ALEXANDER L. GERBES, 1 VEIT GÜLBERG, 1 TOBIAS WAGGERSHAUSER, 2 JOSEF HOLL, 1 AND MAXIMILIAN REISER 2

Renal Effects of Transjugular Intrahepatic Portosystemic Shunt in Cirrhosis: Comparison of Patients With Ascites, With Refractory Ascites, or Without Ascites ALEXANDER L. GERBES, 1 VEIT GÜLBERG, 1 TOBIAS WAGGERSHAUSER, 2 JOSEF HOLL, 1 AND MAXIMILIAN REISER 2 Renal effects of the transjugular intrahepatic portosystemic shunt (TIPS) were compared in 6 patients without ascites (group 1), 11 patients with ascites responding to diuretic treatment (group 2), and 6 patients with refractory ascites (group 3). Seven days after insertion of TIPS, 24-hour urinary sodium excretion had increased in patients with ascites: 113 16 mmol to 170 30 mmol (P.012) in group 2, and 22 8 mmol to 77 27 mmol (P.050) in group 3. In group 3, fractional sodium excretion tended to increase from 0.26% 0.14% to 0.62% 18% (P.081). The relative increase of urinary sodium excretion (to 444% 122%) and fractional sodium excretion (to 413% 127%) in group 3 was significantly (P F.05) higher than in group 1 and group 2, respectively. Creatinine clearance and 24-hour urinary volume were not significantly changed in either group. Patients with Child-Pugh class C had a more pronounced effect of TIPS on urinary sodium excretion (increase to 396% 115% vs. 139% 15%; P.066) and on fractional sodium excretion (increase to 415% 103% vs. 94% 15%; P.020) than patients with less-severe liver disease. Fractional sodium excretion of less than 0.35% before TIPS was found to be an indicator of renal response to TIPS. The effect of TIPS on urinary sodium excretion and on fractional sodium excretion was related to the patients Child-Pugh score (r.55; P.007 and r.68; P.001, respectively) and inversely to their fractional sodium excretion (r.44; P.047 and r.54; P.012, respectively) before TIPS. These data demonstrate that TIPS affects renal sodium handling in patients with ascites, particularly in patients with refractory ascites. Severity of liver disease and fractional sodium excretion before TIPS are parameters to predict the extent of the renal response. (HEPATOLOGY 1998;28:683-688.) The transjugular intrahepatic portosystemic shunt (TIPS), a nonsurgical side-to-side shunt, has been introduced to Abbreviation: TIPS, transjugular intrahepatic portosystemic shunt. From the 1 Department of Medicine II, and 2 Institute of Diagnostic Radiology, Klinikum Groβhadern, Ludwig-Maximilians-Universität München, Germany. Received June 20, 1997; accepted April 21, 1998. Presented in part at the annual meeting of the AASLD, Chicago, IL, November 7-12, 1997. Supported in part by grant Ge 576/13-1 by the Deutsche Forschungsgemeinschaft to A. L. Gerbes. Address reprint requests to: Alexander L. Gerbes, M.D., Professor of Medicine, Department of Medicine II, Klinikum Groβhadern, Ludwig-Maximilians-University of Munich, Marchioninistr. 15, 81377 Munich, Germany. Fax: 49-89-7095-2392. Copyright 1998 by the American Association for the Study of Liver Diseases. 0270-9139/98/2803-0013$3.00/0 683 prevent variceal rebleeding or to control refractory variceal bleeding by reducing the portal pressure gradient. 1-4 A considerable portion of these patients who had ascites at the time of TIPS insertion reduced the degree of ascites following TIPS. Therefore, patients with refractory or recurrent ascites were treated with TIPS, most of them with remarkable success. 5-9 These observations prompted studies on the effects of TIPS on sodium handling in patients with refractory ascites. Little is known, however, about the effects of TIPS on renal function in patients without ascites or with ascites not refractory to diuretic treatment. Moreover, predictors of renal response to TIPS have not yet been identified. Furthermore, information about the early influence of TIPS on renal function is limited. 