hyponatremia. In some patients, hyponatremia is due to important losses of extracellular fluid, most commonly from the kidneys (because of overdiuresis due to treatment with excessive doses of diuretics) or from the gastrointestinal tract. This condition, known as hypovolemic hyponatremia, is characterized by low serum sodium associated with contraction of plasma volume, lack of edema and ascites, signs of dehydration, and prerenal renal failure. Hepatic encephalopathy is a common finding, probably because of the effects of a rapid reduction of serum osmolality on brain function (discussed later). In contrast to hypovolemic hyponatremia, in most patients with cirrhosis, hyponatremia develops in the setting of expanded extracellular fluid volume and plasma volume with ascites and edema.1, 11 This condition is known as hypervolemic or dilutional hyponatremia and is due to a marked impairment of renal solute–free water excretion, resulting in disproportionate renal retention of water with respect to sodium retention. Renal impairment is frequent but not constant in this type of hyponatremia. Both conditions differ markedly with respect to volume status. In hypovolemic hyponatremia, the actual plasma volume is reduced, and there is also a reduction in the total extracellular fluid volume with a lack of ascites and edema. In hypervolemic hyponatremia, plasma volume is increased in absolute values but is low with respect to the marked vasodilation of the arterial circulation, a condition known as effective arterial hypovolemia,1, 11 and the total extracellular fluid volume is increased, with ascites and/or edema. From now on, this review refers to hypervolemic hyponatremia unless indicated otherwise.
In healthy subjects, total body water is maintained within narrow limits despite marked variations in daily fluid intake in such a way that any increase in water intake is followed by an increase in renal solute–free water excretion, thus preventing the dilution of body fluids and the development of hypo-osmolality. Conversely, a decrease in water intake is associated with a decrease in solute-free water excretion to prevent hyperosmolality and dehydration. This homeostatic mechanism allows for the maintenance of water balance not only with the daily variations in water intake (usually 1.5-3.0 L/day) but also under conditions of marked changes in fluid intake (0.5-20 L/day). These changes in water excretion take place within minutes and depend on the existence of intact osmoreceptors located in the hypothalamus to detect minute changes in plasma osmolality and effector mechanisms to induce the appropriate modifications in the kidneys (AVP and the water channel aquaporin-2).13 The intracellular mechanisms involved in water reabsorption mediated by AVP in the principal cells of the collecting ducts of the kidney are summarized in Fig. 1.
Patients with cirrhosis and ascites frequently have impairment in the renal capacity to eliminate solute-free water.1, 11 In some patients, the impairment in solute-free water excretion is moderate and can be detected only by the measurement of urine volume after water loading. These patients are able to eliminate water normally and maintain a normal serum sodium concentration as long as their water intake is kept within normal limits, but they may develop hyponatremia when water intake is increased. In other patients, the severity of the disorder is such that they retain most of the water ingested, and this causes hyponatremia and hypo-osmolality. The mechanisms contributing to the impaired renal water handling in cirrhosis are shown in Fig. 2. The main factor responsible for hyponatremia is increased production of AVP from the neurohypophysis due to a nonosmotic hypersecretion related to the circulatory dysfunction present in advanced cirrhosis (Table 1 and Fig. 3). A detailed description of the pathogenesis of hyponatremia in cirrhosis can be found elsewhere.1, 11, 13
|1. Plasma levels of AVP are increased in patients with cirrhosis and ascites and correlate closely with the reduction in solute-free water excretion, patients with higher plasma AVP levels having more severe impairment in solute-free water excretion.|
|2. In experimental cirrhosis, a chronological relationship exists between AVP hypersecretion and impairment in water excretion.|
|3. Increased trafficking of aquaporin-2 (the AVP-regulated water channel that mediates the transport of water from the tubular side to the capillary side of the cell) has been demonstrated in experimental cirrhosis.|
|4. Impairment in water excretion does not develop in rats with a congenital deficit of AVP (Brattleboro rats).|
|5. The administration of vaptans, specific antagonists of the V2 AVP receptors, restores the renal ability to eliminate solute-free water and normalizes the serum sodium concentration in a high proportion of patients with cirrhosis and hyponatremia. The interruption of drug administration is associated with recurrence of hyponatremia.|
Brain Adaptation to Hyponatremia
