Vasoactive drugs for adults with cirrhosis and large ascites treated with paracentesis

Objectives

This is a protocol for a Cochrane Review (intervention). The objectives are as follows:

To assess the benefits and harms of vasoactive drugs versus placebo or no intervention, or versus another vasoactive drug, or versus a plasma expander during or after paracentesis in people with cirrhosis and large ascites.

Background

Description of the condition

Liver cirrhosis is within the top 20 causes of disability‐adjusted life‐years and years of life lost (Pimpin 2018; Asrani 2019; GBD 2017 Cirrhosis Collaborators 2020). Globally each year, liver cirrhosis causes 1.16 million deaths, while numbers of deaths and the proportion of all global deaths due to cirrhosis have increased. Deaths due to cirrhosis constituted 2.4% (2.3% to 2.6%) of total deaths globally in 2017 compared with 1.9% (1.8% to 2.0%) in 1990 (GBD 2017 Cirrhosis Collaborators 2020), with exceptions in some geographic areas where the burden of cirrhosis is lower (for example, in richer countries; Pimpin 2018; GBD 2017 Cirrhosis Collaborators 2020). The main risk factors for the development of liver cirrhosis are viral hepatitis B and C, and alcoholic and non‐alcoholic fatty liver diseases (Pimpin 2018; Asrani 2019; GBD 2017 Cirrhosis Collaborators 2020). Although hepatitis B caused the greatest proportion of cirrhosis deaths in 2017, non‐alcoholic steatohepatitis might become the leading cause of cirrhosis in the future (GBD 2017 Cirrhosis Collaborators 2020).

Cirrhosis is characterised by a long‐lasting natural history with an usually prolonged asymptomatic, compensated phase, followed by a phase of decompensation with appearance of ascites, variceal bleeding, jaundice, and encephalopathy (D'Amico 2018; EASL 2018). The transition from compensated cirrhosis into decompensated cirrhosis is due to the worsening of portal pressure and liver function with a vasodilatory–hyperdynamic circulatory state, leading to a progressive decrease in effective blood volume, cardiac function, and renal perfusion. Inflammation from overt or covert (bacterial translocation) infections is considered a major driver of progression (Bernardi 2015; Adebayo 2019). Ascites is the most common form of decompensation in cirrhosis as 5% to 10% of patients with compensated cirrhosis develop it each year (Ginés 1987a; EASL 2018). The appearance of ascites heralds a poor prognosis, as the five‐year survival drops from about 80% in compensated patients with cirrhosis to about 30% in patients with decompensated cirrhosis and ascites (D'Amico 2006). Ascites can be graded as grade 1 (mild), grade 2 (moderate), and grade 3 (large or gross, if it provokes marked abdominal distension; Arroyo 1996; Moore 2003; EASL 2018). About 10% of patients with cirrhosis and ascites develop refractory ascites, defined as ascites that cannot be mobilised, or the early recurrence of which cannot be satisfactorily prevented by medical therapy (Arroyo 1996; Moore 2003). Refractory ascites is associated with a median survival of only about six months because of the development of complications (Salerno 1993), e.g. hyponatraemia, spontaneous bacterial peritonitis, and progressive renal dysfunction.

No data on the evolution of grade 1 ascites are available, nor is it known whether treatment modifies its natural history (EASL 2018). A moderate restriction of sodium intake (to 80 mmoL/day to 120 mmoL/day) and diuretics are recommended in people with moderate ascites (EASL 2018).

Large‐volume (> 5 litres) paracentesis is the first‐line therapy for the treatment of large ascites, usually removed completely in a single session (total paracentesis; EASL 2018). Randomised clinical trials have shown that paracentesis is more effective than diuretics, and paracentesis is associated with a lower rate of complications and reduced length of hospitalisation (Quintero 1985; Ginés 1987b). There is a general agreement that large‐volume paracentesis can also be used to treat refractory ascites. Moreover, patients with refractory ascites should be evaluated for liver transplantation because of the poor prognosis of this condition (EASL 2018).

Paracentesis with the removal of large amounts of ascitic fluid may impair circulatory function, characterised by marked activation of vasoactive systems ‐ namely, increased plasma renin activity and plasma aldosterone concentration (Ginès 1988; Luca 1995). This circulatory derangement is related to an accentuation of arteriolar vasodilatation, already present in people with advanced cirrhosis. The origin of arteriolar vasodilatation occurring after paracentesis is probably multifactorial and includes the dynamics of paracentesis, increased synthesis of nitric oxide from the vascular endothelium, and mechanical changes secondary to mechanical decompression (Pozzi 1994; Cabrera 2001; Coll 2004). The impairment of circulatory function after large‐volume paracentesis could be associated with a shorter survival (Ginès 1996).

Description of the intervention

Vasoactive medications, including vasopressin and its analogue terlipressin, and somatostatin along with its analogues octreotide and vapreotide, produce splanchnic vasoconstriction and reduce portal blood inflow and portal pressure (Seo 2014). Vasopressin, terlipressin, somatostatin, and octreotide are used currently to control acute portal hypertensive gastrointestinal bleeding (Ioannou 2003; Gøtzsche 2008; Seo 2014; EASL 2018). Terlipressin together with albumin, noradrenaline, and midodrine together with octreotide have been evaluated in the treatment of refractory ascites and renal failure (EASL 2018). Terlipressin, noradrenaline, midodrine, vasopressin, and octreotide alone or together with albumin have been used in connection with paracentesis (Moreau 2002; Singh 2006a; Appenrodt 2008; Singh 2008).

Terlipressin is a long‐acting vasopressin analogue which induces splanchnic vasoconstriction. It requires intravenous infusion. Recently, continuous intravenous infusion of terlipressin was proposed, demonstrating a similar rate of response, but with lower adverse effects than the administration of the drug by intravenous boluses (EASL 2018). Terlipressin administration in connection with large paracentesis has been used at a dose of 3 mg, of which 1 mg is administered as an intravenous bolus at the onset of paracentesis, and the remaining as 1 mg doses intravenously at 8 hours and 16 hours after the first dose (Singh 2006a).

The most common adverse effects of terlipressin are hyponatraemia, diarrhoea, abdominal pain, circulatory overload, and cardiovascular and intestinal ischaemic complications which ranged from 13% to 45% of patients when delivered by intravenous bolus with a very high rate of discontinuation (Seo 2014; Zhou 2018).

