Ascites is a serious complication of cirrhosis, occurring

Transcription

1 Terlipressin Improves Renal Function in Patients with Cirrhosis and Ascites Without Hepatorenal Syndrome Aleksander Krag, 1,2 Søren Møller, 2 Jens H. Henriksen, 2 Niels-Henrik Holstein-Rathlou, 3 Fin Stolze Larsen, 4 and Flemming Bendtsen 1 Patients with advanced cirrhosis and ascites are characterized by circulatory dysfunction with splanchnic vasodilatation and renal vasoconstriction, which often lead to ascites. The vasoconstrictor terlipressin improves renal function in hepatorenal syndrome (HRS). The aim of this study was to evaluate if terlipressin also improves renal function in patients with ascites without HRS. Twenty-three patients with cirrhosis participated; 15 with nonrefractory ascites were randomized to either terlipressin (N group, n 11) or a placebo (P group, n 4), and 8 had refractory ascites and received terlipressin (R group). The glomerular filtration rate (GFR), sodium clearance (C Na ), lithium clearance (C Li ), osmolal clearance (C Osm ), and urine sodium concentration (U Na ) were assessed before and after the injection of 2 mg of terlipressin or the placebo. GFR increased in the N group (69 19 versus ml/min, P < 0.005) and in the R group (31 19 versus ml/min, P < 0.05) after terlipressin. In the N group, terlipressin induced an increase in C Na ( versus ml/min, P < 0.05), C Li ( versus ml/min, P < 0.05), and C Osm ( versus ml/min, P<0.05). In the R group, terlipressin induced an increase in C Na ( versus ml/min, P < 0.05) and C Li ( versus ml/min, P < 0.05). U Na increased in both groups after terlipressin (P < 0.005). Plasma norepinephrine (P < 0.05) and renin (P < 0.05) decreased after terlipressin. All parameters remained unchanged after the placebo. Conclusion: The vasopressin 1 receptor agonist terlipressin improves renal function and induces natriuresis in patients with cirrhosis and ascites without HRS. Vasoconstrictors may represent a novel future treatment modality for these patients. (HEPATOLOGY 2007;46: ) See Editorial on Page 1685 Abbreviations: ANP, atrial natriuretic peptide; CGRP, calcitonin gene related peptide; C H2O, free water clearance; C Li, lithium clearance; C Na, sodium clearance; C Osm, osmolal clearance; C x, renal clearance of substance x; DFRNa, distal fractional resorption of sodium; EDTA, ethylene diamine tetraacetic acid; FE, fractional excretion; FF, filtration fraction; GCP, good clinical practice; GFR, glomerular filtration rate; HR, heart rate; HRS, hepatorenal syndrome; MAP, mean arterial blood pressure; MELD, model for end-stage liver disease; NS, not significant; PFR, proximal fractional resorption; pro-anp, pro-atrial natriuretic peptide; P x, mean plasma concentration of substance x in the clearance period; RBF, renal blood flow; SD, standard deviation; U Li, urine lithium concentration; U Na, urine sodium concentration; U Osm, urine osmolality; U x, concentration in urine of substance x; V1, vasopressin 1-receptor; V2, vasopressin 2-receptor; V u, urinary flow rate. From the 1 Department of Gastroenterology, Hvidovre Hospital, 2 Department of Clinical Physiology, Hvidovre Hospital, 3 Department of Biomedical Sciences, Panum Institute, and 4 Department of Hepatology, Rigshospitalet, University of Copenhagen, Copenhagen, Denmark. Received May 16, 2007; accepted June 28, Supported by grants from the Lundbeck Foundation, the Hvidovre Hospital Foundation for Liver Disease, Ferring Pharmaceuticals Inc., and the University of Copenhagen (the last to A.K.). Address reprint requests to: Aleksander Krag, M.D., Department of Gastroenterology 439, Hvidovre University Hospital, DK-2650 Hvidovre, Denmark. aleksander.krag@hvh.regionh.dk; fax: Copyright 2007 by the American Association for the Study of Liver Diseases. Published online in Wiley InterScience ( DOI /hep Potential conflict of interest: The study was in part supported by an unrestricted grant from Ferring Pharmaceuticals Inc. Investigator conflicts of interest were disclosed to study participants. Ascites is a serious complication of cirrhosis, occurring in about 50% of patients within 10 years after diagnosis, and it is associated with 50% mortality in 2 years. 1,2 Patients with ascites have hyperdynamic circulation, which is characterized by peripheral and splanchnic arterial vasodilatation and reduced arterial blood pressure and systemic vascular resistance. 3 Portal hypertension and splanchnic vasodilatation are major factors in the development of ascites. 4,5 Secondary to vasodilatation, vasoactive hormones, such as the reninangiotensin-aldosterone system and the sympathetic nervous system, are activated. This leads to renal vasoconstriction and reduced renal perfusion and filtration pressure. The current standard treatment of ascites involves diuretics and paracentesis, which are not designed to improve the underlying pathophysiology. The changes in the renal handling of water and salt seem to develop stepwise from a stage that can be controlled by diuretics to a stage refractory to diuretics and finally to full-blown renal failure, which is known as hepatorenal syndrome (HRS) type 1. 6 Several clinical studies have evaluated the efficacy of the long-term administration of terlipressin [a vasopressin 1 (V1) receptor agonist] in patients with HRS, 7-13 and there is evidence that terlipressin may improve renal 1863