The pathophysiological consequences of changes in the osmolality of the extracellular fluid are related to the movement of water from the extracellular compartment to the intracellular compartment or vice versa. Because serum sodium concentration is the major determinant of the osmolality of the extracellular fluid, changes in serum sodium are associated with parallel changes in osmolality. When serum sodium increases over normal values (hypernatremia) and so does the osmolality of the extracellular fluid, water moves out of the cells to maintain an osmotic equilibrium between the intracellular and extracellular spaces. This shift causes cell dehydration and shrinkage. In contrast, when serum sodium decreases below normal values (hyponatremia), water moves into the cells to attain the osmotic balance, causing cell swelling.14, 15 Increases of cell volume may affect all cells but are particularly important in the brain, as the brain size enlargement is limited by the skull. For this reason, brain cells have defensive mechanisms to limit cerebral edema.16 Brain cell swelling leads to extrusion of intracellular solutes to decrease intracellular osmolality until it matches that of plasma so that cell swelling will be limited. Early after the development of hyponatremia, there is a rapid loss of intracellular electrolytes, particularly potassium, usually within the first 24 hours. Subsequently, there is a loss of low-molecular-weight organic compounds, known as organic osmolytes, including myoinositol, glutamine, choline, and taurine. The coordinate losses of both electrolytes and organic osmolytes from the brain cells enable a very effective regulation of brain volume during hyponatremia.16, 17 The effectiveness of this mechanism in preventing lethal edema depends, among other factors, on the severity of hyponatremia and rate of reduction of the serum sodium concentration, adaptation being more efficient in chronic hyponatremia than in acute hyponatremia. There is evidence that such cerebral adaptation to hyponatremia also occurs in cirrhosis. A few experimental and clinical studies have shown a reduced cerebral concentration of organic osmolytes, which is consistent with the existence of adaptive osmoregulatory mechanisms.18–20
Of similar importance to changes that occur in the brain during hyponatremia are changes that occur after recovery of hyponatremia. When the serum sodium concentration returns to normal, there is a regain of electrolytes and osmolytes in brain cells; electrolytes correct quickly, whereas correction of organic osmolytes is slow, particularly if the duration of hyponatremia has been long.16 This is of major clinical importance because when correction of hyponatremia occurs rapidly, lack of adequate brain adaptation to the normalized osmolality of the extracellular fluid may lead to important brain damage; this is known as osmotic demyelination syndrome.14, 16
Clinical Significance of Hyponatremia
There is limited information on the clinical consequences of hyponatremia in cirrhosis because hyponatremia almost always occurs in the setting of advanced liver failure, which causes a wide array of clinical manifestations. Therefore, the precise identification of consequences of hyponatremia versus those of other causes has so far not been possible. This has been further hindered by the lack of an effective treatment of hyponatremia.
Hyponatremia and Neurological Function
In patients without liver disease, hyponatremia is primarily associated with a broad variety of neurological manifestations related to the existence of brain edema, such as headache, disorientation, confusion, focal neurological deficits, seizures, and, in some cases, death due to cerebral herniation.14 Severity of neurological symptoms in patients with hyponatremia correlates roughly with the levels of osmolality and sodium in the extracellular fluid. However, rather than the absolute reduction in serum sodium levels, the most important factor in determining the severity of neurological symptoms is the rate of fall in serum sodium levels, patients with acute hyponatremia having a much higher incidence of neurological symptoms than patients with chronic hyponatremia.
Studies specifically assessing neurological symptoms in cirrhosis with hyponatremia are lacking. However, the clinical experience indicates that significant neurological manifestations such as headache, focal motor deficits, seizures, and cerebral herniation are very uncommon. It is likely that the relatively low incidence of neurological manifestations in patients with cirrhosis and hypervolemic hyponatremia is related to the fact that in most of these patients hyponatremia is chronic rather than acute, and this gives sufficient time for the brain to adjust to hypo-osmolality of the extracellular fluid.
The effects of hyponatremia on brain function have to be discussed in light of the recent hypothesis that proposes a role for a low-grade cerebral edema in the pathogenesis of hepatic encephalopathy.21 According to this hypothesis, ammonia and other neurotoxins act synergistically to induce a low-grade cerebral edema as a result of swelling of astrocytes, which is mainly due to increased intracellular content of glutamine, secondary to ammonia metabolism. The cerebral edema would not be sufficient to cause an increase in intracranial pressure, but astrocyte swelling would result in a number of alterations of neurological function, which would facilitate the development of hepatic encephalopathy. Evidence for such a low-grade cerebral edema derives from experimental and human studies using magnetic resonance.20, 22, 23 In this context of low-grade cerebral edema, hyponatremia may represent a second osmotic hit to astrocytes, causing further depletion of osmotic counteractive systems. In this situation, cells would probably not tolerate a further challenge to cell volume, and encephalopathy would develop because of any other osmotic stimulus, including situations associated with an increased ammonia load to the brain or further impairment in serum sodium concentration (Fig. 4).
