SUMMARYThe effect of prophylactic treatment of oesophageal varices by endoscopic injection sclerotherapy before the first episode of variceal haemorrhage was studied in patients with cirrhosis in a prospective, randomised and controlled multicentre trial. From February 1984 to March 1987 patients with liver cirrhosis and large varices (stage III-IV according to Paquet) were treated and followed up. The sample comprised 87 patients: 45 in the prophylactic treatment and 42 in the control group. After excluding drop outs, 41 patients were treated in each group. Twenty nine per cent of patients in the sclerotherapy group and 34% in the control group had a variceal haemorrhage during the period of observation. There was no significant difference in the distributions of the bleeding free intervals between the sclerotherapy and the control groups. During the follow up period 24% of patients in the sclerotherapy group and 46% in the control group died. The distribution of survival times indicates a tendency towards longer survival of patients with prophylactic sclerotherapy, particularly in those with alcoholic cirrhosis.Severe haemorrhages from oesophageal varices belong to the most serious complications in patients with liver cirrhosis and portal hypertension. About 30 to 70% of patients with liver cirrhosis develop oesophageal varices' and some 20-40% of these will eventually bleed.2 The mortality of the first haemorrhage has been reported to be between 30 and 80% .3 Although many studies from all over the world show that sclerotherapy during the acute bleeding phase will efficiently stop variceal bleeding-9 and repeated sclerotherapy after terminating the initial bleeding episode can increase the longterm survival of such patients,"' the first massive variceal haemorrhage is nevertheless fatal for many patients. It therefore