2 1864 KRAG ET AL. HEPATOLOGY, December 2007 function and survival in patients with HRS type 1. The mechanisms behind the effect of terlipressin in HRS have not been thoroughly investigated. Terlipressin seems to revert the systemic vasodilatation and increase blood pressure 14 and thereby improve renal perfusion pressure and renal function. As HRS is most often preceded by the development of ascites, vasoconstrictors may also have a beneficial effect in decompensated cirrhosis without severe renal impairment and thereby represent a novel future treatment for ascites. This is a pathophysiological study in which we investigated if the acute administration of the vasoconstrictor terlipressin improves renal perfusion, glomerular filtration, and renal excretion of salt and water in patients with cirrhosis with nonrefractory ascites and refractory ascites. Patients and Methods Patients. The study included 23 patients, between 18 and 75 years of age, with alcoholic cirrhosis and ascites; 2 patients also had a chronic hepatitis C infection. Of the patients, 15 with nonrefractory ascites were investigated in a randomized, double-blinded, placebo-controlled design, and 8 with diuretic-resistant ascites were investigated in an unblinded design. We chose not to randomize the patients with diuretic-resistant ascites because they are rare and, on account of the advanced disease, are difficult to handle in a clinical study. Patients were ineligible if they had HRS as defined by the International Ascites Club, 3 gastrointestinal bleeding within the week before the study, insulin-dependent diabetes mellitus, acute or chronic intrinsic renal or cardiovascular disease, arterial hypertension, abnormal urine analyses or electrocardiograms, hepatocellular carcinoma or portal vein thrombosis, or any other acute medical conditions such as infections or lung diseases. Furthermore, alcohol abstinence for 2 months was required. Refractory ascites was defined by the diagnostic criteria set by the International Ascites Club, 15 which are in brief as follows: ascites that cannot be mobilized or early recurrence that cannot be satisfactorily prevented by medical therapy. All patients with nonrefractory ascites had visible ascites verified by an abdominal ultrasound examination within the 2 months before the investigation. All patients received diuretics prior to the study. The diuretic treatment was administered according to the current guidelines with stepwise increases until a clinical effect or the development of hyperkalemia, hyponatremia, encephalopathy, or an increase in serum creatinine. 16 Five of the patients could not tolerate the maximum dose of diuretics, which was 160 mg of furosemide and 400 mg of spironolactone, because of an increase in creatinine or potassium. Patients with refractory ascites had a mean of 6 therapeutic paracenteses (range: 2-12) performed, although none were performed within the last week before the investigations. Diuretics and beta-blockers were temporarily discontinued 60 hours before the investigations. The patients were on a sodium-restricted diet (60 mmol/ day) the last 72 hours before the investigations. They were carefully instructed orally and given written information about the diet by a dietician. During the last 24 hours before the investigations, all patients were hospitalized, and the nutrition unit of the hospital prepared their food with a 60 mmol/day sodium diet. Design and Methods. A nurse who was unrelated to the study drew opaque sealed envelopes prepared by an external statistician. Among the 15 patients with nonrefractory ascites, 11 were randomized to 2 mg of terlipressin, and 4 patients were randomized to an infusion of a placebo. Only 4 received the placebo because of computer randomization. A computer made the randomization code with 21 envelopes, with one-third for the placebo and two-thirds for terlipressin, and only 15 were used. The 8 patients with refractory ascites all received 2 mg of terlipressin. The nurse also administered the infusion of terlipressin or the placebo. As the placebo, 10 ml of an isotonic saline solution was infused. Unblinding was not performed until the data had been typed into the database, the calculations had been performed, and the data had been monitored by the good clinical practice (GCP) unit. The patients were studied at 9:00 AM after a 9-hour fast. At 10:00 PM of the previous day, an oral dose of 300 mg of lithium carbonate was administered. An oral water load of 200 ml of tap water was given every half hour from 9:00 AM to the end of the clearance periods. The patients were in the supine position throughout the investigation. All patients had a bladder catheter placed before the clearance periods to ensure correct urine sampling. The infusion of tracers was prepared in 60 ml of isotonic saline with 16 MBq of 51 Cr ethylene diamine tetraacetic acid (EDTA; GE-Healthcare, Hilleroed, Denmark) and 10 MBq of 131 I-hippuran (GE-Healthcare). At 9:00 AM, a priming dose of 8 ml of the 51 Cr- EDTA and 131 I-hippuran solution was given as a rapid intravenous bolus injection together with 6.5 MBq of 51 Cr-EDTA, which was followed by a constant infusion of 8 ml/h (P 300 pump, Kivex, Hoersholm, Denmark) for 3.5 hours for a total of 10 MBq. 51 Cr-EDTA and 131 I-hippuran were used to determine the glomerular filtration rate (GFR) and effective renal plasma flow. After an equilibrium period of 2 hours, blood samples were drawn for analyses of the plasma lithium, sodium, osmolality, 51 Cr-EDTA, and 131 I-hippuran. The blood sam-

