Management of drainage for malignant ascites in gynaecological cancer

Abstract

Background

Ascites is the accumulation of fluid within the abdominal cavity. Most women with advanced ovarian cancer and some women with advanced endometrial cancer need repeated drainage for ascites. Guidelines to advise those involved in the drainage of ascites are usually produced locally and are generally not evidence‐based. Managing drains that improve the efficacy and quality of the procedure is key in making recommendations that could improve the quality of life (QoL) for women at this critical period of their lives.

Objectives

To evaluate the effectiveness and adverse events of different interventions for the management of malignant ascites drainage in the palliative care of women with gynaecological cancer.

Search methods

We searched CENTRAL, MEDLINE, and Embase to 4 November 2019. We checked clinical trial registries, grey literature, reports of conferences, citation lists of included studies, and key textbooks for potentially relevant studies.

Selection criteria

We included randomised controlled trials (RCTs) of women with malignant ascites with gynaecological cancer. If studies also included women with non‐gynaecological cancer, we planned to extract data specifically for women with gynaecological cancers or request the data from trial authors. If this was not possible, we planned to include the study only if at least 50% of participants were diagnosed with gynaecological cancer.

Data collection and analysis

Two review authors independently selected studies, extracted data, evaluated the quality of the included studies, compared results, and assessed the certainty of the evidence using Cochrane methodology.

Main results

In the original 2010 review, we identified no relevant studies. This updated review included one RCT involving 245 participants that compared abdominal paracentesis and intraperitoneal infusion of catumaxomab versus abdominal paracentesis alone. The study was at high risk of bias in almost all domains. The data were not suitable for analysis. The median time to the first deterioration of QoL ranged from 19 to 26 days in participants receiving paracentesis alone compared to 47 to 49 days among participants receiving paracentesis with catumaxomab infusion (very low‐certainty evidence). Adverse events were only reported among participants receiving catumaxomab infusion. The most common severe adverse events were abdominal pain and lymphopenia (157 participants; very low‐certainty evidence). There were no data on the improvement of symptoms, satisfaction of participants and caregivers, and cost‐effectiveness.

Authors' conclusions

Currently, there is insufficient evidence to recommend the most appropriate management of drainage for malignant ascites among women with gynaecological cancer, as there was only very low‐certainty evidence from one small RCT at overall high risk of bias.

Plain language summary

Methods for the drainage of fluid containing cancer cells that collect in the abdomen in women with gynaecological cancer

Background

Malignant ascites is the build‐up of fluid within the abdominal cavity caused by underlying cancer. Women with advanced ovarian cancer and some women with advanced uterine cancer (also known as womb cancer) often need drainage for malignant ascites to alleviate discomfort. Guidelines to advise healthcare professionals involved in the drainage of ascites are usually produced locally and are generally based on clinicians' experience.

Study characteristics

We searched for studies up to Novemeber 2019 that compared different ways of managing the drainage of fluid collected in the abdomen of women with gynaecological cancer (cancer that starts in a woman's reproductive organs).

Key results and quality of evidence

The original 2010 review found no relevant studies. This updated review included one randomised controlled trial (RCT: a type of study in which people are randomly assigned to receive different treatments) involving 245 women that compared drainage combined with catumaxomab (a medicine used to treatment malignant ascites) versus drainage alone. However, the results were insufficient to assess the difference between these treatments. Although women receiving drainage combined with catumaxomab had better quality of life (the general well‐being of a person) for longer compared to drainage alone, we are very unsure of this evidence due to the small number of participants and trials. There were some side effects in the drainage plus catumaxomab group (e.g. pain, low white blood cell count), but they were not well reported. At present, there is insufficient data regarding the best management of drainage for malignant ascites among women with gynaecological cancer.

The evidence is current to 4 November 2019.

Background

Description of the condition

The peritoneum is a membrane that lines the abdomen. It is made up of two layers: one encloses the organs (such as the lower intestines) and the other lines the inside of the muscle wall of the abdomen. The peritoneum produces small amounts of fluid that lubricate the two layers so that they slide easily over one another as a person moves. Ascites is the name given to an accumulation of this fluid within the abdominal cavity. It is probably caused by a combination of several factors, including the production of excess fluid in response to inflammation, and fluid not draining away as it would normally.

In malignant disease, ascites may be present at diagnosis and also when the disease recurs (Tamsma 2007). Treatment with chemotherapy will often be successful in preventing ascites but in recurrence, when treatment is no longer effective or not a therapeutic option, ascites can be a persistently problematic symptom. Ascites is most commonly associated with cancer of the ovary which, together with tumours originating in the breast, bowel, pancreas, and endometrium, account for 80% of cases of malignant ascites in women (Wilailak 1999).