Several lines of evidence support the existence of a relationship between hepatic encephalopathy and low serum sodium concentration. First, serum sodium levels and serum ammonia levels are major factors determining electroencephalographic abnormalities in cirrhosis.24 Second, in patients treated with transjugular intrahepatic portosystemic shunts, hyponatremia is a major risk factor for hepatic encephalopathy.25 Third, in patients treated with diuretics (a clinical situation associated with a high incidence of hepatic encephalopathy), hyponatremia is a risk factor for hepatic encephalopathy (P. Ginès, unpublished data) (Fig. 5). Finally, in a prospective study in patients with cirrhosis using a time-dependent statistical analysis, serum sodium was an independent predictive factor of hepatic encephalopathy.26
Hyponatremia and Complications of Cirrhosis
Besides hepatic encephalopathy, hyponatremia has also been reported to be associated with other complications of cirrhosis, yet information is limited. Specifically, hyponatremia is a frequent finding in patients with cirrhosis and bacterial infections.27 In the majority of patients, hyponatremia occurs in close association with renal failure and correlates with a poor prognosis. The mechanisms leading to hyponatremia in these patients are not known and deserve specific investigation. Finally, it is important to note that patients with hyponatremia constitute a unique population with a very high risk of developing hepatorenal syndrome.28 On the other hand, low serum sodium levels are a very common finding in patients with hepatorenal syndrome. In this situation, hyponatremia may be due not only to increased AVP levels but also to a markedly reduced glomerular filtration rate and increased proximal sodium reabsorption.1
Information on the impact of hyponatremia on health-related quality of life in patients both with and without liver disease is very limited. In patients with cirrhosis, hyponatremia impairs quality of life because patients require a restriction of daily fluid intake to prevent further reductions in serum sodium concentration, and this is usually poorly tolerated. In a recent study in a large population of patients with cirrhosis, hyponatremia was an independent predictive factor of the impaired health-related quality of life, the predictiveness being independent of that of liver function tests and Child-Pugh or Model for End-Stage Liver Disease scores.29
Hyponatremia and Liver Transplantation
Patients with cirrhosis and hyponatremia are at increased risk of developing central pontine myelinolysis after transplantation, and this is related to a rapid change in serum sodium in the early postoperative period.30, 31 Two recent studies have shown that the existence of hyponatremia before transplantation in patients with cirrhosis is associated not only with an increased risk of neurological complications after transplantation but also with an increased risk of renal failure and infectious complications, greater use of blood products, longer duration of hospital stay, and, more importantly, increased short-term mortality after transplantation.8, 9
Management of Hyponatremia
The distinction between hypovolemic and hypervolemic hyponatremia is very important from a therapeutic perspective. Patients with hypovolemic hyponatremia must be treated with saline solutions aimed at increasing plasma volume and normalizing the low total body sodium along with the removal of the precipitating factor (usually diuretics). In contrast, patients with hypervolemic hyponatremia should be managed with interventions aimed at increasing renal solute–free water excretion with the final goal of reducing the excess of water with respect to sodium in the circulation. In this regard, vaptans increase solute-free water excretion and improve serum sodium concentration in patients with hypervolemic hyponatremia but may be deleterious in patients with hypovolemic hyponatremia because these patients do not have an excess of water and its use may cause a further impairment of hypovolemia and renal function. The rest of this section is devoted to the management of hypervolemic hyponatremia.
Fluid restriction (1-1.5 L/day) is currently the standard of care for the management of hypervolemic hyponatremia in cirrhosis, but its efficacy is very limited. In several prospective randomized studies comparing vaptans to placebo, in which patients from both groups were managed with fluid restriction, the efficacy of this latter intervention in the placebo group in improving serum sodium concentration more than 5 mmol/L ranged from 0% to 26%.32–34 Hypertonic sodium chloride has been used in clinical practice in the treatment of severe hyponatremia in cirrhosis, but its efficacy is limited, and it is associated with increasing ascites and edema; therefore, its use in the management of hypervolemic hyponatremia cannot be recommended. Finally, albumin has been shown to improve serum sodium concentration in a few studies, but the number of patients included has been low, and the follow-up has been short.35, 36
Recently, the pharmacological approach to the treatment of hyponatremia has made a big step forward with the discovery of vaptans, drugs that are active orally and cause a selective blockade of the V2 receptors of AVP in the principal cells of the collecting ducts.10 In healthy subjects, vaptans induce a marked and dose-dependent increase in urine volume with a reduction in urine osmolality due to increased solute-free water excretion. In contrast to conventional diuretics, the administration of vaptans in healthy subjects does not increase natriuresis. Table 2 shows a list of the different vaptans, their characteristics, and the current status of their development. No drug has yet been approved specifically for the management of hyponatremia in cirrhosis. Most of the existing information on long-term effects of administration of vaptans in cirrhosis derives from studies reported only in abstract form.