6-13 Most of the previous studies were performed without diuretics or with the dose of diuretics decreasing throughout the observation period. While withholding diuretic treatment from those patients does not represent a practical clinical approach, the change of dosage of diuretics could disguise the effects of TIPS per se. Therefore, diuretic treatment was continued in the present study, and renal function was evaluated before and 7 days after TIPS. The aim of our study was to compare the effects of TIPS on renal function in patients with refractory ascites to patients with ascites not refractory to diuretic treatment and to patients without ascites, and to identify predictors of renal response. PATIENTS AND METHODS The investigational protocol was approved by the ethics committee of the medical faculty of the Ludwig-Maximilians-University of Munich, and all patients gave informed written consent. Patients. Twenty-three patients with cirrhosis with uncomplicated short-term follow-up after TIPS were included and studied prospectively. All patients were in stable condition and had been hospitalized for at least 1 week. During the week preceding the TIPS, there had been no variceal bleeding, and diuretic treatment was held constant. As proposed by Roessle et al. 1 the following exclusion criteria for TIPS insertion were applied: serum bilirubin 85 µmol/l, previous hepatic encephalopathy grade 2 or more unrelated to bleeding or diuretic treatment, primary or secondary hepatic tumors, and severe congestive heart failure. Group 1 comprised 6 patients who received TIPS for prophylaxis of rebleeding in whom presence of ascites was excluded by ultrasound examination of the abdomen. Group 2 consisted of 11 patients with ascites confirmed by clinical examination and ultrasound sonography who received TIPS for prophylaxis of variceal rebleeding. Patients of group 2 received spironolactone (130 25 mg/d) and furosemide (44 11 mg/d). In patients of group 1 and group 2, TIPS was inserted between day 8 and day 12 after the bleeding episode. Six patients who had TIPS for refractory ascites

684 GERBES ET AL. HEPATOLOGY September 1998 formed group 3. Refractory ascites was defined according to a consensus conference 14 as ascites that could not be controlled by maximum doses of spironolactone and furosemide tolerated without significant side-effects. In this group, the daily doses of spironolactone and furosemide averaged 250 34 mg and 53 20 mg, respectively. In all patients, diuretic treatment was held constant until 7 days following insertion of TIPS. Patients were hospitalized throughout the study period and on a diet containing approximately 110 mmol (groups 1 and 2) and 50 mmol (group 3) sodium per day. Severity of liver disease was classified according to Child-Pugh. 15 Study Protocol. A 24-hour urine collection was performed the day before TIPS and 7 days after TIPS insertion. Blood pressure and heart rate were monitored in the morning, after the patients had remained for at least 1 hour in the supine position, and blood samples were withdrawn. The day before, the day after, and 7 days after TIPS, duplex Doppler sonography of the hepatic artery, portal vein, and of the blood flow within the TIPS stent was performed. Before TIPS, patients received a single intravenous dose of a third-generation cephalosporine. The TIPS procedure was performed by a standard technique. Palmaz stents (Johnson & Johnson, Norderstedt, Germany) were used in 15 patients, Memotherm stents (Bard- Angiomed, Karlsruhe, Germany) in 5 patients, and Wall stents (Schneider, Zürich, Switzerland) in 3 patients, respectively. Before and after TIPS insertion, pressure in the portal vein and inferior caval vein was determined; the difference is given as the hepaticvenous pressure gradient. During 1 week after TIPS insertion, partial thromboplastin time was held at 50 to 60 seconds, by use of an intravenous heparin infusion. Laboratory Analysis. Serum and urinary sodium and creatinine were measured using an ion-selective probe (Boehringer Mannheim/ Hitachi rack 917, Mannheim, Germany) and the Jaffe reaction, respectively. Plasma concentration of active renin was determined by an immunoradiometric assay method using a monoclonal antibody against the active site of the renin molecule (Nichols Institute, San Juan Capistrano, CA; normal range, 9-29 µu/ml). Norepinephrine was determined by high-performance liquid chromatography using an electrochemical detection device (internal standard: dihydrobenzylamine; normal range, 185-275 pg/ml). Serum aldosterone concentration was measured by a commercial nonextraction radioimmunoassay (Coat-a-count aldosterone RIA, DPC, Los Angeles, CA; normal range, 20-150 pg/ml) that exhibits no significant cross-reactivity to other steroids. Atrial natriuretic peptide and endothelin-1 were determined as described in detail elsewhere. 