Recurrent malignant ascites causes unpleasant symptoms that significantly reduce the quality of life (QoL) of women with advanced cancer (Harding 2012). The accumulation and volume of fluid are difficult to predict, so women often have to be admitted to hospital as an emergency with a variety of symptoms including distension of the abdomen, anorexia, discomfort, nausea, constipation, and breathlessness (Meyer 2016).

Description of the intervention

Management of refractory malignant ascites takes place in the context of palliative care and aims to improve the QoL of people with cancer (Wimberger 2012). The management of symptomatic malignant ascites consists of both mechanical interventions that aim to drain the ascites from the peritoneal cavity, and pharmacological interventions that prevent and diminish the development of ascites (Hodge 2019).

The most common mechanical intervention is abdominal paracentesis. Paracentesis is effective in relieving symptoms of nausea, vomiting, dyspnoea, and abdominal discomfort. However, paracentesis requires frequently repeated treatments to maintain symptom control (Hodge 2019). Options of mechanical drainage for durable symptom control include indwelling catheters, peritoneal ports, and peritoneo‐venous shunts (Hodge 2019).

Pharmacological interventions include diuretics and immunological agents (Meyer 2016). Pharmacological interventions are typically used as an adjunct to mechanical interventions. The most common diuretics used in management of malignant ascites are spironolactone (potassium‐sparing aldosterone antagonist) and furosemide (loop diuretic) (Meyer 2016). People with cancer most likely to respond to diuretic therapy are those with portal hypertension‐associated ascites caused by large liver metastases (Pockros 1992). Another pharmacological intervention for managing malignant ascites is an immunological agent, catumaxomab, which can reduce the requirement of paracentesis among women with malignant ascites (Berek 2014).

How the intervention might work

Repeated abdominal paracentesis is a widely used and effective procedure that can provide good symptom relief in the short term (Lee 1998). This involves the placement of a fine tube into the abdomen which remains in place for several hours – sometimes days – to allow fluid to be released from the body to provide relief from symptoms. Other methods of drainage include the insertion of permanent tunnelled catheters and peritoneo‐venous shunts which are options for people with cancer who are expected to require frequent paracenteses (MacDonald 2006; Meyer 2016; Hodge 2019).

Diuretics are common pharmacological interventions for managing malignant ascites. The goal of diuretic therapy is a decrease in overall plasma volume to control the rate of reaccumulation of ascites (Meyer 2016).

Another pharmacological intervention for managing malignant ascites is catumaxomab, a trifunctional monoclonal antibody that has two antigen‐binding sites and a functional Fc (fragment crystallisable) domain. The antigen‐binding sites bind to T‐cells through CD3 receptors and to tumour cells through epithelial cellular adhesion molecule (EpCAM) (Meyer 2016). In heavily pretreated women with chemotherapy‐refractory ovarian cancer and recurrent symptomatic malignant ascites, intraperitoneal infusion of catumaxomab reduces the requirement of abdominal paracentesis and improves QoL by relieving ascites symptoms (Wimberger 2012; Berek 2014). Catumaxomab was initially authorised for market by the European Medicines Agency for the treatment of malignant ascites in adults with EpCAM‐positive carcinomas where standard therapy was not available or no longer feasible. Its market authorisation, however, was withdrawn in the EU in June 2017 at the manufacturer's request due to the insolvency of the manufacturer. At present, there are no other manufacturers of this immunological agent.

Why it is important to do this review

Women with advanced gynaecological cancer often need repeated drainage for malignant ascites (Jatoi 2005). Women often wait for as long as possible before seeking intervention or are advised by healthcare professionals to wait for drainage until there is a large enough volume of fluid to ensure that it is suitable for safe drainage. This means that women with malignant ascites often experience fatigue, discomfort, anorexia, breathlessness, constipation, and increased frequency of micturition before hospital admission to be therapeutically drained. Furthermore, the increase in treatment options in recent years would allow time for an increase in incidence of malignant ascites. From 2007 to 2008, malignant ascites accounted for over 28,000 bed‐days in hospitals in England (HES).

Guidelines to advise those involved in the drainage of ascites have mainly been produced locally by teams and much of this is not evidence‐based but mainly based on clinicians' anecdotal evidence and experience (MacDonald 2006). They can be outdated and variable within the same setting as the procedure can take place in several areas within the same hospital. The lack of national standards or guidance may lead to inequity and varying levels of care, which may adversely affect the QoL of women who require regular hospital admission for drainage of ascites. Becker 2006 carried out a systematic review of the evidence from all types of studies on the effectiveness of paracentesis, diuretics, and peritoneo‐venous shunting in the management of malignant ascites, but did not report QoL outcomes. Ascites is a major aspect of the long‐term consequence of gynaecological cancer and effective, consistent, evidence‐based management would clearly impact on the QoL for this group of women (MacDonald 2006). 