3 HEPATOLOGY, Vol. 46, No. 6, 2007 KRAG ET AL pling was immediately followed by urine collection from the bladder catheter. The urine volume was recorded, and the samples were assayed for lithium, sodium, osmolality, 51 Cr-EDTA, and 131 I-hippuran. A gamma counter (1480 Wizard 3, Wallac, Turku, Finland) was used to assess the radioactivity of 51 Cr-EDTA and 131 I-hippuran in the samples. The samplings were repeated at 30-minute intervals, and this resulted in 3 clearance periods after equilibration and before the intervention. Thereafter, patients with nonrefractory ascites were randomized to an infusion of 2 mg of terlipressin (Ferring Pharmaceuticals, Copenhagen, Denmark) or 10 ml of isotonic saline, whereas all patients with refractory ascites received 2 mg of terlipressin. Urine and blood sampling for renal function tests was then repeated for 3 more clearance periods of 30 minutes. Blood samples for vasoactive substances were obtained after the second clearance period at the baseline and after terlipressin or the placebo. Clearance during the steady state was calculated by the standard formula C x U x V u /P x, where C x is the renal clearance of substance x, U x is the concentration in urine of substance x, V u is the urinary flow rate, and P x is the mean plasma concentration of substance x in the clearance period. Lithium clearance (C Li ) is used as a marker of proximal sodium reabsorption under the assumption that lithium is filtered freely across the glomerulus and is reabsorbed in proportion to sodium and water in the proximal tubules and no reabsorption or secretion takes place in the distal tubules. 17 The fractional excretion (FE) of x is calculated as follows: FE C x /GFR. The free water clearance (C H2O ) is determined as follows: C H2O V u C Osm, where C Osm is the clearance of osmotic substances determined by the standard clearance formula. The proximal fractional resorption (PFR) of sodium and water is estimated as 1 C Li /GFR. 17 The distal fractional resorption of sodium (DFRNa) is estimated as 1 C Na /C Li, where C Na is the sodium clearance. 17 Plasma concentrations of norepinephrine were determined by high-performance liquid chromatography, as described elsewhere. 18 The intra-assay and interassay coefficients of variation were 8% and 9%, respectively. The plasma renin concentration was determined with a commercially available 2-site immunoradiometric assay (DGR International, Inc., Hamburg, Germany). The mean plasma concentration of renin in 536 healthy subjects was 26 pg/ml (range: ). 19 Aldosterone was measured with a commercial radioimmune assay kit (DSL-8600, Diagnostic Systems Laboratories, Inc., Webster, TX). The mean morning plasma concentration in 73 healthy adults in the supine position was 192 pmol/l (range: ). 19 Plasma calcitonin gene related peptide (CGRP) was analyzed radioimmunologically, as described elsewhere. 20 The intra-assay and interassay coefficients of variation were 4% and 7%, respectively. The mean concentration of CGRP in the plasma of normal subjects was 37 pmol/l (range: 24-50). 19 N-terminal pro-atrial natriuretic peptide (pro-anp; 1-30) was measured radioimmunologically with antiserum and calibrator material from Peninsula Laboratories (Los Angeles, CA) and a tracer prepared in house, as described elsewhere. 21 The intra-assay and interassay coefficients of variation were 2.3% and 6.8%, respectively. The mean plasma concentration in 48 healthy subjects was 1411 pg/ml (range: ). 19 Plasma and urinary lithium concentrations were measured by atomic absorption spectrophotometry (Perkin- Elmer 2380). Sodium in plasma and urine was measured by flame emission photometry (PerkinElmer 2380). Plasma and urine osmolarities were measured by the method of freezing-point depression (Osmomat 030-D, Gonotec, Berlin, Germany). The study was conducted according to GCP and was approved and monitored by the GCP unit of Copenhagen University Hospital. Furthermore, the study was approved and inspected by the Danish Medicines Agency (EudraCT no ). The regional ethics committee also approved the study (KF /04), which was registered on clinicaltrials.gov (NCT ). Statistics. For the estimation of patient numbers, a type 1 error of 0.05 and a type 2 error of 0.20 were chosen. The standard deviation (SD) of the estimate of renal function in these patients was approximately 20%. 22 In a paired design, 8 patients are required for the detection of a difference of 20% after terlipressin treatment. We did not expect any change in the renal function in the control group during the short study period. This assumption was based on a previous study, the same methods being applied without changes in the placebo group. 22 However, the stimulation of diuresis with 200 ml of water every half hour may increase C H2O. The results are expressed as the mean the SD of the 3 clearance periods before and after terlipressin or the placebo. Illustrating mean values ignores the time response; however, analyses using summary measures (for example, the final level and the change from the first measurement to the last) did not change the statistical outcomes. Vasoactive substances are additionally shown as medians and total ranges because of nonnormal distributions. Statistical analyses were performed with an unpaired Student t test, a Mann-Whitney test and a paired Student t test, or a Wilcoxon test, as appropriate. All reported P values are 2-tailed, with values less than 0.05 considered significant. The SPSS 10.1 statistical package (SPSS, Inc., Chicago, Il) was applied throughout the study.