Noradrenaline is an alpha‐adrenergic agonist with powerful vasoconstrictor activity, which constricts mesenteric vessels and reduces splanchnic vasodilatation. Its use requires a central venous line. In cirrhotic people, noradrenaline has been used in the treatment of type 1 hepatorenal syndrome (Duvoux 2002), as well as in connection with paracentesis. It is administered as continuous dose of 0.5 mg/hour and then titrated hourly in steps of 0.125 mg/hour to increase the mean arterial pressure by 10 mmHg above the baseline (maximum dose 3 mg/hour) and maintained at that level for the next 72 hours with close monitoring (Singh 2006b).

Midodrine is a less expensive and orally available alpha‐1 adrenoceptor agonist, which has been used with paracentesis alone or with octreotide (Singh 2008; Yosry 2019). The dose is usually 12.5 mg thrice daily for three days (Hamdy 2014).

Octreotide is an inhibitor of the release of powerful gastrointestinal vasodilator peptides, producing a net vasoconstrictor effect. In acute hypertensive bleeding, the recommended dose is a continuous infusion of 50 μg/hour after an initial bolus of 50 μg, for five days (Seo 2014). In connection with paracentesis, it has been used together with midodrine (Bari 2012).

How the intervention might work

Vasoactive drugs have been used in association with paracentesis, based on the assumption that if paracentesis‐induced arteriolar vasodilatation precedes and plays a major role in the development of circulatory dysfunction, then the administration of a splanchnic vasoconstrictor or of drugs reducing vasodilatation might be effective in the prevention of this derangement (Sola‐Vera 2004). Splanchnic vasoconstrictors might also balance paracentesis‐induced arteriolar vasodilatation, preventing complications associated with a decrease in effective arterial blood volume (Moreau 2002; Singh 2006a; Appenrodt 2008; Singh 2008). Vasoactive substances have been administered during or after paracentesis (Moreau 2002; Singh 2006a; Singh 2006b; Appenrodt 2008; Singh 2008; Bari 2012).

Why it is important to do this review

Two different therapeutic approaches have been used to counterbalance the circulatory derangement induced by paracentesis: administration of plasma expanders and vasoconstrictors.

Plasma expanders have been evaluated in meta‐analyses showing conflicting results (Wong 2008; Bernardi 2012; Wang 2015; Kütting 2017; Simonetti 2019). In our previous Cochrane systematic review, it was uncertain if differences existed between plasma expanders versus no plasma expander, or one plasma expander versus albumin, in terms of all‐cause mortality, serious adverse events, renal impairment, liver‐related complications, non‐serious adverse events, and recurrence of ascites (Simonetti 2019). Experimental plasma expanders seemed to increase the incidence of hyponatraemia and post‐paracentesis circulatory dysfunction compared to albumin, but again, the evidence in our review was of very low certainty. Considering the apparent lack of effect and high costs of plasma expanders, alternative treatments have been used, such as direct vasoconstrictors (terlipressin, noradrenaline, or midodrine) together or not with indirect vasoconstrictors, through the inhibition of vasodilator substances in the splanchnic bed (such as octreotide). The role of these drugs, after paracentesis in patients with cirrhosis, has been evaluated in small trials with 40 or fewer participants each, and the results were variable (Moreau 2002; Singh 2006a; Appenrodt 2008; Singh 2008; Bari 2012). Such small trials cannot reach definite conclusions on benefit or harm. This is due to the fact that strong prognostic factors cannot be evenly distributed in small randomised clinical trials. Two meta‐analyses, including five and eight trials respectively, did not find any difference between vasoconstrictors (terlipressin, noradrenaline, or midodrine) and albumin in terms of mortality, hyponatraemia, and post‐paracentesis circulatory dysfunction (Bernardi 2012; Kütting 2017). Other randomised clinical trials have been published after these meta‐analyses, so reassessment of the role of vasoconstrictors is needed. Moreover, none of the meta‐analyses were conducted as Cochrane systematic reviews, taking risk of bias (systematic errors) and play of chance (random errors) into consideration.

Objectives

To assess the benefits and harms of vasoactive drugs versus placebo or no intervention, or versus another vasoactive drug, or versus a plasma expander during or after paracentesis in people with cirrhosis and large ascites.

Methods

Criteria for considering studies for this review

Types of studies

We will include randomised clinical trials, irrespective of publication type, publication status, publication date, and language, assessing the benefits and harms of vasoactive drugs given before, during, or after paracentesis in people with cirrhosis and large ascites. We will not include quasi‐randomised studies (i.e. pseudo‐randomised studies), as the method of allocation to the study arms is not random. Randomised trials with cluster designs are unlikely to be found on the subject of our review. If we identify any, we will follow Cochrane instruction on how to include them (Higgins 2021a). If we find cross‐over trials, then we will include the data from the trial's first period (Higgins 2021a).

We will include trials irrespective of whether they report on the outcomes specified in our review.

Types of participants

We will include adults (age > 18 years) with liver cirrhosis and large ascites (with or without refractory ascites), treated by vasoactive drugs for paracentesis.

The diagnosis of cirrhosis can be based on histology or a combination of humoral, ultrasound, or endoscopic signs of portal hypertension (presence of oesophagastric varices and/or increase of portal diameter), or low levels of platelets and splenomegaly and signs of liver failure, or other signs of decompensation (jaundice or portosystemic encephalopathy, or both) and/or increase of aspartate transaminase/alanine transaminase (AST/ALT).

We will exclude randomised trials including people with acute‐on‐chronic liver failure or prior surgical and radiological procedures such as surgical large‐caval anastomosis, liver transplantation, and transjugular intrahepatic portosystemic shunt.

If a trial includes only a subset of relevant participants, we will extract data only on the subset of interest. If the data are not available, we will exclude the trial.

Types of interventions

• Vasoactive drugs (terlipressin, noradrenaline, midodrine, somatostatin, octreotide, and other vasoactive drugs), individually or in combination with each other versus placebo or no intervention, administered in connection to paracentesis.

• Vasoactive drugs (terlipressin, noradrenaline, midodrine, somatostatin, octreotide, and other vasoactive drugs), individually or in combination with each other versus another vasoactive drug, administered in connection to paracentesis.