Objectives

To evaluate the effectiveness and adverse events of different interventions for the management of malignant ascites drainage in the palliative care of women with gynaecological cancer.

Methods

Criteria for considering studies for this review

Types of studies

We planned to include individual randomised controlled trials (RCTs). If there were no individual RCTs, we planned to include non‐randomised studies (NRS) with a parallel comparison. It is unlikely that cross‐over or cluster‐RCTs could be designed to evaluate the interventions of interest.

Types of participants

Women with malignant ascites who had a confirmed histological diagnosis of all types of gynaecological cancer. If studies included women with non‐gynaecological cancer (i.e. pancreatic, colon, gastric, or hepatobiliary cancer), we planned to extract data specifically for women with gynaecological cancers or to contact the trial authors to request data related to women with gynaecological cancer. If this is not possible, we planned to include the study only if at least 50% of participants were diagnosed with gynaecological cancer.

Types of interventions

We included any studies that attempted to compare the following:

• drainage technique (i.e. repeated abdominal paracentesis; insertion of permanent tunnelled catheters; or peritoneo‐venous shunts) versus no drainage;

• drainage technique versus a different drainage technique;

• combinations of intervention (i.e. drainage technique plus medication) versus no drainage;

• combinations of interventions versus single intervention;

• combinations of interventions versus other combinations of interventions.

Types of outcome measures

Primary outcomes

• Quality of life (QoL), measured using a validated scale (i.e. European Organization for the Research and Treatment of Cancer (EORTC) Core Quality of Life Questionnaire (QLQ) C30; EORTC 1995).

Secondary outcomes

• Improvement of ascites‐related symptoms, measured using a validated scale (i.e. Edmonton Symptom Assessment System–Ascites Modification (ESAS:AM; Easson 2007), the Functional Assessment of Chronic Illness Therapy‐Ascites Index (FACIT‐AI; Cella 2013)), or as defined by the trial authors

• Adverse events: overall rate of adverse events and rates of the following adverse events: infectious complication (i.e. intra‐abdominal infection, intra‐abdominal collection, peritonitis, sepsis); digestive complications (i.e. bowel perforation, accidental bowel puncture, gut obstruction); cardiovascular and thromboembolic complications (i.e. hypotension, pulmonary embolism, venous thrombosis, disseminated intravascular coagulopathy); catheter‐related complications (i.e. catheter placement failure, catheter occlusion). We planned to categorise the severity of the adverse events according to the Common Terminology Criteria for Adverse Events (CTCAE).

• Participant satisfaction, measured by either a binary response (i.e. yes/no or satisfactory/unsatisfactory) or a validated rating measure such as the Visual Analogue Scale (VAS), Verbal Rating Scale (VRS), or Numerical Rating Scale (NRS).

• Caregiver satisfaction, determined by either a binary response (i.e. yes/no or satisfactory/unsatisfactory) or a validated rating measure such as the VAS, VRS, or NRS.

• Cost‐effectiveness, measured using a scale that was validated and reported in a peer‐reviewed publication (i.e. European Society for Medical Oncology Magnitude of Clinical Benefit Scale; ESMO‐MCBS; Cherny 2017).

We presented a 'Summary of findings' tables reporting all outcomes listed in order of priority (see Data synthesis).

• QoL.

• Improvement of ascites‐related symptoms.

• Adverse events.

• Participant satisfaction.

• Caregiver satisfaction.

• Cost‐effectiveness.

Search methods for identification of studies

We searched for studies, irrespective of the language of publication, publication status, or sample size.

Electronic searches

For this update, we searched the following electronic databases up to 4 November 2019:

• the Cochrane Central Register of Controlled Trials (CENTRAL; 2019, Issue 11), see Appendix 1;

• MEDLINE via Ovid (March 2009 to October week 4 2019), see Appendix 2;

• Embase via Ovid (March 2009 to 2019 week 44), see Appendix 3.

Searching other resources

Ongoing studies and grey literature

We searched the World Health Organization International Clinical Trials Registry Platform (WHO ICTRP; www.who.int/ictrp/ en/) and ClinicalTrials.gov (clinicaltrials.gov) to identify any ongoing trials. If we had identified ongoing trials that had not been published, we planned to approach the principal investigators to ask for relevant data. We searched the following databases for grey literature: Open‐Grey (www.opengrey.eu/) and Index to Theses (ProQuest Dissertations & Theses: UK & Ireland).