4 1866 KRAG ET AL. HEPATOLOGY, December 2007 Table 1. Demographic, Clinical, and Biochemical Data R Group (Refractory Ascites Terlipressin; n 8) N Group (Nonrefractory Ascites Terlipressin; n 11) P Group (Nonrefractory Ascites Placebo; n 4) Age (years) Gender (male/female) 5/3 7/4 1/3 Child-Pugh score MELD score * Spironolactone (mg/day) Furosemide (mg/day) Beta-blocker treatment 5/8 5/11 0/4 Plasma coagulation factors II, VII, and X (units; ) Serum sodium (mmol/l; ) Serum creatinine ( mol/l; ) Serum albumin ( mol/l; ) Blood hemoglobin (mmol/l; males: , females: ) HR (minute 1 ) MAP (mm Hg) The means the SD are given. Reference intervals are given in parentheses. HR indicates heart rate; MAP, mean arterial blood pressure; and MELD, model for end-stage liver disease. *P 0.05 versus the placebo. P versus the placebo. Results The groups consisting of patients with nonrefractory ascites randomized to either terlipressin (N group) or the placebo (P group) were identical with respect to demographic, clinical, and biochemical variables (Table 1). The serum alanine aminotransferase was higher in the N group (P 0.05), but only 1 patient in the ascites group had a value of 113 U/L, which exceeded the reference interval. In the refractory ascites group (R group), the plasma albumin was, as expected, somewhat lower (P 0.05), and the model for end-stage liver disease score was borderline-significantly higher (P 0.06) than the P group (Table 1). After the infusion of terlipressin, the mean arterial blood pressure (MAP) increased by mm Hg (mean change SD; P 0.005) in the N group and by mm Hg (P 0.005) in the R group, whereas no change was observed in the P group (1 10 mm Hg). A corresponding decrease in the heart rate occurred in the N group (12 9 minutes 1, P 0.005) and in the R group (13 9 minutes 1, P 0.01). The heart rate was unaltered in the P group [ minutes 1, not significant (NS)]. Renal Perfusion and Glomerular Filtration [GFR, Renal Blood Flow (RBF), and Filtration Fraction (FF)]. At the baseline, there were no differences between the N group and the P group with respect to GFR, RBF (Fig. 1), and FF (Table 2); GFR (P 0.005), RBF (P 0.05; Fig. 1), and FF (P 0.05) were significantly lower in the R group than in the P and N groups. A significant increase in GFR of approximately 30% was observed in both treatment groups after terlipressin. In the N group, GFR increased from to ml/min (P 0.005), and in the R group, it increased from to Fig. 1. (A) GFR and (B) RBF (the mean the standard error of the mean). P 0.05 and *P versus the baseline. #P 0.05 and P versus the baseline in the placebo group.

5 HEPATOLOGY, Vol. 46, No. 6, 2007 KRAG ET AL Table 2. Effects of Terlipressin and the Placebo on Renal Function R Group (Refractory Ascites; n 8) N Group (Nonrefractory Ascites; n 11) P Group (Nonrefractory Ascites; n 4) Baseline After Terlipressin Baseline After Terlipressin Baseline Placebo FF (%) * V u (ml/min) C H2O (ml/min) U osm (mosm/kg) * U Li (mmol/l) * U Na (mmol/l) PFR (%) DFRNa (%) FeNa (%) * FeLi (%) * The means the SD are given. C H2O indicates free water clearance; DFRNa, distal fractional resorption of sodium; FeLi, fractional lithium excretion; FeNa, fractional sodium excretion; FF, filtration fraction; PFR, proximal fractional resorption; U Li, urine lithium concentration; U Na, urine sodium concentration; U Osm, urine osmolality; and V u, urinary flow rate. *P 0.05 (comparison between the baseline and after terlipressin or the placebo). P (comparison between the baseline and after terlipressin or the placebo). P 0.05 (comparison between the baseline in the P group and the baseline in the N group or R group). P (comparison between the baseline in the P group and the baseline in the N group or R group) ml/min (P 0.05). No change occurred in the placebo group (Fig. 1). RBF (ml/min) did not change significantly in the N and R groups ( versus ml/min, P 0.31, and versus ml/min, P 0.16, respectively). FF increased significantly in the N group ( versus , P 0.05; Table 2). There was no linear correlation between the improvement in GFR and RBF and the change in MAP. However, the improvement in GFR correlated with the Child-Pugh score (r 0.42, P 0.04). Renal Clearances of Electrolytes and Water. The effects of terlipressin and the placebo on C Na (ml/min), C Li (ml/min), C Osm (ml/min), and C H2O (ml/min) are shown in Fig. 2. In the R group, all clearances at the baseline were significantly lower than those in the P group (Fig. 2). In the N Fig. 2. Renal clearances (the mean the standard error of the mean): (F) refractory ascites and terlipressin, ( ) nonrefractory ascites and terlipressin, and ( ) nonrefractory ascites and placebo. C H2O indicates free water clearance; C Li, lithium clearance; C Na, sodium clearance; C Osm, osmolal clearance; period 1, baseline; and period 2, after terlipressin or the placebo. P 0.05 and *P versus the baseline. #P 0.05 and P versus the baseline in the placebo group.