• Vasoactive drugs (terlipressin, noradrenaline, midodrine, somatostatin, octreotide, and other vasoactive drugs), individually or in combination with each other versus a plasma expander, administered in connection to paracentesis.

The vasoactive drugs could have been administered at any dose, mode of administration, and duration.

We will allow co‐interventions if administered equally to all comparison groups in a trial.

Types of outcome measures

We will assess the following outcomes at the maximum available follow‐up.

Primary outcomes

1. All‐cause mortality.

2. Proportion of participants with one or more serious adverse events at maximal follow‐up, excluding those with liver‐related complications (see below). We will consider an event as a serious adverse event if the authors clearly stated that it was due to any of the trial interventions, and if it fulfilled the International Conference on Harmonization (ICH) Guidelines definition of serious adverse events (any event that leads to death; is life‐threatening; requires in‐patient hospitalisation or prolongation of existing hospitalisation; results in persistent or significant disability or congenital birth defect or anomaly; jeopardises the patient; or requires intervention to prevent it (ICH‐GCP 1997)). We will consider all other adverse events as non‐serious.

3. Health‐related quality of life, measured with validated scales or questionnaires.

Secondary outcomes

1. Proportion of participants with one or more liver‐related complications such as hypertensive gastrointestinal bleeding, hepatic encephalopathy, and bacterial infections.

2. Proportion of participants with adverse events considered not serious, defined as adverse events not comprised under the definition of serious adverse events as well as liver‐related complications.

3. Proportion of participants with kidney impairment (in participants with cirrhosis, it is defined as a serum creatinine (SCr) value ≥ 1.5 mg/dL (EASL 2018).

4. Proportion of participants developing refractory ascites (see the definition in Description of the condition).

5. Recurrence of ascites, defined as ascites that requires repeated paracentesis or hospitalisation, or both.

6. Hyponatraemia (serum Na+ concentration < 135 mmol/L).

7. Post‐paracentesis circulatory dysfunction, defined as an increase in the plasma renin activity of more than 50% of the pretreatment value to a level of more than 4 ng/mL/hour on the sixth day after paracentesis (Ginès 1996).

8. Proportion of dropouts and withdrawals due to adverse events.

We will conduct our primary analysis using the outcome data at the longest follow‐up, except for post paracentesis circulatory dysfunction which will be used according to its definition (see Description of the condition). We will also analyse mortality at 1‐ and 2‐year time points.

In the analysis on kidney failure or hyponatraemia, we will use MedCcalc® (www.mdcalc.com/) to convert to the units most frequently used in the trials.

Search methods for identification of studies

Electronic searches

We will search the Cochrane Hepato‐Biliary Group (CHBG) Controlled Trials Register (maintained and searched internally by the CHBG Information Specialist via the Cochrane Register of Studies Web), the Cochrane Central Register of Controlled Trials (latest issue), MEDLINE Ovid (1946 to the date of the search), Embase Ovid (1974 to the date of the search), LILACS (Bireme; 1982 to the date of the search), Science Citation Index Expanded (Web of Science; 1900 to the date of the search), and Conference Proceedings Citation Index (Web of Science; 1990 to the date of the search). Appendix 1 gives the preliminary search strategies.

Furthermore, we will search the Food and Drug Administration (FDA; www.fda.gov), European Medicines Agency (EMA; www.ema.europa.eu/ema/), World Health Organisation (WHO) International Clinical Trial Registry Platform (ictrptest.azurewebsites.net/Default.aspx), ClinicalTrials.gov (clinicaltrials.gov/), ISRCTN registry (www.isrctn.com/), and Google Scholar (http://http: //googlescholar.org), as well as pharmaceutical company sources, reference lists of potentially eligible studies, and relevant reviews, for ongoing or unpublished trials. We will also search for grey literature in the System for Information on Grey Literature in Europe (OpenGrey; www.opengrey.eu/).

For articles that we deem eligible for inclusion that are not in English or Italian, we will post a public call on the Cochrane TaskExchange page, where a native speaker can voluntarily assist the group with the overall content.

We will not do a systematic search for observational studies, though we are aware that the decisions to not search systematically for all observational studies and to extract data on harms might bias our review towards assessment of benefits and might overlook certain harms such as late or rare harms (Higgins 2021b). If, during the selection of trials, we identify observational studies such as quasi‐randomised studies or controlled or uncontrolled clinical studies with the same characteristics of participants and interventions as in our protocol and reporting adverse events relevant to the outcomes of this review, we will extract the adverse event data for experimental and control groups separately from the data found in the randomised clinical trials. We have decided to do so because uncommon and late adverse events can be most often found in observational studies, or in post‐marketing phase publications of use of the drugs and case‐reports. We will present the data on harm separately, without a formal meta‐analytic assessment. If we demonstrate benefits from vasoactive drugs for participants with cirrhosis and large or refractory ascites undergoing paracentesis, then a systematic review of the harms of this intervention in observational studies ought to be launched (Storebø 2018).

Searching other resources

We will check references of included trials, as well as meta‐analyses and other publications retrieved with the searches for randomised clinical trials in order to identify further trials of relevance to our review.

Data collection and analysis

We will perform the systematic review and meta‐analyses following recommendations of Cochrane (Higgins 2021b). We will perform the analyses using RevMan web (RevMan Web 2020). We will assess the certainty of the evidence with GRADE (GRADEpro GDT; Schünemann 2021a).

Selection of studies

In accordance with the inclusion criteria of this review protocol, two of the review authors (RGS and GP) will independently examine titles and abstracts of the retrieved records to remove obviously irrelevant studies. RGS and GP will then decide the validity of the remaining potentially relevant records based on the retrieved full‐text publications. RGS and GP will link together publications reporting on the same study (including observational reports of the included randomised clinical trials). Where appropriate, RGS and GP will correspond with study authors to verify study eligibility. As a last step in the process of final study selection, RGS and GP will list the excluded studies with the reasons for their exclusion in the 'Characteristics of excluded studies' section. Having made a final decision on the randomised clinical trials to be included, including trials from other resources, RGS and GP will proceed to data extraction. RGS and GP will include trials fulfilling the inclusion criteria of the current protocol no matter if they report on the outcomes of interest to the review. RGS and GP will resolve disagreements through discussion; in case of disagreement, CG will arbitrate. RGS and GP will record the selection process in sufficient detail to complete a PRISMA flow diagram (Moher 2009).

