Pauwels 1994.

Methods	Randomised: yes, but method not described. Concealment: method not described. Double‐blind: no Description of withdrawals/drop‐outs: yes, 11/60 patients were excluded from the analysis because the source of bleeding was not variceal. Length of FU: 30 days. Jadad score: 2/5.
Participants	Country: France Diagnosis of bleeding: clinical. Source of bleeding: unknown. Endoscopy performed in all patients after randomisation. Diagnosis of cirrhosis: unknown. 57 patients with 60 episodes were randomised. 46 patients with 49 episodes were analysed. Child's class: unknown. Alcoholic cirrhosis = unknown.
Interventions	Terlipressin versus somatostatin versus placebo. Rx1: terlipressin 2 mg iv every 6 h until bleeding stopped then 1 mg iv every 6 h for 24 h. Rx2: somatostatin 250 microg iv bolus then 250 micorg/h infusion until 2 h after bleeding arrest. Time from admission to treatment: unknown. Initial sclerotherapy: None.
Outcomes	1. Mortality at 30 days. 2. Failure of initial haemostasis at 48 h. 3. Rebleeding at 48 h. 4. Procedures required for haemostasis. 5. Blood transfusions. 6. LOS: NR.
Notes	Source of funding: unknown. Patients were randomised before endoscopy. However, 11 patients found not to have varices at endoscopy were excluded.
Risk of bias
Bias	Authors' judgement	Support for judgement
Allocation concealment?	Unclear risk	B ‐ Unclear