6 1868 KRAG ET AL. HEPATOLOGY, December 2007 Table 3. Effects of Terlipressin and the Placebo on Vasoactive Hormones R Group (Refractory Ascites; n 8) N Group (Nonrefractory Ascites; n 11) P Group (Nonrefractory Ascites; n 4) Baseline After Terlipressin Baseline After Terlipressin Baseline Placebo Norepinephrine (nmol/l) 0.99 ( ) 0.59 ( )* 0.79 ( ) 0.52 ( )* 1.25 ( ) 1.03 ( ) Renin (pg/ml) 339 (7 4,003) 164 (7 2,221)* 17 (8 1,215) 8 (7 810)* 19 (9 214) 13 (7 253) Aldosterone (pmol/l) 2,229 (118 13,659) 2,269 (612 14,538) 468 (73 10,329) 573 (260 11,286)* 378 (114 1,729) 209 (106 1,574) CGRP (pmol/l) 125 (67 234) 129 (70 237)* 102 (30 185) 106 (29 205)* 107 (90 126) 99 (88 121) Pro-ANP (pg/ml) 1,876 (657 3,793) 3,237 (771 3,551)* 1,075 (477 2,069) 1,352 (1,008 2,462) 927 (871 1,634) 940 (885 1,575) The medians (ranges) are given. CGRP indicates calcitonin gene related peptide; and pro-anp, pro-atrial natriuretic peptide. *P 0.05 (comparison between the baseline and after terlipressin or the placebo). P (comparison between the baseline and after terlipressin or the placebo). group, the baseline C H2O value was lower than that in the P group (P 0.05). In the N group, terlipressin induced a significant increase in C Na ( versus ml/min, P 0.05), C Li ( versus ml/min, P 0.05), and C Osm ( versus ml/min, P 0.05) and a corresponding drop in C H2O ( versus ml/min, P 0.005). In the R group, terlipressin induced a significant increase in C Na ( versus ml/min, P 0.05) and C Li ( versus ml/min, P 0.05). The changes in C Osm ( versus ml/min) and C H2O ( versus ml/min) were not significant in the R group. On the contrary, in the P group, there were decreases in C Osm ( versus ml/min, P 0.05), C Li ( versus ml/min, NS), and C Na ( versus ml/min, NS) and an increase in C H2O ( versus ml/min, NS). Renal Sodium and Lithium Handling. At the baseline in the R group, the urinary flow rate (V u ; P 0.005), urinary lithium concentration (P 0.05), urine sodium concentration (U Na ; P 0.05), fractional sodium excretion (P 0.005), and fractional lithium excretion (P 0.005) were significantly lower than those in the P group, and the proximal and distal fractional sodium reabsorption (PFR, P 0.005; DFRNa, P 0.05) and urine osmolality (U Osm ; P 0.05) were significantly higher (Table 2). In the N group, U Osm (P 0.05) and U Na (P 0.05) were higher than those in the P group. After terlipressin, U Osm increased by 17% in the N group (P 0.05), whereas no change was observed in the R group. In the P group, U Osm decreased by 40% (P 0.07) after the placebo (Table 2). V u did not change in any of the groups. U Na increased significantly in both groups after terlipressin (P 0.005, Table 2). The urinary lithium concentration was unaltered after terlipressin but exhibited a significant drop in the P group (P 0.05; Table 2). PFR and DFRNa did not change after terlipressin. The fractional sodium excretion increased by 233% in the R group (P 0.05, Table 2) and showed a borderline-significant increase in the N group (39%, P 0.07). The fractional lithium excretion did not change after terlipressin in contrast to a significant decrease in the P group (P 0.05). Vasoactive Hormones. Terlipressin induced a significant reduction in the circulating levels of norepinephrine and renin in both the N and R groups (P 0.05) and a significant increase in pro-anp (P 0.05) and CGRP (P 0.05; Table 3). We observed an increase in aldosterone in the N group (median increase: 9%, P 0.05) and no change in the R group ( 5%, NS). No change in epinephrine was observed. All hormones remained unchanged in the P group. Adverse Events. Eleven of the patients had loose stools. Three had abdominal cramps, and 1 also had nausea and vomited. All observed side effects were self-limiting. Discussion The major new findings in this pathophysiological study are that the acute effect of the V1 receptor agonist terlipressin induces a significant improvement in the renal function in patients with refractory and nonrefractory ascites without HRS. It improves the GFR and the urinary clearances of sodium, lithium, and osmoles and ameliorates the activation of potent vasoactive systems responsible for renal vasoconstriction and renal sodium and water retention in patients with ascites without HRS. Several clinical studies have evaluated the efficacy of terlipressin in HRS Only a few studies have investigated the renal effects of vasoconstrictors in patients with ascites without HRS. Two different receptor systems have been approached: the V1 vasopressin receptors and the alpha-1-adrenoceptors. In 8 patients with refractory ascites without HRS, Gadano et al. 23 found an increase in the renal perfusion pressure and RBF after terlipressin but no significant change in GFR or urinary sodium excretion. However, these findings may have been blurred, as the urine production was not stimulated by oral water ingestion and a catheter did not collect the urine, and only 3 patients had