16,17 Statistical Analysis. All results are given as mean and standard error unless stated otherwise. Within-group differences before TIPS and 7 days after TIPS were calculated by the paired t test. Differences between groups were calculated by ANOVA, and post-hoc analysis was performed with the Student-Newman-Keuls test. P.05 was considered statistically significant. Correlation coefficients were calculated as Pearson s r using the least-squares method. Because of its skewed distribution, fractional sodium excretion was analyzed after logarithmic transformation. RESULTS Patient Characteristics. The degree of portal hypertension was comparable in all three groups as shown in Table 1. Child-Pugh score of groups 2 and 3 (with ascites) was higher than of group 1 (without ascites). Neurohumoral substances were determined in 5 patients of group 1, 6 patients of group 2, and 4 patients of group 3. Significant differences were found for plasma concentrations of active renin (groups 1 vs. 2 vs. 3: 40 1 vs. 272 99 vs. 1,504 464 µu/ml) and of aldosterone (251 200 vs. 504 122 vs. 1,140 314 pg/ml). Differences (groups 1 vs. 2 vs. 3) did not reach significance for plasma concentrations of norepinephrine (307 44 vs. 285 49 vs. 869 211 pg/ml), atrial TABLE 1. Characterization of Patients Group 1 (without ascites) Group 2 (with ascites) Group 3 (with refractory ascites) Number 6 11 6 Sex (male/female) 3/3 10/1 5/1 Age (median, range) (yr) 47 (32-62) 54 (28-73) 55 (47-64) Etiology of cirrhosis (alcoholic/postnecrotic/cryptogenic) 4/1/1 10/1/ 5/ /1 Systolic blood pressure (mm HG) 117 4 111 4 110 6 Diastolic blood pressure (mm HG) 70 4 69 3 67 5 Heart rate (min 1 ) 78.7 3.4 78.2 4.6 82.0 5.1 Hepatic-venous pressure gradient (mm HG) 22.2 2.9 27.5 1.0 22.0 2.1 Child-Pugh score 6.2 0.3 8.2 0.4* 9.3 0.6* *P.05 vs. group 1. natriuretic peptide (3.5 1.1 vs. 5.7 1.4 vs. 8.0 3.9 fmol/ml), and endothelin-1 (14.0 0.8 vs. 17.2 1.1 vs 17.2 0.8 pg/ml). Effects of TIPS. The hepatic-venous pressure gradient before TIPS was not different between the groups (Table 1). TIPS insertion caused a similar decrease of hepatic-venous pressure gradients from 22.2 2.9 to 12.6 2.0 mm Hg in group 1, from 27.5 1.0 to 13.9 1.8 mm Hg in group 2, and from 22.0 2.1 to 11.4 0.7 mm Hg in group 3, respectively. There was no significant effect of TIPS on systolic or on diastolic blood pressure, nor on heart rate, in either group. Renal function after TIPS is shown in Table 2. There was no significant effect of TIPS on creatinine clearance or on 24-hour urinary volume in either group. In group 1, TIPS did not affect 24-hour urinary sodium excretion or fractional sodium excretion. In contrast, 24-hour sodium excretion in group 3 increased (P.050) from 22 8 mmol to 77 27 mmol. This effect was also significant in group 2 (from 113 16 mmol to 170 30 mmol; P.012). The patients with refractory ascites (group 3) exhibited a trend toward an increase of fractional sodium excretion from 0.26% 0.14% to 0.62% 0.18% (P.081). The effects of TIPS on 24-hour TABLE 2. Renal Function Before and 7 Days After TIPS Group 1 (without ascites) Group 2 (with ascites) Group 3 (with refractory ascites) Creatinine clearance (ml/min) before 116 22 81 13 55 15 day 7 119 9 95 13 63 8 24-h urinary volume (ml) before 2,028 471 1,462 176 898 130 day 7 1,783 199 1,664 192 1,490 216 24-h urinary sodium excretion (mmol) before 131 39 113 16 22 8 day 7 124 18 170 30* 77 27* Fractional sodium excretion (%) before 0.65 0.18 0.83 0.16 0.26 0.14 day 7 0.55 0.14 0.89 0.15 0.62 0.18 *P.05 vs. before TIPS.

HEPATOLOGY Vol. 28, No. 3, 1998 GERBES ET AL. 685 urinary volume and creatinine clearance were not different between the groups (Fig. 1). In contrast, the mean individual increase in 24-hour urinary sodium excretion in group 3 by 444% 122% was significantly higher than in group 1 (114% 18%) and group 2 (150% 20%), respectively. Similarly, mean fractional sodium excretion increased to 413% 127% in group 3, significantly higher than in group 1 (72% 20%) and in group 2 (139% 30%), respectively (Fig. 1). Following TIPS, plasma aldosterone and active renin concentrations decreased significantly in patients with ascites. The influence of TIPS on vasoactive neurohumoral substances is summarized in Table 3. Indicators of Renal Effects. The change of urinary sodium excretion was found to correlate to the Child-Pugh score in the total of 23 patients investigated (r.55; P.007), as well as in the patients with ascites (r.50; P.041). Similarly, there was a highly significant correlation between the change of fractional sodium excretion after TIPS and the Child-Pugh score in all patients (r.68; P.001) (Fig. 2), and in patients with ascites (r.66; P.005). The effect of TIPS on urinary sodium excretion was inversely related to fractional sodium excretion before TIPS in the total of patients investigated (r.44; P.047), as well as in the groups comprising patients with ascites (r.48; P.058) (Fig. 3). Fractional sodium excretion before TIPS was found to correlate inversely to TIPS-induced changes of fractional sodium excretion in the total of patients (r.54; P.012), as well as in the patients with ascites (r.61; P.013). Plasma concentrations of active renin before TIPS were significantly correlated to TIPS-induced changes of urinary sodium excretion and to fractional sodium excretion in all patients (r.71; P.003 and r.80; P.001, respectively), and in patients with ascites (r.70; P.025 and r.79; P.007, respectively). Severity of liver disease with Child-Pugh class C vs. A and B and fractional sodium excretion before TIPS with a cut-off value of 0.35% were found to discriminate the effects on renal sodium excretion and on fractional sodium excretion (Table 4). DISCUSSION Our results demonstrate that: 1) under practical clinical conditions, TIPS increases renal sodium excretion in patients with ascites, particularly with refractory ascites, but not in patients without ascites; 2) this can be observed 7 days after TIPS insertion; 3) in patients with refractory ascites, an increase in fractional renal sodium excretion seems to account for the change of sodium handling; 4) severity of liver disease as indicated by the Child-Pugh score and pre-tips fractional sodium excretion correlate to the effects of TIPS on renal sodium handling; and 5) patients with Child-Pugh class C or a fractional sodium excretion of less than 0.35% exhibit a marked increase of renal sodium excretion and fractional sodium excretion upon TIPS. Several studies observed a reduction of ascites following insertion of TIPS in patients with cirrhosis. 1-4 The influence of TIPS on renal function, however, has been investigated only in patients with refractory or recurrent ascites. 6-13 Compared with a practical clinical setting, most of these studies lack important information: they investigated renal function the day after TIPS, when it might be influenced by the procedure itself (e.g., contrast medium, analgesics, or anesthesia) and at 1 month, when it might be influenced by other factors (e.g., influence of TIPS on the nutritional status). So far, there has been little information on the renal function at 1 week after TIPS, the interval when most patients will be considered for release of hospital care. One of the studies 8 completely withheld diuretic treatment for all patients during 1 month after TIPS and placed patients on a severely sodium-restricted diet, which is clearly not a practical approach (comment by Siuta 11 ). In other studies, 6,7,12 diuretic treatment was changed following TIPS. While this may be advisable in many patients several weeks after TIPS, it precludes an analysis of the effects of TIPS per se on renal sodium handling. We therefore decided to investigate patients 7 days after insertion of TIPS. Furthermore, we selected patients in a stable clinical condition. Patients with ascites FIG. 1. Effect of TIPS on renal function in patients without ascites (group 1 [ ]), patients with ascites (group 2 [i]), and patients with refractory ascites (group 3 [ ]). (A) Twenty-four hour urinary volume; (B) creatinine clearance; (C) 24- hour sodium excretion; (D) fractional sodium excretion., P.05 vs. group 1; ( ), P.05 vs. group 2.

686 GERBES ET AL. HEPATOLOGY September 1998 TABLE 3. Neurohumoral Vasoactive Substances Before and 7 Days After TIPS All Patients Patients With Ascites Active renin (µu/ml) before 523 198 765 269 day 7 436 189 630 268* Aldosterone (pg/ml) before 589 142 758 170 day 7 305 117* 423 164* Norepinephrine (pg/ml) before 629 106 692 154 day 7 448 88 519 126 Atrial natriuretic peptide (fmol/ml) before 5.6 1.1 6.5 1.6 day 7 8.9 1.7* 10.1 2.2 Endothelin-1 (pg/ml) before 16.2 0.7 17.2 0.8 day 7 15.2 0.5 15.6 0.6 *P.05 vs. before TIPS. FIG. 3. Effects of TIPS on urinary sodium excretion (as percentage of pre-tips urinary sodium excretion) are related inversely to the fractional sodium excretion (data of one patient not available) before TIPS in patients with ascites (r.66; P.006). Because of skewed distribution of data, logarithmic transformation of fractional sodium excretion was performed. were treated with diuretics to achieve a balanced sodium homeostasis. In patients with refractory ascites, sodium balance obviously could not be maintained despite diuretic treatment. These patients also had a more severe restriction of sodium intake. Thus, comparison of absolute values of sodium excretion between the groups is not appropriate with our experimental design. The dose of diuretics was held constant in every patient until day 7 after TIPS. Altogether, our experimental conditions allow for novel results regarding the effects of TIPS on renal function in cirrhosis. We found that TIPS increases renal sodium excretion only in patients with ascites, not in patients without ascites. The increase of natriuresis was most marked in patients with refractory ascites. Natriuresis was observed at 1 week after TIPS. This is in contrast to the statement in one study 8 that natriuresis was delayed for as long as 1 month. These authors, however, in a recent publication, found a trend toward augmented natriuresis at 7 days after TIPS insertion. 13 A complex interaction of hemodynamics and neurohumoral systems governs ascites formation 18-25 and must be considered for the renal response to therapy. Regarding the mechanism of TIPS-induced natriuresis, we observed an increase of fractional sodium excretion in patients with refractory ascites. Following TIPS, plasma concentrations of renin and of aldosterone were significantly reduced in patients with ascites. These data are in accordance with a study by Quiroga et al., 6 who observed similar increases of fractional sodium excretion at 1 month after TIPS. In their investigation, plasma renin and aldosterone decreased in all patients and plasma norepinephrine decreased in those patients in whom it was initially elevated. This indicates that both proximal and distal tubular sodium reabsorption could be influenced by TIPS via neurohumoral effects. Wong et al. 8 also reported a decrease of plasma renin activity and plasma aldosterone concentration, but without an effect on plasma norepinephrine at 1 month after TIPS. Still, they observed a decrease of proximal tubular reabsorption of sodium. It has been claimed 13 that the decrease of the portal venous pressure gradient should be the pivotal mechanism contributing to improvement of renal function upon TIPS. However, in TABLE 4. Renal Response to TIPS Depends on Severity of Liver Disease and Fractional Sodium Excretion Before TIPS Number of Patients Urinary Sodium Excretion After TIPS (percent of values before TIPS) P Fractional Sodium Excretion After TIPS (percent of values before TIPS) P FIG. 2. Effects of TIPS on fractional sodium excretion (data of two patients not available; expressed as percentage of pre-tips fractional sodium excretion) correlate to Child-Pugh score (r.68; P.001). Child Pugh class A B 16 139 15 94 15 C 7 396 115.066 415 103.020 Fractional sodium excretion before TIPS* 0.35% 14 130 16 98 15 0.35% 7 396 114.059 407 107.027 *Data not available in two patients.

HEPATOLOGY Vol. 28, No. 3, 1998 GERBES ET AL. 687 contrast to our study, Wong et al. 13 did not compare the effects of TIPS in patients with refractory ascites to patients without refractory ascites. In our study, TIPS reduced the portal venous pressure gradients to a similar extent in all three groups of patients. We did not determine gradients at 7 days after insertion, the time of the renal function studies. Still, it is reasonable to assume that there was no major difference in the pressure gradients between the groups at day 7. Our data demonstrate marked differences of renal response in the different groups of patients despite a similar initial decrease of the hepatic-venous pressure gradient by TIPS, and thus suggest other mechanisms responsible for the improvement of renal function. This might be mediated by an increase of right atrial pressure and presumably of centrally effective blood volume upon TIPS, as reported recently. 26,27 Right atrial pressure, however, was not determined in our protocol. Such hemodynamic effects of TIPS would result in augmented renal perfusion and increased distal tubular sodium delivery, comparable, for example, with the effects of water immersion. 16,28 Under such conditions, diuretic treatment can be much more effective. Altogether, the marked increase of fractional sodium excretion observed in our patients with refractory ascites can be explained by hemodynamic and neurohumoral effects of TIPS. Interestingly, the change of renal sodium handling following TIPS was found to be related to the severity of liver disease indicated by the Child-Pugh score and to the renal function before TIPS. Patients with Child-Pugh class C had a greater increase of renal sodium excretion and of fractional renal sodium excretion than patients with class A and B. Furthermore, patients with a markedly reduced fractional sodium excretion had a greater benefit from TIPS insertion. If these findings could be extended to patients with even more impaired renal function and often most severe liver disease, i.e., with hepatorenal syndrome, one could expect a tremendous effect of TIPS. Indeed, recent observations 29,30 indicate that TIPS can improve or even revert hepatorenal syndrome. In conclusion, our data show that under practical clinical conditions, TIPS increases renal sodium excretion in patients with cirrhosis and ascites, but not in patients without ascites. The effect can be observed as early as 7 days after TIPS. Augmentation of sodium excretion and of fractional sodium excretion is most significant in patients with refractory ascites. The effect of TIPS on renal sodium handling correlates to the Child-Pugh score and inversely to the fractional sodium excretion before TIPS. Acknowledgment: The authors thank B. Aleith, U. Keller, and K. Simon for careful preparation of the manuscript. The authors further thank Dr. P. Luppa for determination of plasma renin, aldosterone, and norepinephrine concentrations. The laboratory assistance of I. Liβ is greatly appreciated. REFERENCES 1. Rössle M, Haag K, Ochs A, Sellinger M, Nöldge G, Perarnau JM, Berger E, et al. The transjugular intrahepatic portosystemic stent-shunt procedure for variceal bleeding. N Engl J Med 1994;330:165-171. 2. La Berge JM, Somberg KA, Lake JR, Gordon RL, Kerlan RK, Ascher NL, Roberts JP, et al. Two-year outcome following transjugular intrahepatic portosystemic shunt for variceal bleeding: results in 90 patients. Gastroenterology 1995;108:1142-1151. 3. Conn HO. Transjugular intrahepatic portal-systemic shunts: the state of the art. HEPATOLOGY 1993;17:148-158. 4. Sanyal AJ, Freedman AM, Luketic VA, Purdum PP, Shiffman ML, Tisnado J, Cole PE. Transjugular intrahepatic portosystemic shunt for patients with active variceal hemorrhage unresponsive to sclerotherapy. Gastroenterology 1996;111:138-146. 5. Ochs A, Rössle M, Haag K, Hauenstein KH, Deibert P, Siegerstetter V, Huonker M, et al. The transjugular intrahepatic portosystemic stent shunt procedure for refractory ascites. N Engl J Med 1995;332:1192-1197. 6. Quiroga J, Sangro B, Núñez M, Bilbao I, Longo J, García-Villarreal L, Zozaya JM, et al. Transjugular intrahepatic portal-systemic shunt in the treatment of refractory ascites: effect on clinical, renal, humoral and hemodynamic parameters. HEPATOLOGY 1995;21:986-994. 