Data extraction and management

Two of the authors (RGS and GP) will independently extract data from the included trials and studies. RGS and GP will use a pre‐piloted data extraction sheet based on our protocol and one or two trials fulfilling the inclusion criteria. RGS and GP will resolve disagreements by discussion; in case of disagreement, CG will arbitrate.

We will extract the following.

1. Details of the study: country of origin, type of study report (article, conference abstract, letter or comment, etc.), study start date and study completion date, when the study was published, and the contact for correspondence.

2. Details of study methods: trial inclusion and exclusion criteria.

3. Details of participants: the comparability between the groups randomised to alternative treatments regarding baseline prognostic variables (aetiology of cirrhosis; mean age; proportion of males/females; participants with Child‐Pugh stages A, B, or C (Pugh 1973); proportion of participants with hepatic encephalopathy; previous bleeding episodes; type of ascites (refractory or not), or with hypotension; mean arterial pressure; kidney impairment; and hyponatraemia). Furthermore, we will record plasma renin activity, plasma aldosterone levels, and urinary sodium excretion.

4. Setting and context.

5. Interventions: type and dose of vasoconstrictors, and timing of administration of vasoconstrictors in connection to paracentesis.

6. Details on the associated treatments: type of paracentesis (partial or total paracentesis); total amount of ascitic fluid removed; length of the procedure; sodium restriction and diuretics (type and dose) before and after paracentesis; timing for clinical and laboratory assessment.

7. Outcomes (of the trial and for our protocol).

8. Results, and the time points of reporting the outcome results.

9. Other details: sample size calculation performed and reported; intention‐to‐treat or not; and study support.

10. Funding.

If a trial is published in a language other than English or Italian, we will find someone to translate the data we are interested in.

We will make a table with study characteristics based on PICOT (participants, interventions, comparison, outcomes, and time) in order to explore and compare study elements across studies. This will enable us to group studies for each of our comparisons. This information will then be entered in the 'Characteristics of included studies' table.

Assessment of risk of bias in included studies

We will assess the influence of the risk of bias on our results, using the Cochrane 'Risk of bias 2' (RoB 2) tool (Sterne 2019; Higgins 2021c).

Each assessment of the RoB 2 tool focuses on a specific outcome result from a randomised trial. The judgements are suggested by the assessor's answers to each of the signalling questions for each domain, and an algorithm in the tool. To record and manage the RoB 2 assessment and process, we will use the Excel tool developed for this purpose (Higgins 2020). We plan to provide detailed information in appendices. Two authors (RGS and GP) will assess the risk of bias independently. We will resolve disagreements through discussion, and if we fail to reach consensus, then another author (CG) will arbitrate.

We will assess the effect of assignment to the intervention. We will analyse participants in the intervention groups to which they were randomised, regardless of the intervention they actually received, and we will include all randomised participants in the outcome analyses; i.e. we will perform our analyses based on the intention‐to‐treat principle.

We will assess the risk of bias of all‐cause mortality; serious adverse events; health‐related quality of life; liver‐related complications; adverse events considered not serious; renal impairment; and development of refractory ascites outcomes.

We will use the below five domains to assess the risk of bias in the individual randomised trials (Sterne 2019; Higgins 2021c).

A. Bias arising from the randomisation process.
B. Bias due to deviations from intended interventions.
C. Bias due to missing outcome data.
E. Bias in measurement of the outcome.
F. Bias in selection of the reported result.

Table 1 shows the signalling questions for each of the five domains to reach the risk of bias judgements. The response options for the signalling questions are: (1) Yes; (2) Probably yes; (3) Probably no; (4) No; and (5) No information. Elaborations to these signalling questions can be found in Higgins 2021c.