Handsearching

We handsearched the citation lists of included studies and key textbooks, and previous systematic reviews. We also handsearched the reports of conferences in the following sources:

• Annual Meeting of the American Society of Gynecologic Oncology (ASGO);

• Annual Meeting of European Society of Medical Oncology (ESMO);

• Annual Meeting of the American Society of Clinical Oncology (ASCO);

• Annual Meeting of the British Gynaecological Cancer Society (BGCS);

• Biennial Meeting of the Asian Society of Gynecologic Oncology (ASGO);

• Biennial Meeting of Asia and Oceania Federation of Obstetrics and Gynaecology (AOFOG);

• Biennial Meeting of the European Society of Gynaecological Oncology (ESGO);

• Biennial Meeting of the International Gynecologic Cancer Society (IGCS).

We planned to include unpublished studies only if study outcomes and methodological descriptions were adequately provided in written form or via direct contact with the trial authors.

Data collection and analysis

Selection of studies

After the removal of duplicates of all titles and abstracts retrieved by electronic searching, we transferred these data to Covidence (www.covidence.org). Two review authors (CK and SR) independently examined the remaining references. We excluded studies that clearly did not meet the inclusion criteria, and we obtained copies of the full‐text of potentially relevant references. Two review authors (CK and SR) independently assessed the eligibility of the retrieved reports/publications. We planned to resolve any disagreements through discussion or when required, we planned to consult a third review author (JS or PL). We planned to identify and exclude duplicates and collated multiple reports of the same study so that each study rather than each report was the unit of interest in the review. We applied the details regarding the selection process in Covidence (www.covidence.org) to complete a PRISMA flow diagram and Characteristics of excluded studies table (Liberati 2009).

Data extraction and management

Two review authors (CK and SR) independently assessed study characteristics and outcome data from included studies using Covidence (www.covidence.org). We intended to note in the Characteristics of included studies table if outcome data were not reported in a usable way. We planned to resolve disagreements by consensus or by involving a third review author (JS/PL). One review author (NJ) checked study characteristics for accuracy against the trial report.

For the included studies, we planned to extract the following data.

• Author, year of publication, and journal citation (including language).

• Country.

• Inclusion and exclusion criteria.

• Study methodology (RCT or non‐RCTs, type of non‐RCTs).

• Study population and disease characteristics: including numbers of participants with gynaecological cancer (only studies involving mixed types of cancer), the total number enrolled/included; demographic characteristics of participants, performance status of the participants, and stages and types of cancer.

• Details of intervention and comparison: including drainage technique; type of drain; type of healthcare professional who manages the drainage; site for drainage; length of time drains should be in situ; frequency of drainage; volume of ascites drained.

• Risk of bias in the included study (see below).

• Duration of follow‐up, numbers lost to follow‐up, and deviations from the protocol.

• Outcomes: for each outcome, outcome definition (with diagnostic criteria if relevant) and unit of measurement; for adjusted estimates, we planned to record variables adjusted for in analyses.

• Results: number of participants allocated to each intervention group, total number analysed for each outcome and the missing participants.

• Notes: funding for trial and notable conflicts of interest of trial authors.

Where possible, we extracted all data according to an intention‐to‐treat basis, in which participants were analysed in groups to which they were assigned. We planned to note the time points at which outcomes were collected and reported.

If we found more than one publication of the same study, we planned to use the most recent publication for data extraction and collate multiple reports of the same study as the unit of interest in the review.

Assessment of risk of bias in included studies

Two review authors (CK and SR) independently applied the 'Risk of bias' tool and resolved differences by discussion or by appeal to a third review author (JS or PL). We planned to assess and report on the methodological quality and risk of bias in included studies in accordance with the Cochrane Handbook for Systematic Reviews of Interventions (Higgins 2019), which recommends the explicit reporting of the following individual elements for RCTs (Appendix 4).

• Selection bias: random sequence generation and allocation concealment.

• Performance bias: blinding of participants and personnel (people and treatment providers).

• Detection bias: blinding of outcome assessment.

• Attrition bias: incomplete outcome data.

• Reporting bias: selective reporting of outcomes.

• Other possible bias.

If no RCTs were noted, we planned to include NRS. We planned to assess the risk of bias in NRS according to the Cochrane 'Risk of Bias' in non‐randomised Studies of Interventions (ROBINS‐I) tool (Stern 2016). We planned to classify NRS at high risk of bias when they were at serious risk according to the Cochrane ROBINS‐I tool. With the ROBINS‐I, the included studies are assessed for their risk of bias in the following seven domains (Appendix 5).

• Bias due to confounding.

• Bias in the selection of participants into the study.

• Bias in the classification of interventions.

• Bias due to deviations from intended intervention.

• Bias due to missing data.

• Bias in measurement of outcomes.

• Bias in the selection of the reported result.

We planned to provide a quote from the study report or a statement (or both) as a justification for the judgement for each item in the 'Risk of bias' table. When interpreting treatment effects and meta‐analyses, we planned to take into account the risk of bias for the studies that contributed to that outcome. Where information on risk of bias related to unpublished data or correspondence with an original author, we planned to note this in the table.