7 HEPATOLOGY, Vol. 46, No. 6, 2007 KRAG ET AL detectable urine sodium. Furthermore, renal clearances of electrolytes and water were not determined. In a recent study, 12 patients with ascites were treated with the alpha-1-adrenergic agonist midodrine for 7 days. 24 Following a delay of 3-5 days, a significant increase in GFR, U Na, V u, and creatinine clearance was seen. In another study on acute effects of midodrine, an improvement in GFR and U Na was observed in patients with ascites, and no effect was seen in patients with type 2 HRS. 25 Although the V1 receptor agonists induce splanchnic vasoconstriction, the effects of alpha-1-adrenergic agonists on splanchnic vasodilatation, which is considered a key factor in the pathogenesis of ascites, 5 are unknown in patients with cirrhosis. Some of the patients in the treatment groups were treated with beta-blockers (Table 1). For ethical reasons, we could not discontinue beta-blocker therapy for more than 3 days in patients with esophageal varices; however, beta-blockers do not seem to affect RBF and GFR, 26,27 and as propranolol has a half-life of 3-6 hours, the pharmacological effect at that point is probably negligible. None of the patients in the P group were on beta-blocker therapy because the 2 patients (50%) who had indications for beta-blocker therapy both had intolerance to beta-blockers. An important factor contributing to ascites in patients with portal hypertension is splanchnic vasodilatation. 4-6 In this study, we therefore approach the treatment of ascites with the understanding that ascites is due to splanchnic arterial vasodilatation, arterial underfilling, reduced renal perfusion pressure, and activation of vasoconstrictors (that is, the sympathetic nervous system and reninangiotensin-aldosterone system). Terlipressin activates the V1 receptors, which are predominantly located in the vasculature in the splanchnic region, causes vasoconstriction, 28 and thereby reduces the splanchnic arterial vasodilatation and portal pressure and ameliorates the hyperdynamic circulation. 14,29 This improves the effective circulatory volume and renal perfusion pressure. 23,29 In this study, we showed that these improvements in hemodynamics are associated with an increase in GFR and a deactivation of vasoconstrictors and sodium-conserving hormones (that is, norepinephrine and renin). We observed an increase in C Li, which reflects an increase in the delivery of sodium and water from the proximal straight tubules to the loops of Henle. 17,30 As PFR is unchanged, the increase in C Li is most likely the result of the increase in GFR. Because the kidneys maintain the glomerulotubular balance, an increased GFR will, in general, be associated with a nearly constant proximal fractional reabsorption. The result of the increase in GFR and the delivery of sodium and water from the proximal straight tubules (C Li ) is increased sodium excretion, as seen by the increase in C Na and C Osm and the increase in U Na and U Osm. The use of C Li as a marker of intrarenal sodium handling in patients with a severe reduction of fractional sodium clearance has been questioned. The main reason is a potential distal reabsorption of lithium. However, in a study by Angeli et al. 31 of patients with cirrhosis with ascites and avid sodium retention, no reabsorption in the distal tubule was found. 31 A negative C H2O value in the treatment groups represents hypertonic urine, in contrast to the urine in the placebo group, which becomes more hypotonic. The mean value of C H2O in the P group in the 3 baseline clearance periods was higher (Fig. 2B) than that in the N and R groups. The levels of C H2O were identical in the first baseline clearance period; however, the P group showed a more pronounced increase in C H2O during the baseline clearance periods (data not shown). This probably represents a slight difference in the ability to suppress vasopressin in response to the 200-mL oral water load that the patients were given every half hour. 32 This also held true in the N group; however, in the R group, C H2O did not change. This corresponds to the nonrefractory and refractory disease stages, respectively. Although the major reasons for increased natriuresis after terlipressin are most likely the increased MAP and the improvement in the hemodynamics, the natriuretic effect of terlipressin will be enhanced by decreased sympathetic nerve activity (decreased plasma norepinephrine), decreased renin secretion, and increased atrial natriuretic peptide (ANP) secretion. The mechanism of release of ANP from the heart is mainly facilitated by an increase in the atrial pressure and stretching of the atria. However, an additional pressure-independent release may exist. It has been shown that ANP secretion increases in response to vasopressin analogs and pressor agents. 