1. Risk‐of‐bias judgements.

A. Risk‐of‐bias judgements for bias arising from the randomisation process
Low risk of bias	(1) The allocation sequence was adequately concealed AND (2.1) Any baseline differences observed between intervention groups appear to be compatible with chance, OR (2.2) There is no information about baseline imbalances AND (3.1) The allocation sequence was random OR (3.2) There is no information about whether the allocation sequence was random
Some concerns	(1.1) The allocation sequence was adequately concealed AND (1.2.1) The allocation sequence was not random OR (1.2.2) Baseline differences between intervention groups suggest a problem with the randomisation process OR (2.1) There is no information about concealment of the allocation sequence AND (2.2) Any baseline differences observed between intervention groups appear to be compatible with chance OR (3) There is no information to answer any of the signalling questions
High risk of bias	(1) The allocation sequence was not adequately concealed OR (2.1) There is no information about concealment of the allocation sequence AND (2.2) Baseline differences between intervention groups suggest a problem with the randomisation process
B. Risk‐of‐bias judgements for bias due to deviations from intended intervention (effect of assignment to intervention) This domain includes judgement on blinding of participants, carers, and people delivering interventions. Because the domain addresses two somewhat distinct issues, we separate the algorithm into two parts and combine them to reach the judgement.
	Part 1	Part 2	Criteria for the domain
Low risk of bias	(1) Participants, carers and people delivering the interventions were unaware of intervention groups during the trial   OR   (2.1) Participants, carers or people delivering the interventions were aware of intervention groups during the trial AND (2.2) No deviations from intended intervention arose because of the experimental context.	An appropriate analysis will be used to estimate the effect of assignment to intervention	Low risk of bias for Part 1 AND Low risk of bias for Part 2
Some concerns	(1) Participants, carers, or people delivering the interventions were aware of intervention groups during the trial AND (2.1) There is no information on whether there were deviations from intended intervention because of the experimental context OR [[(2.1.1) There were deviations from intended interventions that arose because of the experimental context] AND [[(2.1.1.1) These deviations were balanced between the intervention groups] OR [(2.1.1.2) These deviations were not likely to have affected the outcome]]]	(1) An appropriate analysis was not used to estimate the effect of assignment to intervention AND (2) The potential impact (on the estimated effect of intervention) of the failure to analyse participants in the group to which they were randomised was not substantial	Some concerns for Part 1 OR Some concerns for Part 2 AND Part 1 not High risk of bias AND Part 2 not High risk of bias
High risk of bias	(1) Participants, carers, or people delivering the interventions were aware of intervention groups during the trial AND (2) There were deviations from intended interventions that arose because of the experimental context AND (3) These deviations were unbalanced between the intervention groups AND (4) These deviations were likely to have affected the outcome	(1) An appropriate analysis was not used to estimate the effect of assignment to intervention AND (2) The potential impact (on the estimated effect of intervention) of the failure to analyse participants in the group to which they were randomised was substantial.	High risk of bias for Part 1 OR High risk of bias for Part 2
C. Risk‐of‐bias judgements for bias due to missing outcome data.
Low risk of bias	(1) Outcome data were available for all, or nearly all, randomised participants OR (2) There is evidence that the result was not biased by missing outcome data OR (3) Missingness in the outcome could not depend on its true value
Some concerns	(1) Outcome data were not available for all, or nearly all, randomised participants AND (2) There is no evidence that the result was not biased by missing outcome data AND (3) Missingness in the outcome could depend on its true value AND (4) It is not likely that missingness in the outcome depended on its true value
High risk of bias	(1) Outcome data were not available for all, or nearly all, randomised participants AND (2) There is no evidence that the result was not biased by missing outcome data AND (3) Missingness in the outcome could depend on its true value AND (4) It is likely that missingness in the outcome depended on its true value
D. Risk‐of‐bias judgements for bias in measurement of the outcome
Low risk of bias	(1) The method of measuring the outcome was not inappropriate AND (2) The measurement or ascertainment of the outcome did not differ between intervention groups AND (3.1) The outcome assessors were unaware of the intervention received by study participants OR (3.2) The assessment of the outcome could not have been influenced by knowledge of the intervention received
Some concerns	(1.1) The method of measuring the outcome was not inappropriate AND (1.2) The measurement or ascertainment of the outcome did not differ between intervention groups AND (1.3) The assessment of the outcome could have been influenced by knowledge of the intervention received AND (1.4) It is unlikely that assessment of the outcome was influenced by knowledge of intervention received OR   (2.1) The method of measuring the outcome was not inappropriate AND (2.2) There is no information on whether the measurement or ascertainment of the outcome could have differed between intervention groups AND (2.3.1) The outcome assessors were unaware of the intervention received by study participants OR (2.3.2) The assessment of the outcome could not have been influenced by knowledge of the intervention received
High risk of bias	(1) The method of measuring the outcome was inappropriate OR (2) The measurement or ascertainment of the outcome could have differed between intervention groups OR (3) It is likely that assessment of the outcome was influenced by knowledge of the intervention received
E. Risk‐of‐bias judgements for bias in selection of the reported result
Low risk of bias	(1) The data were analysed in accordance with a pre‐specified plan that was finalised before unblinded outcome data were available for analysis AND (2) The result being assessed is unlikely to have been selected, on the basis of the results, from multiple outcome measurements (e.g. scales, definitions, time points) within the outcome domain AND (3) Reported outcome data are unlikely to have been selected, on the basis of the results, from multiple analyses of the data
Some concerns	(1.1) The data were not analysed in accordance with a pre‐specified plan that was finalised before unblinded outcome data were available for analysis AND (1.2) The result being assessed is unlikely to have been selected, on the basis of the results, from multiple outcome measurements (e.g. scales, definitions, time points) within the outcome domain AND (1.3) The result being assessed is unlikely to have been selected, on the basis of the results, from multiple analyses of the data OR   (2) There is no information on whether the result being assessed is likely to have been selected, on the basis of the results, from multiple outcome measurements (e.g. scales, definitions, time points) within the outcome domain AND from multiple analyses of the data
High risk of bias	(1) The result being assessed is likely to have been selected, on the basis of the results, from multiple outcome measurements (e.g. scales, definitions, time points) within the outcome domain OR (2) The result being assessed is likely to have been selected, on the basis of the results, from multiple analyses of the data

For the precise wording of signalling questions and guidance for answering each one, see the full risk‐of‐bias tool at www.riskofbias.info.

The judgement options for the individual domains are: low risk of bias, some concerns, or high risk of bias. If we judge one domain with 'some concerns' or at 'high risk', the judgement will apply for the whole outcome.

Overall risk of bias judgement for a specific outcome

The overall risk of bias judgement for an outcome result is the same as for the individual domains.

• Low risk of bias: the trial is judged to be at low risk of bias for all domains for this result.

• Some concerns: the trial is judged to raise some concerns in at least one domain for this result, but not to be at high risk of bias for any domain.

• High risk of bias: the trial is judged to be at high risk of bias in at least one domain for this result, or the study is judged to have some concerns for multiple domains in a way that substantially lowers confidence in the result.

We will use the overall RoB 2 judgement per outcome (i.e. all‐cause mortality; serious adverse events; health‐related quality of life; liver‐related complications; adverse events considered not serious; renal impairment; and development of refractory ascites) to feed into the GRADE 'Summary of findings' table.

For cross‐over trials, we will use only data from the first period of the cross‐over, and we will use the standard version of RoB 2.

We will store our RoB 2 data, in Microsoft Excel, on a publicly available web site, the address of which we will provide at the review stage.

Measures of treatment effect

Dichotomous outcomes

We will use the risk ratio (RR) with 95% confidence interval (CI) for dichotomous outcomes.

Continuous outcomes

We will use the mean difference (MD) with 95% CI or the standard mean difference (SMD) for continuous outcomes, depending on whether the scales used in the trials were the same or different.

Unit of analysis issues

The unit of analysis will be the participant as randomised to the intervention group of a clinical trial. In trials with two‐group parallel design, we will compare the experimental intervention group versus the control group. In trials with more than two‐group parallel design, we will compare each experimental group with half of the control group if used within the same comparison.

In cross‐over trials, we will only include the relevant data from the first trial period in order to avoid residual effects from the treatment (Higgins 2021a). In order to avoid repeated observations on trial participants, we will analyse the trial data at the longest follow‐up, except for data on post paracentesis circulatory dysfunction, which will be used according to its definition (see Description of the condition; Higgins 2021a).

We do not expect cluster randomised trials, but if we find such trials, we will include them and assess bias risk according to the Cochrane Handbook for Systematic Reviews of Interventions (Higgins 2021a).