Measures of treatment effect

We planned to apply the following measures for the effect of treatment.

• For dichotomous outcomes (e.g. adverse events, participant satisfaction, caregiver satisfaction), we planned to analyse data based on the number of events and the number of people assessed in the intervention and comparison groups. We planned to use these to calculate the risk ratio (RR) and 95% confidence interval (CI).

• For continuous outcomes (e.g. QoL measures, cost‐effectiveness, and satisfaction score), we planned to analyse data based on the mean, standard deviation (SD), and number of people assessed for both the intervention and comparison groups to calculate the mean difference (MD) between treatment arms with the 95% CI. If the MD was reported without individual group data, we planned to use this to report the study results. If more than one study measured the same outcome using different tools, we planned to calculate the standardised mean difference (SMD) and 95% CI using the inverse variance method.

However, available information from the included study was not suitable to measure the difference of the effects in interventions between the comparison groups. Therefore, we applied the narrative review approach to report the effects of interventions.

Unit of analysis issues

The unit of analysis of individual RCT was per woman randomised and for NRS, the participant receiving the intervention. We planned to follow the methods stated in the Cochrane Handbook for Systematic Reviews of Interventions for carrying out the calculations or determining the statistical outcomes (Higgins 2019). In a study with multiple intervention groups, where possible, we planned to combine all relevant experimental intervention groups into a single group to create a single pair‐wise comparison (Higgins 2019). However, only one study that randomised individual participants into one of two interventions met the review inclusion criteria. Therefore, there was no requirement to review unit of analysis issues. This may be required in updates of this review, should additional studies become available.

Dealing with missing data

We planned to report the percentage of observations with missing data in each included study. We planned to not impute missing outcome data for any of the outcomes.

Assessment of heterogeneity

Heterogeneity is clinically assessed by visual inspection of forest plots. We planned to also assess statistical heterogeneity in each meta‐analysis using the I² statistic and Chi² test (Higgins 2003; Higgins 2019). We planned to regard heterogeneity as substantial if the I² statistic was greater than 50%, or there was a low P value (less than 0.10) in the Chi² test for heterogeneity (Deeks 2001; Higgins 2019). If there was substantial statistical heterogeneity (I² greater than 60%), we planned to carry out subgroup analyses to assess differences between the included studies. We did not assess heterogeneity because this review included only one trial.

Assessment of reporting biases

We planned to examine funnel plots corresponding to a meta‐analysis of the primary outcome to assess the potential for small‐study effects if the search had identified more than 10 studies. We planned to assess funnel plot asymmetry visually, and if asymmetry of funnel plots was identified, we planned to perform exploratory analyses to investigate it (Sterne 2011). However, we did not assess the potential of reporting biases, as there was only one included study available for this review.

Data synthesis

If sufficient, clinically similar studies were available, we planned to pool their results in meta‐analyses. We planned to use adjusted summary statistics if available; otherwise we planned to use unadjusted results. We planned to apply the random‐effects model with an inverse variance weighting for all meta‐analyses (DerSimonian 1986). We planned to perform statistical analysis using Review Manager 2014.

• For any dichotomous outcome (e.g. adverse events, participant satisfaction, and caregiver satisfaction), we planned to calculate the RRs for each study and then pool them.

• For continuous outcome (e.g. QoL measures, cost‐effectiveness, and satisfaction score), we planned to pool the MDs between the treatment arms, if all trials measure the outcome on the same scale; otherwise, we planned to pool the SMDs.

We planned to use forest plots to display the results of studies examining the same outcome. If there was evidence of substantial clinical and methodological heterogeneity across the included studies, we planned to apply a narrative review approach to data synthesis.

Main outcomes of 'Summary of findings' table for assessing the certainty of the evidence

We prepared a 'Summary of findings' table to summarise the results of the review based on the methods described in Chapter 11 of the Cochrane Handbook for Systematic Reviews of Interventions (Schünemann 2011). We planned to present the results of the meta‐analysis for the outcomes as outlined in the Types of outcome measures section.

We presented, where possible, the overall certainty of the evidence for each outcome according to the GRADE approach, which takes into account issues related to internal validity (risk of bias, inconsistency, imprecision, publication bias) and to external validity such as directness of results (Langendam 2013). We created a 'Summary of findings' table based on the methods described in the Cochrane Handbook for Systematic Reviews of Interventions (Higgins 2019) and using GRADEpro GDT. We used the GRADE checklist and GRADE Working Group quality of evidence definitions (Meader 2014). We downgraded the evidence from high quality by one level for each serious (or by two for each very serious) limitation.

• 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.

