General
Chronic liver diseases that lead to liver cirrhosis may be associated with the development of portal hypertension. This is an increase in pressure within the portal venous system, the vein that carries blood from the intestines, spleen and stomach to the liver. Other causes of portal hypertension are specific liver conditions characterized by changes in the liver architecture, for example Nodular Regenerative Hyperplasia, and thrombosis of the portal vein (vena portae trombose) or of the veins leaving the liver (Budd-Chiari syndrome). When portal hypertension occurs, blood is detoured into other smaller veins that ultimately allow blood to flow back to the heart, so partially avoiding the liver. However, these smaller veins can enlarge in the oesophagus, stomach or at other places - and form varicose veins or varices, which may give rise to bleeding. Bleeding varices are most commonly observed in the oesophagus and in the stomach, but may also occur in other sites such as the rectum or duodenum.

Another frequent complication of portal hypertension is the development of ascites, i.e. accumulation of fluid in the abdominal cavity. Ascites may cause abdominal swelling; other problems are the development of an infection of this fluid and the development of hernias, such as umbilical, inguinal and scar-related hernias.

Current study
The TIPS-TRUE Trial. Erasmus MC is one of the Dutch hospitals participating in the TIPS-TRUE Trial. This study aims to determine the optimal treatment strategy for patients with a first or second occurrence of variceal bleeding in the oesophagus or stomach.

Introduction
Variceal bleeding in the oesophagus or stomach is a serious complication in patients with liver cirrhosis and portal hypertension. Around 20% of patients do not survive a first bleeding, 17% will experience a recurrent bleeding within six weeks, and 70% within two years. Dependent on the severity of the underlying liver disease and of the bleeding, the mortality associated with each bleeding episode ranges from 20-40% (1). Apart from treatment of the acute bleeding episode and of the underlying disease, treatment of patients with variceal bleeding is aimed at preventing recurrences (secondary prophylaxis).

Current therapeutic approach
Several options are available for the prevention of recurrent variceal bleedings: endoscopic therapy, treatment with the non-selective ß-blocker propranolol, and TIPS (Transjugular Intrahepatic Portosystemic Shunt). There is consensus that endoscopic therapy, notably rubber band ligation, and treatment with propranolol are equivalent alternatives (2). From the results of a large number of trials and meta-analyses it is generally assumed that TIPS, in comparison with endoscopic therapy, is significantly more effective in preventing recurrent bleedings, but does not improve survival and increases the risk of hepatic encephalopathy (3, 4). This is why TIPS so far has been applied as a valuable second-line treatment when other therapies fail; but rather not as treatment at an earlier stage (2).

The studies that resulted in the above conclusions concerned TIPS procedures using uncoated metal stents. A major problem in these procedures was loss of shunt function by thrombosis and neo-intima proliferation, leading to recurrent bleeding, need of intensive shunt monitoring, and frequent shunt revisions (5). Recent studies have shown that implantation of polytetrafluoroethylene-coated stents strongly reduces the risk of TIPS obstruction (6-8). In a controlled study the one-year integrity of the shunt with conventional stents was 41% and with coated stents 88% (9). A retrospective study suggested improved survival with the use of coated stents (10). Surprisingly, in these studies the improved integrity of the shunt on the long term appeared not to coincide with a greater likelihood of encephalopathy (7-9). Perhaps this can partly be ascribed to the fact that often stents were used with smaller diameter than the conventional 10 mm stents. There may be other possible explanations, however, such as improved effectiveness with regard to ascites, and its associated lower need of diuretic therapy, as well as the less frequent occurrences of bleeding.

The TIPS-TRUE trial
In summary, recent studies suggest hat TIPS procedures involving implantation of coated stents are more effective in the treatment of complications of portal hypertension, and perhaps could lead to reduced mortality in patients suffering from variceal bleeding, in comparison with the current standard treatments. This consideration lies at the basis of a randomised multicenter trial in the Netherlands: de TIPS-TRUE trial (Transjugular Intrahepatic Porto-systemic Shunt with Gore-tex® covered stent-graft versus endoscopic treatment for secondary prevention of gastro-oesophageal bleeding). In this multicenter study TIPS with coated stents are compared with endoscopic therapy in combination with propranolol, regarding effectiveness in the secondary prevention of variceal bleeding. The first-line treatment of variceal bleeding in the stomach is often endoscopic therapy as well (injection of tissue glue); should this fail, TIPS is currently used as second-line treatment. This is why this trial will also include patients with gastric variceal bleeding.

Aim of the trial
The primary aim of the trial is to evaluate to which extent TIPS treatment using implantation of coated stents is better in preventing recurrent bleeding, as compared with endoscopic therapy. A secondary aim is determining differences in the incidences of therapy failure (combined endpoint: necessity of starting alternative therapy or death), hepatic encephalopathy, and death. Another secondary aim is comparing costs and quality of life. Implicit in the study is the question whether TIPS should not only be regarded a second-line treatment option but could also be considered a primary therapy for patients with variceal bleeding.

Design
The trial went ahead on January 1, 2008. The expected inclusion period is one year with minimum follow-up one year. A power calculation showed that 124 patients are needed. A central randomisation procedure is in place.

Inclusion criteria

• First or second variceal bleeding in oesophagus or stomach;
• Initial stabilisation (no indications of continued bleeding) achieved through combined endoscopic, vasoactive and endoscopic therapy;
• Informed consent;
• Age >18 and <76 years.

Exclusion criteria

• Previous spontaneous (not bleeding-related) hepatic encephalopathy;
• Severe heart failure (NYHA class III or IV);
• Portal hypertension not related to liver disease (vena portae or vena lienalis trombose);
• Previous TIPS implantation;
• Extensive hepatocellular carcinoma;
• Severe liver insufficiency (Child-Pugh class > 13 points);
• Sepsis and/or multi-organ failure.

Centres
The trial is organized by the four university medical centres in the Netherlands which perform TIPS procedures: AMC Amsterdam, LUMC Leiden, UMC St Radboud Nijmegen, and Erasmus MC Rotterdam. In addition, there are a number of participating centres in which the protocol has been assessed as well and the study has been approved. Patients in these centres can be enrolled, treated and followed when they have been randomized to the endoscopic therapy group. Patients randomized to the TIPS procedure group are referred to one of the four TIPS-centres. All other centres may refer patients with a first or second variceal bleeding to one of the (regional) TIPS centres.

No doubt, nationwide support and willingness to participate from the side of many centres are essential to be able to include the required number of patients. We would like to call upon all professionals involved in the care for patients with variceal bleeding and invite them to contribute to this important trial.

This project was financially supported by ZonMw (the Netherlands Organisation for Health Research and Development).