Dealing with missing data

We will try to obtain missing data from authors of the included trials. We will perform our analyses based on the intention‐to‐treat principle whenever possible, including all randomly assigned participants, irrespective of completeness of data. Otherwise, we will use the data available to us.

Regarding the primary outcomes, we will include participants with incomplete or missing data in sensitivity analyses according to the following two extreme case scenarios (Hollis 1999):

• Extreme case analysis favouring the experimental intervention ('best‐worst' case scenario): none of the drop‐outs/participants lost from the experimental arm, but all of the drop‐outs/participants lost from the control arm experienced the outcome, including all randomised participants in the denominator.

• Extreme case analysis favouring the control ('worst‐best' case scenario): all drop‐outs/participants lost from the experimental arm, but none from the control arm experienced the outcome, including all randomised participants in the denominator.

Assessment of heterogeneity

We will use the random‐effects meta‐analyses to allow for heterogeneity by assuming that the underlying effects follow a normal distribution, but we will interpret the results carefully (Deeks 2021). We expect clinical heterogeneity because of the stage and form of cirrhosis that participants in the included trials might have, and because of the different interventions administered across the trials. We will assess methodological heterogeneity by looking into trial design and if the outcomes across the trials are similarly defined and measured. We will check for statistical heterogeneity through visual inspection of the forest plots by using a standard Chi² test and a significance level of α = 0.1. In view of the low power of such tests, we will also examine heterogeneity by using the I² statistic (Higgins 2002). We will interpret the I² statistic as in Deeks 2021.

• 0% to 40%: might not be important.

• 30% to 60%: may represent moderate heterogeneity.

• 50% to 90%: may represent substantial heterogeneity.

• 75% to 100%: considerable heterogeneity.

Assessment of reporting biases

We will assess reporting biases that arise from missing outcome results in the identified trial publications, on all our primary and secondary outcomes (Primary outcomes; Secondary outcomes). To ascertain the missing data, we will compare outcome information and data results across the identified trial publications, in addition to contacting trial authors, and checking trial registers, trial protocols, or whatever else we can find, to verify that data were indeed missing. If we find non‐reporting of any study outcome of relevance to our review, we will try to explore the potential reasons. If the non‐reporting of a result is associated with an undesirable finding for the trial authors, then we will use the RoB 2 domain, 'bias in the selection of the reported result’, to judge the risk of bias of the specific outcome.

Whenever we have 10 or more trials, we will draw funnel plots to assess reporting biases from the individual trials by plotting the risk ratio (RR) on a logarithmic scale against its standard error (Egger 1997; Page 2021). We will examine the degree of asymmetry of the funnel plot.

Data synthesis

Meta‐analysis

We will use Review Manager Web to conduct meta‐analyses (RevMan Web 2020). We will follow the principles and methods of meta‐analysis as described in Chapter 10.2.1 of the Cochrane Handbook for Systematic Reviews of Interventions (Deeks 2021). We will use a random‐effects analysis approach because we expect that the included trials will be heterogeneous. However, before presenting the data synthesis, i.e. producing summary estimates of the treatment effect and presenting data using forest plots (where possible), we will ensure that the data presented in the study publication do not differ in magnitude and direction of effects with those in our review analysis. In case of differences, we will recheck the data. Our primary analysis will include all eligible trials, where outcome data are provided, no matter the risk of bias judgements.

If we deem a meta‐analysis not possible or not meaningful to perform, we will provide the reasons for our decision, and we will tabulate the results. Depending on the reason(s) for not performing a meta‐analysis, we will choose one of the alternative synthesis methods suggested in Table 12.1.a. of Chapter 12 of the Handbook (McKenzie 2021).

Subgroup analysis and investigation of heterogeneity

We plan to perform the following subgroup analyses.

• Trials at overall low risk of bias compared to trials with some concern of bias or at high risk of bias, because of the risk of overestimation of beneficial intervention effects and underestimation of harmful intervention effects in randomised clinical trials at risk of bias (Schulz 1995; Moher 1998; Kjaergard 2001; Wood 2008; Savović 2012a; Savović 2012b; Savović 2018).

• Type of vasoactive drugs, analysing separately randomised clinical trials according to the drug used to assess if there are differences of effect between different vasoactive drugs.

• For profit support, analysing separately randomised clinical trials without for‐profit support and trials with or unknown for‐profit support. There is no consensus on the role of for‐profit support on the results of the trials (Lundh 2017).

• Refractory ascites, analysing separately randomised clinical trials including participants with refractory ascites compared to trials including participants without refractory ascites. It is not known if paracentesis in the presence of refractory ascites is safe and if vasoactive drugs protect patients from its potentially deleterious effects.

• Severe hyponatraemia, analysing separately randomised clinical trials including participants with serum Na ≤ to 120 mEq/litre, and trials in which serum Na was > 120 mEq/litre.

• If we observe heterogeneity with I² > 40%, we will compare the trials with follow‐up less than the median follow‐up to the trials with a follow‐up above the median.

To determine whether a statistically significant subgroup difference is detected, we will consider the P value from the test for subgroup differences. We will use the test to assess the difference between the pooled effect estimates for each subgroup. A P value of less than 0.1 will show a significant subgroup effect.

Sensitivity analysis

We plan to do the following sensitivity analyses.

• Restricting the analysis to trials at overall ‘low’ risk of bias.

• Restricting the analysis to trials at overall ‘low/some concerns’.

• Excluding trials in which partial repeated paracentesis was used. The extraction of large amount of ascites in one session could affect the balance between the intra and extra vascular bed differently  in comparison with partial repeated paracentesis. Therefore, the different modality of paracentesis could affect the results.

• Excluding trials with a short follow‐up, as the haemodynamic effect of paracentesis, and the effect of the intervention could be different immediately after paracentesis compared to after a longer interval (≥ 1 month). Therefore, in case we observe substantial heterogeneity, we plan to evaluate the effect of durations of the interventions on our outcomes, using several durations of follow‐up.

• Fixed‐effect method to explore presence or absence of statistical heterogeneity of outcome results.

• Assessment of imprecision with Trial Sequential Analysis (see below).