Subgroup analysis and investigation of heterogeneity

As we included only one RCT (two reports), there was no need to conduct sensitivity analyses. If appropriate, we will carry out subgroup analyses in updates of the review to assess the effect of the different types of drainage procedure. We also plan to assess subgroup differences by interaction tests available within Review Manager 2014.

Sensitivity analysis

We were unable to conduct a sensitivity analysis because we included only one study. If we include more studies in review updates, we will conduct a sensitivity analysis for the primary outcome to determine whether the conclusions are robust to arbitrary decisions made regarding the methodological aspects, as follows.

• Repeating the analysis excluding RCTs high or unclear risk of bias for allocation concealment.

• Repeating the analysis excluding studies with more than 20% missing data at the time points at which outcomes were reported.

• Repeating the analysis excluding unpublished studies, if any.

Results

Description of studies

See Figure 1; Characteristics of included studies table; and Characteristics of excluded studies table.

1.

Study flow diagram.

Results of the search

The original 2010 review identified 1664 unique records (Keen 2010). Eighteen articles were retrieved in full. The full‐text screening excluded all 18 studies.

In this updated review, an additional search of the literature in November 2019 yielded the following results; 44 references from CENTRAL, 668 references from MEDLINE, and 1596 references from Embase. Searching for other sources yielded two additional references. After deduplication, we additionally screened titles and abstracts of 2145 references and excluded 2117 that obviously did not meet the review inclusion criteria. Of the 28 references that potentially met the review inclusion, we excluded 26 reports after reviewing the full texts and add one ongoing study (see Characteristics of excluded studies table; Characteristics of ongoing studies table). Figure 1 displays the PRISMA flow chart for study selection. We included two reports of one RCT in this updated review.

Included studies

One RCT (two reports) met the review inclusion criteria and was included in this review (Heiss 2010).

Participants

Heiss 2010 was a randomised, multicentre, phase II/III study of people with symptomatic malignant ascites due to EpCAM‐positive cancer. The intervention was paracentesis plus catumaxomab (catumaxomab group) compared to paracentesis alone (control group). Participants were classified as ovarian cancer and non‐ovarian cancer cases. The study included 129 women with ovarian cancer and 129 women with non‐ovarian cancer. These 258 participants were randomised to the treatment groups in a 2:1 ratio. See Characteristics of included studies table for details of inclusion and exclusion criteria. Of 129 women with ovarian cancer, 85 were randomised to the catumaxomab group and 44 were randomised to the control group. In a second report from the study (Wimberger 2012, in: Heiss 2010), data regarding QoL of 160 women allocated to the catumaxomab group and 85 women allocated to the control group were available for analyses. Approximately 50.6% of participants were women diagnosed with ovarian cancer (Wimberger 2012, in: Heiss 2010).

Interventions

The intervention was paracentesis plus catumaxomab (catumaxomab group) compared to paracentesis alone (control group). After draining the ascites fluid, catumaxomab was administered by intraperitoneal infusion on day 0 at 10 μg, day 3 at 20 μg, day 7 at 50 μg, and day 10 at 150 µg. The control group was treated with one therapeutic paracentesis only on day 0. In both groups, repeated therapeutic paracentesis was carried out if the participant required relief of ascites symptoms. In addition, a therapeutic paracentesis was performed when ascites volume was greater than 1 L and there was the presence of signs and symptoms of ascites assessed by the investigators.

Outcomes

Outcomes of interest in the first report included puncture‐free survival which was defined as the time to first need for therapeutic puncture or death after treatment, whichever occurred first, time to next paracentesis, signs and symptoms of ascites, and overall survival (Heiss 2010). In Heiss 2010, outcomes of women with ovarian cancer were reported separately to those of women with non‐ovarian cancer. In the second report (Wimberger 2012, in: Heiss 2010), outcome of interest was time to first deterioration in health‐related QoL. QoL was evaluated using the EORTC QLQ‐C30 questionnaire (qol.eortc.org/questionnaire/eortc‐qlq‐c30/). The deterioration of QoL was defined as a decrease of at least 5 points in scores. Outcomes of women with ovarian cancer were reported together with those non‐ovarian cancer participants (Wimberger 2012 in: Heiss 2010).

Compliance and follow‐up

Of 85 women with ovarian cancer who were allocated to catumaxomab, five did not receive at least one dose of catumaxomab and 13 were excluded due to major protocol deviation. Of 44 women in the control group, two were excluded secondary to major protocol deviation (Heiss 2010). Approximately 5% of participants did not complete at least one of the EORTC QLQ‐C30 items at screening and then were excluded from the analysis of the second report (Wimberger 2012, in: Heiss 2010).

Excluded studies

In the original 2010 review (Keen 2010), the authors excluded 18 studies. In this updated review, we excluded a further 26 reports after reviewing the full texts from the latest search. Overall, 43 full‐text articles were excluded due to wrong study design (37), wrong participant population (five), and wrong intervention of interest (one) (Characteristics of excluded studies table), and we identified one ongoing trial awaiting evaluation once the results are made available (Characteristics of ongoing studies table).