|Name||Compound||Receptor||Route of Administration||Specific Studies in Patients with Cirrhosis||Current Status of Clinical Development|
|Conivaptan||YM-087||V1a/V2||Intravenous||No||Approved in the United States for the management of hyponatremia in hospitalized patients|
|Lixivaptan||VPA-985||V2||Oral||Yes32, 37, 38||Phase 2|
|Satavaptan||SR-121463||V2||Oral||Yes33, 34, 39, 40||Phase 3|
|Mozavaptan||OPC-31260||V2||Oral||Yes41||Approved in Japan for the treatment of the syndrome of inappropriate antidiuretic hormone secretion|
Effects of Vaptans in Patients with Cirrhosis
The first studies on vaptans in cirrhosis were performed in patients with ascites but without hyponatremia.37, 41 In this population, the oral administration of vaptans is associated with a marked increase in urine volume, reduction in urine osmolality, and increase in solute-free water excretion, which is responsible for a negative fluid balance. Urine sodium excretion does not change significantly. The effects on urine volume start 1 to 2 hours after the administration of the drug and last 4 to 12 hours, depending on the dose. In all studies, there was large interindividual variability, with some patients showing almost no increase of urine volume and others having remarkable diuresis (4-5 L in patients receiving high doses).
On this background, subsequent studies were performed to evaluate whether treatment with vaptans, including lixivaptan, tolvaptan, and satavaptan, is effective in improving serum sodium concentration in patients with hyponatremia.32, 33, 38, 43, 44 The results of these studies have consistently demonstrated that the administration of vaptans for a short period of time (1-2 weeks in most of the studies) is associated with a significant improvement of the low serum sodium levels (Table 3). The increase in serum sodium concentration occurs within the first few days of treatment and ranges from 2 to 7 mmol/L on average (Fig. 6). A normalization of serum sodium concentration is observed in 27% to 54% of patients. Moreover, in approximately one-third of additional patients, serum sodium increases more than 5 mmol/L but does not reach normal values. In short-term studies, no significant effects have been observed on renal function, circulatory function, and activity of the renin-angiotensin-aldosterone system. The effects of treatment on blood volume distribution (that is, central versus splanchnic) or response to vasoconstrictors in the peripheral circulation have not been evaluated. Plasma AVP levels increase consistently during treatment. An increase in urinary sodium excretion has not been observed consistently in all studies.32, 33, 38, 43–45
|Authors||Compound||Dosage (Number of Patients)||Duration of Treatment||Baseline Serum Sodium||End-of-Treatment Serum Sodium||Responders (% of Patients)*|
|Wong et al.38||Lixivaptan||Placebo (8)||7 days||127 ± 1||126 ± 1||NR|
|25 mg bid (8)||126 ± 1||129 ± 2||NR|
|125 mg bid (10)||122 ± 2||127 ± 3||NR|
|250 mg bid (7)||125 ± 1||132 ± 1||NR|
|Gerbes et al.32||Lixivaptan||Placebo (20)||7 days||127 ± 3||128 ± 4||5%|
|100 mg (22)||128 ± 4||130 ± 7||45%|
|200 mg (18)||126 ± 4||132 ± 7||67%|
|Ginès et al.33||Satavaptan||Placebo (28)||14 days||126 ± 4||128 ± 7||26%|
|5 mg (28)||127 ± 5||131 ± 6||50%|
|12.5 mg (26)||128 ± 4||133 ± 5||54%|
|25 mg (28)||126 ± 6||134 ± 6||82%|
Currently, only one study has reported the long-term effects of vaptans on serum sodium concentration in patients with cirrhosis and hyponatremia.34 In that study, patients received a variable dose of satavaptan, according to the effect on the serum sodium concentration, together with diuretic therapy. The main finding of this study was that the improvement in serum sodium concentration obtained after the first days of therapy was maintained for up to 1 year. Although these results are encouraging, further studies are needed to determine the efficacy of vaptans in the long-term management of hyponatremia in cirrhosis.