Trial Sequential Analysis and assessment of imprecision 

Trial Sequential Analysis considers the choice of statistical model (fixed‐effect or random‐effects) and diversity (Thorlund 2017; TSA 2017). We will calculate the diversity‐adjusted required information size (DARIS; i.e. the number of participants needed in a meta‐analysis to detect or reject a certain intervention effect; Brok 2008; Wetterslev 2008; Brok 2009; Wetterslev 2009; Thorlund 2010; Wetterslev 2017).

The underlying assumption of Trial Sequential Analysis is that testing for statistical significance may be performed each time a new trial is added to the meta‐analysis. We will add the trials according to the year of publication and, if more than one trial was published in a year, we will add the trials alphabetically according to the last name of the first author. On the basis of the DARIS, we will construct the trial sequential monitoring boundaries for benefit, harm, and futility (Wetterslev 2008; Wetterslev 2009; Thorlund 2017; Wetterslev 2017). These boundaries determine the statistical inference one may draw regarding the cumulative meta‐analysis that has not reached the DARIS; if the trial sequential monitoring boundary for benefit or harm is crossed before the DARIS is reached, firm evidence may be established and further trials may be superfluous. If the boundaries for benefit or harm are not crossed, it is most probably necessary to continue doing trials in order to detect or reject a certain intervention effect.

In our Trial Sequential Analysis of the two primary dichotomous outcomes, we will base the DARIS on the event proportion in the control group; assuming a plausible relative risk reduction for mortality of 10% and a relative risk reduction for serious adverse events of 5%; a risk of type I error of 2.5% due to the three primary outcomes (Jakobsen 2014); a risk of type II error of 10%; and the diversity of the included trials in the meta‐analysis. For the continuous outcome ‐ health‐related quality of life ‐ we plan to estimate the DARIS using a minimal relevant difference of half the standard deviation of the meta‐analysis; the variance of the meta‐analysis; alpha of 2.5% due to the three primary outcomes (Jakobsen 2014); beta of 10%; and the diversity as estimated from the trials in the meta‐analysis (Wetterslev 2009). We also will calculate the Trial Sequential Analysis‐adjusted CIs (Thorlund 2017; Wetterslev 2017).

In our Trial Sequential Analysis of secondary outcomes, we will base the DARIS for dichotomous outcomes on the event proportion in the control group; we will make an assumption of a relative risk reduction of 10% for liver‐related complications, non‐serious adverse events, renal impairment, and refractory ascites; a type I error risk of 1.1% due to the eight secondary outcomes (Jakobsen 2014); a risk of type II error of 10%; and the diversity of the included trials in the meta‐analysis.

We will use Trial Sequential Analysis as sensitivity analysis in order to compare evaluation of imprecision with GRADE and with Trial Sequential Analysis for the same outcomes that we will present in the 'Summary of findings' tables (Jakobsen 2014; Castellini 2018; Gartlehner 2019); i.e. all‐cause mortality; serious adverse events; health‐related quality of life; liver‐related complications; adverse events considered not serious; renal impairment; and development of refractory ascites outcomes.  In Trial Sequential Analysis, we will downgrade our assessment of imprecision in GRADE (see below) by two levels if the accrued number of participants is below 50% of the DARIS, and one level if between 50% and 100% of the DARIS. We will not downgrade if futility or DARIS is reached. For continuous outcomes we will use a minimal relevant difference equal to SD/2, where SD is the standard deviation of the control group; a type I error risk of 2.0%; and a type II error risk of 10%.

A more detailed description of Trial Sequential Analysis and the software program can be found at www.ctu.dk/tsa/ (Thorlund 2017).

Summary of findings and assessment of the certainty of the evidence

We will assess the certainty of the evidence using the GRADE five factors to assess and present review comparisons and outcome results in 'Summary of findings' tables (GRADEpro GDT; Schünemann 2013; Schünemann 2021a; Schünemann 2021b). In these 'Summary of findings' tables, we will present the period of follow‐up with median/mean and our assessment of the following outcomes: all‐cause mortality; serious adverse events; health‐related quality of life; liver‐related complications; adverse events considered not serious; renal impairment; and development of refractory ascites. RGS and GP will grade the evidence of the listed outcomes independently of each other. RGS and GP will resolve disagreements through discussion. In case of disagreement, CG will arbitrate.

A 'Summary of findings' table consists of three parts: information about the review, a summary of the statistical results, and the grade of certainty of the evidence, i.e. 'high', 'moderate', 'low', or 'very low'. The certainty assessment of the available evidence comprises the number of studies, the types of studies (randomised or observational), and five factors including within study risk of bias, inconsistency of results (substantial heterogeneity), indirectness of evidence (differences in populations, interventions, controls, and outcomes), imprecision of results (calculating the optimal information size), and publication bias (see above) that affects the certainty of the evidence (Guyatt 2008; Balshem 2011; Guyatt 2011a; Guyatt 2011b; Guyatt 2011c; Guyatt 2011d; Guyatt 2011e; Guyatt 2011f; Guyatt 2011g; Guyatt 2011h; Guyatt 2013a; Guyatt 2013b; Guyatt 2013c; Guyatt 2013d; Mustafa 2013; Santesso 2016; Guyatt 2017). The five factors are used to make a judgement on whether the quality of the collected evidence should be downgraded or upgraded.

The below definitions are for grading of the certainty of evidence.

• High certainty: we are very confident that the true effect lies close to that of the estimate of the effect.

• Moderate certainty: 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 certainty: our confidence in the effect estimate is limited: the true effect may be substantially different from the estimate of the effect.

• Very low certainty: we have very little confidence in the effect estimate: the true effect is likely to be substantially different from the estimate of effect.