Risk of bias in included studies

Allocation

There was no statement regarding the methods of random sequence generation applied in this study. Thus, this was at unclear risk of bias for this domain. Heiss 2010 was an open‐label RCT which had high risk of bias related to allocation concealment (Figure 2).

2.

Risk of bias summary: review authors' judgements about each risk of bias item for each included study.

Blinding

As Heiss 2010 was a randomised, open‐label study, it had a high risk of performance bias and detection bias, as it is likely to have had a significant effect on the outcome (Figure 2).

Incomplete outcome data

After randomisation, approximately 15.3% of the women with ovarian cancer in catumaxomab group were excluded from the analysis of the first report due to major protocol deviation. Approximately 5% of participants did not complete at least one of the EORTC QLQ‐C30 items at screening and then were excluded from the analysis of the second report. This indicated a low risk of attrition bias (Figure 2).

Selective reporting

There was no reported comparison of adverse event between the treatment groups (Heiss 2010). In addition, there was no information regarding adverse event in the subsequent report of this RCT (Wimberger 2012, in: Heiss 2010). Therefore, this study was at high risk of reporting bias (Figure 2).

Other potential sources of bias

This study received a grant from Fresenius Biotech GmbH, the marketing authorisation holder for Removab (catumaxomab) who developed this trifunctional antibody together with its partner TRION Pharma GmbH, Germany. Several authors of the included study served as consultant and advisory board members for this company (Figure 2).

Effects of interventions

See: Table 1

Summary of findings for the main comparison. Abdominal paracentesis and intraperitoneal infusion of catumaxomab compared with abdominal paracentesis alone for management of drainage for malignant ascites in gynaecological cancer.

Abdominal paracentesis and intraperitoneal infusion of catumaxomab compared with abdominal paracentesis alone for management of drainage for malignant ascites in gynaecological cancer
Patient or population: women with malignant ascites Settings: inpatient settings Intervention: abdominal paracentesis plus intraperitoneal infusion of catumaxomab Comparison: abdominal paracentesis alone
Outcomes	Illustrative comparative risks* (95% CI)		Relative effect (95% CI)	No of participants (studies)	Certainty of the evidence (GRADE)	Comments
	Assumed risk	Corresponding risk
	Abdominal paracentesis alone	Abdominal paracentesis with catumaxomab infusion
QoL	Median time to the first deterioration of QoL ranged from 19 to 26 days (see comment)	Median time to the first deterioration of QoL varied from 47 to 49 days	See comment	245 participants (1 study)	⊕⊝⊝⊝a,b Very low	Data from Wimberger 2012, in: Heiss 2010 was not suitable to analyse the relative measure between the comparison groups.
Improvement of ascites‐related symptoms	—		—	—	—	Not reported
Adverse events Follow‐up: end of the trial	—		See comment	157 participants (1 study)	⊕⊝⊝⊝ Very lowa,c	Heiss 2010 reported only adverse events among 157 participants receiving catumaxomab infusion (see Table 2).
Participant satisfaction	—		—	—	—	Not reported
Caregiver satisfaction	—		—	—	—	Not reported
Cost‐effectiveness	—		—	—	—	Not reported
*The basis for the assumed risk (e.g. the median control group risk across studies) is provided in footnotes. The corresponding risk (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; QoL: quality of life.
GRADE Working Group grades of evidence High quality: further research is very unlikely to change our confidence in the estimate of effect. Moderate quality: further research is likely to have an important impact on our confidence in the estimate of effect and may change the estimate. Low quality: further research is very likely to have an important impact on our confidence in the estimate of effect and is likely to change the estimate. Very low quality: we are very uncertain about the estimate.

^aDowngraded one level for serious imprecision (small sample sizes).
 ^bDowngraded two levels for serious risk of biases (high risk of performance and detection biases due to lack of blinding and other bias secondary to the potential conflict of interest of trial authors).
 ^cDowngraded two level for serious risk of biases (high risk of performance and detection biases due to lack of blinding, selective reporting bias (see comment), and other bias secondary to the potential conflict of interest of trial authors).

The data provided from the study were not suitable to measure the difference in the effects of interventions between the comparison groups. Therefore, we applied a narrative approach to report the effects of interventions (see Table 1).