33 High activity in the sympathetic nervous system decreases RBF by alpha-1-adrenergic receptors and increases sodium reabsorption and renin secretion by beta-1-adrenergic receptors in the juxtaglomerular cells. 34,35 The decrease in norepinephrine therefore contributes to the increased perfusion and increased sodium excretion. The increase in plasma aldosterone in group R was, however, small (median 9%) in light of the range of 72-10,320, despite a reduction of plasma renin probably due to the delay in this system. Thus, the half-life of aldosterone in cirrhosis is rather long (mean: minutes). 36,37 Perspectives. Although this is a pathophysiological study of the acute effects of terlipressin and extensions to therapy should be conducted very carefully, the study has implications for further research and a new hypothesis. Dosage regimens and especially ischemic side effects should be further evaluated. In patients with HRS treated with terlipressin, about 15% discontinue therapy because of side ef-

8 1870 KRAG ET AL. HEPATOLOGY, December 2007 fects However, we believe that our study provides a proof of concept of the arterial vasodilation hypothesis of sodium retention in cirrhosis. From a pathophysiological viewpoint, the perfect drug for treating ascites in cirrhosis would be a combination of the V1 receptor agonist and vasopressin 2 (V2) receptor antagonist. The V1 receptor agonist terlipressin ameliorates the hyperdynamic circulation and increases GFR and the delivery of tubular fluid to the distal nephron, causing natriuresis. However, the increased natriuresis is associated with increased U Osm and decreased C H2O. The decreased C H2O value points to increased water reabsorption in the collecting ducts, which are the sites of action of the V2 receptor antagonists. The V2 receptor antagonists are drugs that interfere with the renal effects of antidiuretic hormones and inhibit water reabsorption in the collecting ducts and cause aquaresis. 38 A V2 receptor antagonist has been proven effective in patients with cirrhosis and hyponatremia. 39 However, the isolated effect of aquaretics in patients with ascites without hyponatremia is questionable as natriuresis is essential for the treatment and there is a risk of hypernatremia. In conclusion, the V1 receptor agonist terlipressin improves renal function and induces natriuresis in patients with cirrhosis and ascites without HRS. Vasoconstrictors may represent a novel future treatment modality for these patients. References 1. D Amico G, Morabito A, Pagliaro L, Marubini E. Survival and prognostic indicators in compensated and decompensated cirrhosis. Dig Dis Sci 1986; 31: Gines P, Quintero E, Arroyo V, Teres J, Bruguera M, Rimola A, et al. Compensated cirrhosis: natural history and prognostic factors. HEPATOL- OGY 1987;7: Arroyo V, Gines P, Gerbes AL, Dudley FJ, Gentilini P, Laffi G, et al., for the International Ascites Club. Definition and diagnostic criteria of refractory ascites and hepatorenal syndrome in cirrhosis. HEPATOLOGY 1996;23: Schrier RW, Arroyo V, Bernardi M, Epstein M, Henriksen JH, Rodes J. Peripheral arterial vasodilation hypothesis: a proposal for the initiation of renal sodium and water retention in cirrhosis. HEPATOLOGY 1988;8: Gines P, Cardenas A, Arroyo V, Rodes J. Management of cirrhosis and ascites. N Engl J Med 2004;350: Arroyo V, Colmenero J. Ascites and hepatorenal syndrome in cirrhosis: pathophysiological basis of therapy and current management. J Hepatol 2003;38(suppl 1):S69-S Halimi C, Bonnard P, Bernard B, Mathurin P, Mofredj A, di Martino V, et al. Effect of terlipressin (Glypressin) on hepatorenal syndrome in cirrhotic patients: results of a multicentre pilot study. Eur J Gastroenterol Hepatol 2002;14: Moreau R, Durand F, Poynard T, Duhamel C, Jean P, Ichai P, et al. Terlipressin in patients with cirrhosis and type 1 hepatorenal syndrome: a retrospective multicenter study. Gastroenterology 2002;122: Mulkay JP, Louis H, Donckier V, Bourgeois N, Adler M, Deviere J, et al. Long-term terlipressin administration improves renal function in cirrhotic patients with type 1 hepatorenal syndrome: a pilot study. Acta Gastroenterol Belg 2001;64: Ortega R, Gines P, Uriz J, Cardenas A, Calahorra B, De Las HD, et al. Terlipressin therapy with and without albumin for patients with hepatorenal syndrome: results of a prospective, nonrandomized study. HEPATOL- OGY 2002;36: Solanki P, Chawla A, Garg R, Gupta R, Jain M, Sarin SK. Beneficial effects of terlipressin in hepatorenal syndrome: a prospective, randomized placebo-controlled clinical trial. J Gastroenterol Hepatol 2003;18: Uriz J, Gines P, Cardenas A, Sort P, Jimenez W, Salmeron JM, et al. Terlipressin