Appendices

Appendix 1. Search strategies

Database	Time span	Search strategy
The Cochrane Hepato‐Biliary Group Controlled Trials Register	Date will be given at review stage	(vasoactiv* or vasoconstrict* or vasodilat* or vasopress* or inotrop* or terlipressin* or somatostatin* or octreotid* or noradrenalin* or norepinephrin* or midodrin*) and (paracentes* or drain* or punctur* or centes*) and (cirrho* or fibro* or ascit*)
Cochrane Central Register of Controlled Trials (CENTRAL)	Latest issue	#1 MeSH descriptor: [Vasoconstrictor Agents] explode all trees #2 MeSH descriptor: [Vasodilator Agents] explode all trees #3 MeSH descriptor: [Vasopressins] explode all trees #4 MeSH descriptor: [Somatostatin] explode all trees #5 MeSH descriptor: [Octreotide] explode all trees #6 MeSH descriptor: [Norepinephrine] explode all trees #7 MeSH descriptor: [Midodrine] explode all trees #8 (vasoactiv* or vasoconstrict* or vasodilat* or vasopress* or inotrop* or terlipressin* or somatostatin* or octreotid* or noradrenalin* or norepinephrin* or midodrin*) #9 #1 or #2 or #3 or #4 or #5 or #6 or #7 or #8 #10 MeSH descriptor: [Paracentesis] explode all trees #11 (paracentes* or drain* or punctur* or centes*) #12 #10 or #11 #13 MeSH descriptor: [Fibrosis] explode all trees #14 MeSH descriptor: [Liver Cirrhosis] explode all trees #15 MeSH descriptor: [Ascites] explode all trees #16 (cirrho* or fibro* or ascit*) #17 #13 or #14 or #15 or #16 #18 #9 and #12 and #17
MEDLINE Ovid	1946 to the date of the search	1. exp vasoconstrictor agents/ or exp vasodilator agents/ 2. exp Vasopressins/ 3. exp Somatostatin/ 4. exp Octreotide/ 5. exp Norepinephrine/ 6. exp Midodrine/ 7. (vasoactiv* or vasoconstrict* or vasodilat* or vasopress* or inotrop* or terlipressin* or somatostatin* or octreotid* or noradrenalin* or norepinephrin* or midodrin*).mp. [mp=title, abstract, original title, name of substance word, subject heading word, floating sub‐heading word, keyword heading word, organism supplementary concept word, protocol supplementary concept word, rare disease supplementary concept word, unique identifier, synonyms] 8. 1 or 2 or 3 or 4 or 5 or 6 or 7 9. exp Paracentesis/ 10. (paracentes* or drain* or punctur* or centes*).mp. [mp=title, abstract, original title, name of substance word, subject heading word, floating sub‐heading word, keyword heading word, organism supplementary concept word, protocol supplementary concept word, rare disease supplementary concept word, unique identifier, synonyms] 11. 9 or 10 12. exp Fibrosis/ 13. exp Liver Cirrhosis/ 14. exp Ascites/ 15. (cirrho* or fibro* or ascit*).mp. [mp=title, abstract, original title, name of substance word, subject heading word, floating sub‐heading word, keyword heading word, organism supplementary concept word, protocol supplementary concept word, rare disease supplementary concept word, unique identifier, synonyms] 16. 12 or 13 or 14 or 15 17. 8 and 11 and 16
Embase Ovid	1974 to the date of the  search	1. exp vasoactive agent/ 2. exp inotropic agent/ 3. exp vasopressin/ 4. exp terlipressin/ 5. exp somatostatin/ 6. exp octreotide/ 7. exp noradrenalin/ 8. midodrine/ 9. (vasoactiv* or vasoconstrict* or vasodilat* or vasopress* or inotrop* or terlipressin* or somatostatin* or octreotid* or noradrenalin* or norepinephrin* or midodrin*).mp. [mp=title, abstract, heading word, drug trade name, original title, device manufacturer, drug manufacturer, device trade name, keyword, floating subheading word, candidate term word] 10. 1 or 2 or 3 or 4 or 5 or 6 or 7 or 8 or 9 11. exp paracentesis/ 12. (paracentes* or drain* or punctur* or centes*).mp. [mp=title, abstract, heading word, drug trade name, original title, device manufacturer, drug manufacturer, device trade name, keyword, floating subheading word, candidate term word] 13. 11 or 12 14. exp liver cirrhosis/ 15. exp liver fibrosis/ 16. exp ascites/ 17. (cirrho* or fibro* or ascit*).mp. [mp=title, abstract, heading word, drug trade name, original title, device manufacturer, drug manufacturer, device trade name, keyword, floating subheading word, candidate term word] 18. 14 or 15 or 16 or 17 19. 10 and 13 and 18 20. Randomized controlled trial/ or Controlled clinical trial/ or trial.ti. 21. (random* or blind* or placebo* or meta‐analys*).mp. [mp=title, abstract, heading word, drug trade name, original title, device manufacturer, drug manufacturer, device trade name, keyword, floating subheading word, candidate term word] 22. 19 and (20 or 21)
LILACS (Bireme)	1982 to the date of the search	(vasoactiv$ or vasoconstrict$ or vasodilat$ or vasopress$ or inotrop$ or terlipressin$ or somatostatin$ or octreotid$ or noradrenalin$ or norepinephrin$ or midodrin$) [Words] and (paracentes$ or drain$ or punctur$ or centes$) [Words] and (cirrho$ or fibro$ or ascit$) [Words]
Science Citation Index Expanded (Web of Science)	1900 to the date of the search	#4 #3 AND #2 AND #1 #3 TS=(cirrho* or fibro* or ascit*) #2 TS=(paracentes* or drain* or punctur* or centes*) #1 TS=(vasoactiv* or vasoconstrict* or vasodilat* or vasopress* or inotrop* or terlipressin* or somatostatin* or octreotid* or noradrenalin* or norepinephrin* or midodrin*)
Conference Proceedings Citation Index‐Science (Web of Science)	1990 to the date of the search	#4 #3 AND #2 AND #1 #3 TS=(cirrho* or fibro* or ascit*) #2 TS=(paracentes* or drain* or punctur* or centes*) #1 TS=(vasoactiv* or vasoconstrict* or vasodilat* or vasopress* or inotrop* or terlipressin* or somatostatin* or octreotid* or noradrenalin* or norepinephrin* or midodrin*)

Contributions of authors

RGS and GP wrote the protocol.

CG commented on the protocol.

All authors approved the protocol for publication.

Sources of support

Internal sources

• Rosa Simonetti, Italy

  No support

• GIoavanni Perricone, Italy

  No support

• Christian Gluud, Denmark

  Copenhagen Trial Unit/Cochrane Hepato‐Biliary Group

External sources

• Rosa Simonetti, Italy

  No support

• GIoavanni Perricone, Italy

  No support

• Christian Gluud, Denmark

  No support

Declarations of interest

RS: RS is a CHBG editor; nothing else to declare.
CP: nothing to declare.
CG: CG is the Co‐ordinating Editor of the CHBG; nothing else to declare.

New