The most frequent side effect of vaptans in patients with cirrhosis is thirst. In randomized, double-blind studies, thirst was reported as a side effect in up to 29% of patients treated with vaptans.32, 33, 38 Potential theoretical concerns of the administration of vaptans in patients with cirrhosis include the following: (1) hypernatremia due to a markedly negative fluid balance, (2) a rapid increase in serum sodium concentration, and (3) renal failure due to depletion of the intravascular volume. In short-term studies, hypernatremia (serum sodium > 145 mmol/L) occurred uncommonly (only 2%-4% of treated patients).33 Patients with increased risk of hypernatremia are those who have an altered mental state (that is, encephalopathy) and are not able to drink fluid in an amount sufficient to compensate for urine losses. In this situation, vaptans should be used, with great caution, in the hospital setting and with serum sodium monitored frequently. If these conditions cannot be met, vaptans should not be used until patients have recovered a normal mental state. An important concern in all studies has been to avoid a rapid increase in serum sodium that could lead to severe neurological disturbances, particularly central pontine myelinolysis. An increase of serum sodium of greater than 8 mmol/L within the first days of therapy has been reported with a low and similar frequency, ranging from 4% to 14%, in patients treated with vaptans compared to patients treated with placebo.32, 33, 38 Moreover, central pontine myelinolyis has not been reported thus far in any of the studies. Nevertheless, it is important to emphasize that patients were treated in the hospital for the first days of therapy, had free access to water, and followed strict investigation protocols with daily measurement of serum sodium and interruption of drug administration if serum sodium increased more than 8 mmol/day. Finally, in short-term studies, no significant impairment of renal function was found in vaptan-treated groups compared to placebo. Moreover, no significant effects were observed in effective arterial blood volume, as assessed by the activity of the renin-angiotensin system.33 Nonetheless, it should be pointed out that in these studies patients were treated for short periods of time, under strict clinical and analytical surveillance, and with low doses of diuretics. Therefore, it is not known whether the frequency of renal impairment could be higher under different study conditions. Taken together, the available information indicates that these drugs are safe and are associated with a frequency of side effects that is not significantly different from that in patients treated with placebo, except for thirst. Nevertheless, it should be kept in mind that these drugs have a very powerful effect in increasing urine volume and may cause important shifts of extracellular fluid in very short periods of time and therefore should be given under strict clinical and analytical surveillance, particularly at the initiation of therapy or when the dose of the drug is increased. Moreover, it is important to note that in all studies vaptans have been given either without diuretics or with low doses of diuretics (spironolactone, 100 mg/day, in most studies). Therefore, more information is clearly required with respect to the effects of the combined administration of vaptans and diuretics on plasma volume and renal function, particularly when high doses of diuretics are used.
Potential Beneficial Effects of Treating Hyponatremia in Cirrhosis
There are several potential advantages of treating hyponatremia in cirrhosis. First, the reversal of hyponatremia would allow patients to drink fluids normally and thus avoid fluid restriction. Second, treatment with vaptans may prevent the reduction in serum sodium levels commonly seen in patients under diuretic therapy. This may help achieve effective doses of diuretics and improve the response to therapy in patients with difficult-to-treat ascites. Third, because hyponatremia is a factor predisposing to hepatic encephalopathy,25, 26 the improvement of serum sodium concentration may reduce the risk of this complication. Fourth, treatment of hyponatremia may theoretically improve quality of life in patients with cirrhosis. Finally, in patients awaiting liver transplantation, the normalization of the serum sodium concentration before transplantation may help reduce the frequency and severity of neurological complications after transplantation. Nevertheless, all these potentially beneficial effects of treating hyponatremia in cirrhosis should ideally be evaluated in prospective studies.
Finally, a comment on the possible beneficial effect of vaptans in the management of ascites seems pertinent because in short-term studies the administration of satavaptan has been shown to reduce ascites volume.33, 39 On the basis of these data, a phase 2, randomized, double-blind study compared the efficacy of satavaptan in the prevention of ascites recurrence in patients with cirrhosis treated with large-volume paracentesis.40 The results of this study showed that patients treated with satavaptan in combination with diuretics had a lower recurrence rate of ascites and required fewer paracenteses during a 12-week period in comparison with a group of patients treated with placebo. If these results are confirmed in phase 3 studies, which are currently underway, the prevention of ascites recurrence could be another indication of vaptan therapy.