plus albumin infusion: an effective and safe therapy of hepatorenal syndrome. J Hepatol 2000;33: Sanyal A, Boyer TD, Garcia-Tsao G, Regenstein F, Rossaro L, Teuber P. A prospective, randomized, double blind, placebo-controlled trial of terlipressin for type 1 hepatorenal syndrome. HEPATOLOGY 2006;44(suppl 1):LB Moller S, Hansen EF, Becker U, Brinch K, Henriksen JH, Bendtsen F. Central and systemic haemodynamic effects of terlipressin in portal hypertensive patients. Liver 2000;20: Moore KP, Wong F, Gines P, Bernardi M, Ochs A, Salerno F, et al. The management of ascites in cirrhosis: report on the consensus conference of the International Ascites Club. HEPATOLOGY 2003;38: Runyon BA. Management of adult patients with ascites due to cirrhosis. HEPATOLOGY 2004;39: Thomsen K. Lithium clearance: a new method for determining proximal and distal tubular reabsorption of sodium and water. Nephron 1984;37: Moller S, Becker U, Schifter S, Abrahamsen J, Henriksen JH. Effect of oxygen inhalation on systemic, central, and splanchnic haemodynamics in cirrhosis. J Hepatol 1996;25: Moller S, Norgaard A, Henriksen JH, Frandsen E, Bendtsen F. Effects of tilting on central hemodynamics and homeostatic mechanisms in cirrhosis. HEPATOLOGY 2004;40: Bendtsen F, Schifter S, Henriksen JH. Increased circulating calcitonin gene-related peptide (CGRP) in cirrhosis. J Hepatol 1991;12: Melander O, Frandsen E, Groop L, Hulthen UL. Plasma ProANP(1-30) reflects salt sensitivity in subjects with heredity for hypertension. Hypertension 2002;39: Ottesen LH, Aagaard NK, Kiszka-Kanowitz M, Rehling M, Henriksen JH, Pedersen EB, et al. Effects of a long-acting formulation of octreotide on renal function and renal sodium handling in cirrhotic patients with portal hypertension: a randomized, double-blind, controlled trial. HEPA- TOLOGY 2001;34: Gadano A, Moreau R, Vachiery F, Soupison T, Yang S, Cailmail S, et al. Natriuretic response to the combination of atrial natriuretic peptide and terlipressin in patients with cirrhosis and refractory ascites. J Hepatol 1997; 26: Kalambokis G, Fotopoulos A, Economou M, Pappas K, Tsianos EV. Effects of a 7-day treatment with midodrine in non-azotemic cirrhotic patients with and without ascites. J Hepatol 2007;46: Angeli P, Volpin R, Piovan D, Bortoluzzi A, Craighero R, Bottaro S, et al. Acute effects of the oral administration of midodrine, an alpha-adrenergic agonist, on renal hemodynamics and renal function in cirrhotic patients with ascites. HEPATOLOGY 1998;28: Bernardi M, De PR, Trevisani F, Tame MR, Ciancaglini GC, Pesa O, et al. Renal function and effective beta-blockade in cirrhosis with ascites. Relationship with baseline sympathoadrenergic tone. J Hepatol 1989;8: Stanley AJ, Bouchier IA, Hayes PC. Acute effect of propranolol and isosorbide-5-mononitrate administration on renal blood flow in cirrhotic patients. Gut 1998;42: Hirasawa A, Shibata K, Kotosai K, Tsujimoto G. Cloning, functional expression and tissue distribution of human cdna for the vascular-type vasopressin receptor. Biochem Biophys Res Commun 1994;203: Kiszka-Kanowitz M, Henriksen JH, Hansen EF, Moller S, Bendtsen F. Effect of terlipressin on blood volume distribution in patients with cirrhosis. Scand J Gastroenterol 2004;39: Thomsen K, Shirley DG. The validity of lithium clearance as an index of sodium and water delivery from the proximal tubules. Nephron 1997;77: Angeli P, De BE, Dalla PM, Caregaro L, Ceolotto G, Albino G, et al. Effects of amiloride on renal lithium handling in nonazotemic ascitic cirrhotic patients with avid sodium retention. HEPATOLOGY 1992;15:

9 HEPATOLOGY, Vol. 46, No. 6, 2007 KRAG ET AL Bichet D, Szatalowicz V, Chaimovitz C, Schrier RW. Role of vasopressin in abnormal water excretion in cirrhotic patients. Ann Intern Med 1982; 96: Manning PT, Schwartz D, Katsube NC, Holmberg SW, Needleman P. Vasopressin-stimulated release of atriopeptin: endocrine antagonists in fluid homeostasis. Science 1985;229: Bichet DG, Van Putten VJ, Schrier RW. Potential role of increased sympathetic activity in impaired sodium and water excretion in cirrhosis. N Engl J Med 1982;307: Hackenthal E, Paul M, Ganten D, Taugner R. Morphology, physiology, and molecular biology of renin secretion. Physiol Rev 1990;70: Coppage WS, Island DP, Cooner AE, Liddle GW. The metabolism of aldosterone in normal subjects and in patients with hepatic cirrhosis. J Clin Invest 1962;41: Vecsei P, Dusterdieck G, Jahnecke J, Lommer D, Wolff HP. Secretion and turnover of aldosterone in various pathological states. Clin Sci 1969;36: Yeates KE, Morton AR. Vasopressin antagonists: role in the management of hyponatremia. Am J Nephrol 2006;26: Schrier RW, Gross P, Gheorghiade M, Berl T, Verbalis JG, Czerwiec FS, et al. Tolvaptan, a selective oral vasopressin V2-receptor antagonist, for hyponatremia. N Engl J Med 2006;355: