STOP‐HE 2017

Methods	Ornithine phenylacetate (OCR‐002): multicentre, double‐blind, placebo‐controlled, randomised clinical trial
Participants	Included participants: cirrhosis, with an acute episode of overt hepatic encephalopathy ≥ Grade 2 and hyperammonaemia, whose mental status did not improve after a minimum of 12 h of 'standard care' (n = 231) Age: not reported Proportion of men: USA 59%; elsewhere 72% Aetiology of liver disease: not reported MELD score (mean): USA 20; elsewhere 18
Interventions	Intervention comparison: ornithine phenylacetate vs placebo Ornithine phenylacetate: 10 g/day, 15 g/day or 20 g/day, by continuous IV infusion (n = 116); dose pre‐determined by the level of hepatic decompensation Placebo: identically presented 5% dextrose by continuous IV infusion (n = 115) Duration of treatment: 5 days Co‐intervention: all participants received 'standard care'" based on 'gut flora modification'" but this was not specified although 135 (58.4%) participants were taking rifaximin (mean dose: 1057 mg/day; range: 400‐1650/day); 12 (5.2%) were taking L‐ornithine L‐aspartate (dose range 9‐20g/day) and an unspecified proportion were taking lactulose.
Outcomes	Neurocognitive assessment: clinical improvement in hepatic encephalopathy symptoms assessed with HESA; no details provided venous blood ammonia median time to improvement time to normal plasma ammonia levels reduction in plasma ammonia levels correlation of ammonia levels with clinical improvement
Inclusion period	November 2013‐December 2016
Country of origin	Australia, Austria, Belgium, Bulgaria, Czech Republic, Denmark, Estonia, France, Germany, Hungary, Israel, Italy, Netherlands, New Zealand, Russian Federation, Spain, USA
Outcomes included in meta‐analyses	Mortality Serious adverse events Non‐serious adverse events Venous blood ammonia
Notes	Publication status: 3 separate abstracts at the same conference No extractable data are available in the abstracts for the primary end‐points. Differences, some of them significant, were observed between participants enrolled in the USA (n = 130) and elsewhere (n = 101) in baseline characteristics, the severity of the underlying liver disease, the precipitant factors and use of additional anti‐encephalopathy treatments Funding: Trial sponsored by Ocera Therapeutics Inc
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	Computer‐generated randomisation schedule
Allocation concealment (selection bias)	Low risk	Central allocation
Blinding of participants and personnel (performance bias) All outcomes	Low risk	Participants, care providers, investigators and outcomes assessors were blinded.
Blinding of outcome assessment (detection bias) Non‐mortality outcomes	Low risk	Participants, care providers, investigators and outcomes assessors were blinded.
Incomplete outcome data (attrition bias) All outcomes	Low risk	No missing outcome data. All participants were included in the analyses
Selective reporting (reporting bias)	High risk	The primary end‐point in the registered trial protocol is 'change from baseline in hepatic encephalopathy stage'. However, in the published abstracts the primary endpoint reported is 'the time to meaningful clinical improvement in hepatic encephalopathy symptoms assessed using the Hepatic Encephalopathy Staging Tool' together with a number of other time‐related variables.
Other bias	High risk	Differences were reported for a number of variables in participants recruited in the USA and those recruited elsewhere in the world, which is an additional source of bias.
Overall assessment Non‐mortality outcomes	High risk
Overall assessment Mortality outcomes	High risk