7. Somberg KA, Lake JR, Tomlanovich SJ, LaBerge JM, Feldstein V, Bass NM. Transjugular intrahepatic portosystemic shunts for refractory ascites: assessment of clinical and hormonal response and renal function. HEPATOLOGY 1995;21:709-716. 8. Wong F, Sniderman K, Liu P, Allidina J, Sherman M, Blendis L. Transjugular intrahepatic portosystemic stent shunt: effects on hemodynamics and sodium homeostasis in cirrhosis and refractory ascites. Ann Intern Med 1995;122:816-822. 9. Ochs A, Gerbes AL, Haag K, Holl J, Hauenstein KH, Waggershauser T, Reiser M, et al. TIPS and paracentesis for the treatment of refractory ascites (RA): interims analysis of a randomized controlled trial [Abstract]. HEPATOLOGY 1995;22:297A. 10. Lebrec D, Giuily N, Hadengue A, Vilgrain V, Moreau R, Poynard T, Gadano A, et al. Transjugular intrahepatic portosystemic shunts: comparison with paracentesis in patients with cirrhosis and refractory ascites: a randomized trial. J Hepatol 1996;25:135-144. 11. Siuta M, Sanyal AJ, Schubert ML. TIPS for the treatment of refractory ascites. Gastroenterology 1996;110:956-958. 12. Martinet JP, Fenyves D, Legault L, Roy L, Dufresne MP, Spahr L, Lafortune M, et al. Treatment of refractory ascites using transjugular intrahepatic portosystemic shunt (TIPS). A caution. Dig Dis Sci 1997;42: 161-166. 13. Wong F, Sniderman K, Liu P, Blendis L. The mechanism of the initial natriuresis after transjugular intrahepatic portosystemic shunt. Gastroenterology 1997;122:899-907. 14. Arroyo V, Gines P, Gerbes AL, Dudley F, Gentilini P, Laffi G, Reynolds TB, et al. Definition and diagnostic criteria of refractory ascites and hepatorenal syndrome in cirrhosis. HEPATOLOGY 1996;23:164-176. 15. Pugh RNH, Murray-Lyon IM, Dawson JL, Pietroni MC, Williams R. Transection of the oesophagus for bleeding oesophageal varices. Br J Surg 1973;60:646-649. 16. Gerbes AL, Wernze H, Arendt RM, Riedel A, Sauerbruch T, Paumgartner G. Atrial natriuretic factor (ANF) and renin-aldosterone in volume regulation of patients with cirrhosis. HEPATOLOGY 1989;9:417-422. 17. Gülberg V, Gerbes AL. Relation of endothelins to volume regulating neurohumoral systems in patients with cirrhosis of the liver. Eur J Clin Invest 1995;25:893-898. 18. Bernardi M, Trevisani F, Gasbarrini A, Gasbarrini G. Hepatorenal disorders: role of the renin-angiotensin-aldosterone system. Semin Liver Dis 1994;14:23-34. 19. Bernardi M, Gasbarrini A, Trevisani F, Caraceni P, De Collibus C, Colantoni A, Andreone P, et al. Hemodynamic and renal effects of ascites apheresis, concentration and reinfusion in advanced cirrhosis. J Hepatol 1995; 22:10-16. 20. Bernardi M, Fornalè L, Di Marco C, Trevisani F, Baraldini M, Gasbarrini A, De Collibus C, et al. Hyperdynamic circulation of advanced cirrhosis: a re-appraisal based on posture-induced changes in hemodynamics. J Hepatol 1995;22:309-318 21. Bosch J, Pizcueta MP, Fernandez M, Feu F, Cirera I, Luca A, Garcia-Pagan JC. Hepatic, splanchnic and systemic haemodynamic abnormalities in portal hypertension. Baillieres Clin Gastroenterol 1992;6:425-436. 22. Dudley FJ. Pathophysiology of ascites formation. Gastroenterol Clin North Am 1992;21:215-235. 23. Gerbes AL. Pathophysiology of ascites formation. Hepatogastroenterology 1991;38:360-364. 24. Henriksen JH, Ring-Larsen H. Hepatorenal disorders: role of the sympathetic nervous system. Semin Liver Dis 1994;14:35-43. 25. Henriksen JH. Cirrhosis: ascites and hepatorenal syndrome. Recent advances in pathogenesis. J Hepatol 1995;23:25-30. 26. Azoulay D, Castaing D, Dennison A, Martino W, Eyraud D, Bismuth H. Transjugular intrahepatic portosystemic shunt worsens the hyperdynamic circulatory state of the cirrhotic patient: preliminary report of a prospective study. HEPATOLOGY 1994;19:129-132. 27. Van der Linden P, Le Moine O, Ghysels M, Ortinez M, Devière J. Pulmonary hypertension after transjugular intrahepatic portosystemic

688 GERBES ET AL. HEPATOLOGY September 1998 shunt: effects on right ventricular function. HEPATOLOGY 1996;23:982-987. 28. Epstein M. Renal effects of head-out water immersion in humans: a 15-year update. Physiol Rev 1992;72:563-621. 29. Ochs A, Rössle M, Haag K, Gerbes AL, Morgenroth A, Deibert P, Hauenstein KH, et al. TIPS for hepatorenal syndrome [Abstract]. HEPATOLOGY 1994;20:184A. 30. Brensing KA, Textor J, Strunk H, Klehr HU, Schild H, Sauerbruch T. Transjugular intrahepatic portosystemic stent-shunt for hepatorenal syndrome. Lancet 1997;349:697-698.