Summary of findings for the main comparison. Flumazenil versus placebo for people with cirrhosis and hepatic encephalopathy.

Flumazenil versus placebo for people with cirrhosis and hepatic encephalopathy
Patient or population: people with hepatic encephalopathy Setting: hospital Intervention: flumazenil Comparison: placebo
Outcomes	Anticipated absolute effects* (95% CI)		Relative effect (95% CI)	No of participants (studies)	Quality of the evidence (GRADE)	Comments
	Risk with placebo	Risk with flumazenil
All‐cause mortality follow‐up: range 1 day to 2 weeks	Study population		RR 0.75 (0.48 to 1.16)	842 (11 RCTs)	⊕⊕⊝⊝ Low 1,2	The only RCT with low risk of bias found no effect of flumazenil on all‐cause mortality (RR 0.76, 95% CI 0.37 to 1.53). The Trial Sequential Analysis found insufficient evidence to support or refute an intervention benefit/harm.
	93 per 1000	70 per 1000 (45 to 108)
Hepatic encephalopathy	Study population		RR 0.75 (0.71 to 0.80)	824 (9 RCTs)	⊕⊕⊝⊝ Low 1,2	The only RCT with a low risk of bias reported a beneficial effect of flumazenil on hepatic encephalopathy (RR 0.78, 95% CI 0.72 to 0.84; Barbaro 1998). The Trial Sequential Analysis found that flumazenil was associated with a beneficial effect on hepatic encephalopathy (Figure 1). The methods used to assess this outcome varied considerably (Table 2) and the duration of follow‐up was very short in the majority of RCTs.
	933 per 1000	700 per 1000 (662 to 746)
Serious adverse events	See comment	See comment	Not estimable	842 (11 RCTs)	⊕⊕⊝⊝ Low 1,2	All‐cause mortality was the only serious adverse event reported for both the intervention and control group (Table 7). The narrative text in 4 RCTs described that causes of death included liver failure, progressive liver disease, and infections.
*The risk in the intervention group (and its 95% confidence interval) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI). CI: confidence interval; RCT: randomised clinical trial; RR: risk ratio.
GRADE Working Group grades of evidence High quality: We are very confident that the true effect lies close to that of the estimate of the effect. Moderate quality: We are moderately confident in the effect estimate: The true effect is likely to be close to the estimate of the effect, but there is a possibility that it is substantially different. Low quality: Our confidence in the effect estimate is limited: The true effect may be substantially different from the estimate of the effect. Very low quality: We have very little confidence in the effect estimate: The true effect is likely to be substantially different from the estimate of effect.

¹ Downgraded due to risk of bias: only one RCT had a low risk of bias.

² Downgraded due to imprecision: wide confidence intervals.