Brunati 1996.

Methods	Randomised: yes, by random number table (PC). Concealment: method not described (PC). Double‐blind: no (PC). Description of withdrawals/dropouts: no. Length of FU: 5 days. Jadad score: 1/5.
Participants	Country: Italy. Diagnosis of bleeding: by endoscopy. Source of bleeding: all confirmed or unconfirmed varices. Endoscopy: performed in all patients prior to randomisation. Diagnosis of cirrhosis: histological or clinical. 85 patients. Child's A = 21%, B = 42%, C = 37%. Alcoholic cirrhosis = 25%
Interventions	Terlipressin versus ocreotide versus no treatment. R x 1: terlipressin 2 mg iv every six hours for two days. R x 2: octreotide 0.1 mg iv every eight hours for two days. Time from admission to treatment: unknown. Initial sclerotherapy: all patients.
Outcomes	1. Mortality at five days. 2. Failure of initial haemostasis at five days. 3. Rebleeding: NR. 4. Procedures required for haemostasis: NR. 5. Blood transfusions: (median and range). 6. LOS: NR.
Notes	Source of funding: None Published in abstract only.
Risk of bias
Bias	Authors' judgement	Support for judgement
Allocation concealment?	Unclear risk	B ‐ Unclear