Dursun 2003.

Methods	Double‐blind, single‐centre, placebo‐controlled, parallel‐arm RCT.
Participants	40 participants with cirrhosis and hepatic encephalopathy classified as subclinical (corresponding to minimal; 10 participants) or overt Grade I to III (30 participants; Table 3). Type of overt hepatic encephalopathy (acute or chronic) not specified. Mean ± SD age: flumazenil: 44.5 ± 12.9 years; placebo: 43.7 ± 11.9 years. Proportion of men: 73%. Aetiology of cirrhosis: alcohol 0%; hepatitis B/C 100%. Proportion testing positive for benzodiazepines at baseline (Table 6): investigators did not screen for benzodiazepines.
Interventions	Intervention comparison: intravenous infusion flumazenil 1 mg/hour for 5 hours versus placebo (saline) administered similarly. Total dose of flumazenil: 5 mg. Cointerventions: lactulose 30 mL 6‐hourly
Outcomes	Outcomes included in meta‐analyses: mortality, hepatic encephalopathy (Table 2), serious adverse events (Table 7), and Number Connection Test assessed after a maximum of 5 hours.
Neuropsychiatric assessment	Baseline and post infusion: Clinical assessment of mental status (Table 3) assessed at baseline and every 30 minutes after infusion for a maximum of 5 hours; Glasgow Coma Score (Table 3) assessed at baseline; Electroencephalography (Table 4) assessed at baseline and 1 hour after infusion; Number Connection Test assessed at baseline and every 30 minutes after infusion for a maximum of 5 hours; Blood ammonia concentrations assessed at baseline.
Inclusion period (date)	December 1999 to January 2002.
Country	Turkey.
Notes	Included data: the trial report included information about participants with minimal and overt hepatic encephalopathy. We have analysed these 2 groups separately.
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Unclear risk	Not reported.
Allocation concealment (selection bias)	Low risk	Concealed drug containers.
Blinding of participants and personnel (performance bias) All outcomes	Low risk	Blinding of participants and personnel.
Blinding of outcome assessment (detection bias) All outcomes	Low risk	Blinding of outcome assessment.
Incomplete outcome data (attrition bias) All outcomes	Low risk	No missing outcome data and all participants are included in analyses.
Selective reporting (reporting bias)	Low risk	Trial describes clinically relevant outcomes. We had no access to information about outcomes described in the original protocol or information in trial registries.
For‐profit funding	Unclear risk	No information provided.
Other bias	Low risk	No other biases.
Overall assessment	High risk	High risk of bias.