Trial for treatment of refractory ascites TIPS the scales

Abstract

Salerno F, Merli M, Riggio O, Cazzaniga M, Valeriano V, Pozzi M, Nicolini A, Salvatori F, Gruppo Italiano per lo Studio TIPS (Departments of Internal Medicine and Radiology, University and Ospedale Maggiore Policlinico Instituto di Ricera e Cura a carattere Scientifico, Milan; Departments of Gastroenterology and Radiology, University “La Sapienza” and Policlinico “Umberto 1,” Rome; Department of Internal Medicine, 2nd University of Milan, Ospedale San Gerardo, Monza, Italy). Randomized controlled study of TIPS versus paracentesis plus albumin in cirrhosis with severe ascites. Hepatology 2004;40:629–635. The authors performed a multi-center, randomized, controlled, clinical trial (RCT) in 66 cirrhotic patients with refractory or recidivant ascites, comparing transjugular intrahepatic portosystemic shunt (TIPS) and repeated large volume paracentesis plus intravenous albumin (LVP+A). Thirteen patients treated by TIPS died, compared to 20 patients treated by LVP+A. The probability of survival without transplantation in the TIPS group was 77% at 1 year and 59% at 2 years, compared to 52%, and 29% respectively in the LVP+A group (P = .21). In a multivariate analysis, death was independently predicted by treatment with LVP+A and MELD score. The number of rehospitalizations was similar in each group but episodes of hepatic encephalopathy were more frequent in the TIPS group. The development of refractory ascites in a cirrhotic patient, as defined as resistance to maximal doses of diuretics using a combination of spironolactone and furosemide, is associated with a deterioration in survival rate to 50% in 1 to 2 years (J Hepatol 1999;31:1088–1097). Unless reversed, refractory ascites should be an indication for liver transplantation. However, in the present RCT only 4 patients from each group of 33 cirrhotic patients with refractory ascites were transplanted, a remarkably small number. Up until the mid 1980s, ascites were essentially treated by increasing doses of diuretics. When the ascites became resistant to diuretics, the option was to try some form of shunt procedure such as the peritoneovenous shunt (Gastroenterology 1979;71:250–257). However, although effective, these shunts gave rise to serious life-threatening complications (Am J Surg 1980;139:125–131). In 1987, Gines and colleagues showed in a RCT that in cirrhotic patients with tense ascites, LVP+A, resulted in fewer side effects and reduced hospitalization, compared to diuretics (Gastroenterology 1987;93:234–241). LPV+A then became the standard management for patients with refractory ascites. At the same time, the new radiological shunt technique known as TIPS (Am J Surg 1971;121:588–592, Can Med Assoc J 1982;126:267–268, AJR Am J Roentgenol 1983;140:709–714) was shown to significantly, but transiently, reduce portal pressure in cirrhotic patients and thus to be an effective method of treating bleeding varices. With the use of a permanent metallic stent between the hepatic and portal veins (JAMA 1991;266:390–393), TIPS was shown to be a successful long-term treatment for bleeding varices (Ann Intern Med 1992;116:304–309, Surgery 1993;114:719–726). Then a large study of cirrhotic patients with bleeding varices found that the TIPS procedure was also beneficial in reducing ascites in those patients with associated ascites (N Engl J Med 1994;330:165–171). The pathophysiology of this effect seemed to be a reduction in sinusoidal portal pressure resulting in a fall in the plasma renin activity and serum aldosterone levels, and a rise in renal blood flow and glomerular filtration rate, associated with a natriuresis and diuresis (Ann Intern Med 1995;122:816–822). This effect occurred despite an exacerbation of the hyperdynamic circulation, with a further fall in systemic vascular resistance and a further increase in cardiac output (Gastroenterology 1997;112:899–907, Hepatology 1999;29:632–639). Although systemic vasodilatation, with vascular underfilling, had been proposed as the underlying mechanism for sodium retention in cirrhosis (Hepatology 1988;8:1151–1157), the increased vasodilatation after TIPS is not associated with further vascular underfilling because the shunting causes an increase in total central and cardiac blood volumes (Gastroenterology 1997;112:899–907). These pathophysiological changes continue to improve long term so that these patients can cautiously increase their salt intake (Am J Med 1999;106:315–322), resulting in increased food intake, an improvement in nutritional status (Am J Gastroenterol 2001;96:2442–2447, J Hepatol 2004;40:228–233), and their quality of life (Digestion 2002;66:127–130). However, it is not all good news, because chronic hemolysis is universal (Hepatology 1996;23:32–39), the incidence of hepatic encephalopathy, as with any shunt procedure, increases (Hepatology 1994;20:46–55, Am J Gastroenterol 1995;90:549–555), and the shunts tend to stenose because of mucosal hyperplasia (Hepatology 1998;28:22–32) and form vegetative infections (Gastroenterology 1998;115:110–115). In 1996, the first RCT comparing TIPS with LVP+A for cirrhotic patients with refractory ascites was published (J Hepatol 1996;25:135–144). The total number of patients included in the study was small (n = 25), and although TIPS was superior in terms of clearing the ascites, the mortality rate in the most severely decompensated patients with TIPS was increased. However, the percentage of TIPS patients in whom the stent malfunctioned was high. In contrast, the next published RCT comparing TIPS and LVP+A from Germany included 66 patients. Again, the TIPS group showed a benefit in terms of reducing the recurrence of ascites. Furthermore, the probability for survival without liver transplantation was 69% at 1 year and 58% at 2 years in the TIPS group, compared with 52% and 32%, respectively, in the LVP+A group (N Engl J Med 2000;342:1701–1707). These results are almost identical with the present study. In a multivariate analysis in the German study, treatment with TIPS was independently associated with survival, without the need for transplantation (P = .02). In addition, age <60 years, female sex, serum bilirubin <3 mg/dL, and serum sodium >125 mmol/L were all significantly related to increased survival (N Engl J Med 2000;342:1701–1707). There was no difference in the incidence of hepatic encephalopathy between the 2 groups. This study was then followed by 2 further RCT studies in which there were no significant differences in mortality between the 2 groups (Gastroenterology 2002;123:1839–1847, Gastroenterology 2003;124:634–641). In these latter 2 studies, the survival, free of transplantation for both TIPS and LVP+A, was lower: 26% vs 30% at 18 months and 35% vs 33% at 500 days, respectively. The percentage of patients treated by liver transplantation, 21% and 31%, respectively, was much greater than in the 2 studies showing a significant benefit for TIPS, in which 5% and 12% of patients required transplantation, respectively. Comparing other parameters that might have impacted on the results, the mean ages, and the percent of patients over the age of 60 years, the percent of female patients, the percent of patients with alcoholic liver disease, the mean Child-Pugh scores, the percent of Child-Pugh C patients, and the serum bilirubin levels were all similar. Even in a study that showed no overall benefit of TIPS for survival, TIPS did result in a reduction in the number of patients with hyponatremia, and a significant reduction in the number with renal failure and the hepatorenal syndrome compared to the LVP+A group (Gastroenterology 2002;123:1839–1847). The increase in hepatic encephalopathy with TIPS was variable. Surprisingly, there was no advantage with TIPS as far as quality of life was concerned when measured (Gastroenterology 2003;124:634–641), and TIPS was more expensive than LVP+A (Gastroenterology 2002;123:1839–1847). In conclusion, I agree with Dr Salerno and his colleagues, in confirming the results of the German study, that the benefit of TIPS over LVP+A in managing refractory ascites seems to be translated into an improved survival rate. This justified the high salvage rate using TIPS for patients with failure of LVP+A. Might the much higher transplantation rates in the 2 studies showing no benefit prevented those studies from showing a survival advantage for TIPS? ReplyGastroenterologyVol. 128Issue 3PreviewDr Blendis discusses possible explanations for the different survival rates following TIPS reported by 5 RCTs. His suggestion that a different rate of liver transplantation could have affected a survival advantage for TIPS is appealing but difficult to prove. A meta-analysis performed with individual data is the right tool to do this, but the total number of patients included in the 5 RCTs is quite low (n = 330) and this makes such an analysis likely to be unsuccessful. Full-Text PDF