To undertake European multicentre randomised trials in the treatment of patients with variceal haemorrhage, acquiring adequate sample sizes for meaningful results.
To develop European collaboration through these clinical trials, interchanging best practice between units, and working towards gold standards that are applicable across Europe in this field.
To determine the relative place of treatments that have uncertain benefits in the management of patients that suffer variceal haemorrhage, we will take advantage of the extensive number of units specialising in this area, and the widespread enthusiasm to collaborate in randomized clinical trials. The project will focus on the role of drugs, both acutely and long-term, and the intervention of portosystemic shunting, either by surgical or radiological means; these all being treatments that complement endoscopic therapy. Three trials that relate to different stages in the management of patients with variceal haemorrhage will clarify the role of these therapies, with each participating centre conducting any number of the three simultaneously. Trial 1 will test the relative efficacy of a single bolus of 250 ug somatostatin followed by 250 ug/h intravenous infusion for 5 days and/or a single bolus of 2 mg terlipressin followed by 1 mg 4 hourly intravenously with 10 mg/24 h transdermal patches of nitro-glycerine 12 hourly for 5 days versus placebo as adjuvant to endoscopic treatment for the control of acute variceal haemorrhage; a double-blind protocol will be used Trial 2 will compare low dose 6 weekly intramuscular depot octreotide for one year and/or beta blockade for one year versus no treatment as adjuvant to programmed endoscopic treatment to prevent recurrent variceal haemorrhage; an open protocol will be used. Trial 3 will compare transjugular intrahepatic portosystemic shunting versus narrow diameter Hgraft portosystemic shunting for recurrent variceal haemorrhage for which endoscopic and pharmacological treatment has failed; an open protocol will be used. Because expected treatment differences are modest, large sample sizes are required; to develop the collaboration effectively, and ensure its success, Trial l will be started first, with the choice of timing to start Trials 2 and 3 dependent on progress.