Primary outcome

Quality of life

In the second report from the included study (Wimberger 2012, in: Heiss 2010), participants receiving paracentesis alone experienced more rapid deterioration in QoL than those who received paracentesis plus catumaxomab infusion. The median time to first deterioration of QoL ranged from 19 to 26 days in participants receiving paracentesis alone, while it varied from 47 to 49 days among participants receiving paracentesis with intraperitoneal infusion of catumaxomab. However, only QoL scores evaluated at screening were reported. Available information, therefore, was not suitable to analyse the difference of a change in the QoL score between the comparison groups. We were unable to obtain additional data by contacting the authors of the study. Cox proportional hazards models adjusting for the baseline QoL score, country, and type of cancer confirmed that paracentesis plus intraperitoneal infusion of catumaxomab was superior to paracentesis alone in terms of prolonging the time to first deterioration of QoL across all domains measured (Table 3; Wimberger 2012, in: Heiss 2010).

2. Cox proportional hazards models for European Organization for Research and Treatment of Cancer Quality of Life Questionnaire‐Core 30 items (EORTC‐QLQ‐C30).

Domain assessed	Relative measures (paracentesis with catumaxomab (n = 160) vs paracentesis alone (n = 85)
	Hazard ratio (95% confidence interval)
Emotional functioning	0.24 (0.14 to 0.42)
Global quality of life	0.17 (0.10 to 0.28)
Fatigue	0.23 (0.13 to 0.39)
Nausea and vomiting	0.08 (0.04 to 0.16)
Pain	0.18 (0.10 to 0.32)
Dyspnoea	0.17 (0.08 to 0.36)
Sleep disturbance	0.14 (0.07 to 0.28)
Appetite loss	011 (0.06 to 0.21)

Adapted from Wimberger 2012, in: Heiss 2010.

The model adjusted for baseline quality of life score, country, and type of primary cancer.

Hazards ratio of less than 1 in favour the paracentesis plus catumaxomab compared to paracentesis alone.

n: number of participants.

Secondary outcomes

Improvement of ascites‐related symptoms

The trial did not report improvement of ascites‐related symptoms.

Adverse events

There was no reported comparison of adverse events between the treatment groups. Heiss 2010 reported adverse events among 157 participants receiving abdominal paracentesis plus catumaxomab infusion (Table 2). The most common severe adverse events were abdominal pain, lymphopenia, and cytokine release‐related symptoms (pyrexia, nausea, and vomiting). There were no cases of catheter‐related adverse events (Heiss 2010).

1. Severe adverse events related to catumaxomab infusion.

Common adverse events among participants receiving catumaxomab (n = 157)	NCI‐CTC Grade 3 or more
	Number (%)
Abdominal pain	15 (9.6)
Lymphopenia	11 (7.0)
Pyrexia	9 (5.7)
Gamma‐glutamyltransferase increased	9 (5.7)
C‐reactive protein increased	7 (4.5)
Fatigue	5 (3.2)
Nausea	5 (3.2)
Ileus	5 (3.2)
Anorexia	5 (3.2)
Vomiting	4 (2.5)
Blood alkaline phosphatase increased	4 (2.5)
Aspartate aminotransferase increased	4 (2.5)
Alanine aminotransferase increased	3 (1.9)
Diarrhoea	3 (1.9)
Hypotension	3 (1.9)

n: number of participants; NCI‐CTC: National Cancer Institute Common Toxicity Criteria.

Adapted from Heiss 2010.

Participant satisfaction

The trial did not report participant satisfaction.

Caregiver satisfaction

The trial did not report caregiver satisfaction.

Cost‐effectiveness

The trial did not report cost‐effectiveness.

Discussion

We carried out this systematic review of the literature and included only one RCT reported in two reports (Heiss 2010). However, there were not enough quantitative data to conduct a meta‐analysis. Therefore, we reported the available outcomes narratively.

Summary of main results

See Table 1; Table 2.

We found limited evidence from one small study. The included study compared a combination of abdominal paracentesis and intraperitoneal infusion of catumaxomab versus abdominal paracentesis alone for women with malignant ascites. This trial showed that combined abdominal paracentesis and catumaxomab infusion was superior to paracentesis alone in term of lengthening the time to first deterioration of QoL. The common severe adverse events reported among participants receiving abdominal paracentesis and catumaxomab infusion were abdominal pain, lymphopenia, and cytokine release‐related symptoms. Adverse events among participants receiving abdominal paracentesis alone were not reported. The trial did not report improvement of ascites‐related symptoms, satisfaction of participants and caregiver(s), and cost‐effectiveness.

Overall completeness and applicability of evidence

Malignant ascites is associated with a poor prognosis and leads to poor QoL among terminally ill people with cancer. The major goal of treatment for malignant ascites is to alleviate the symptoms that cause patient discomfort. Therefore, the primary outcome of this review was the impact of the intervention on QoL. Secondary outcomes were the improvement of symptoms, satisfaction of participants and caregiver(s), and cost‐effectiveness.

This review aimed to assess various comparisons of the intervention for managing malignant ascites (see Types of interventions). However, we found only one small RCT that met the inclusion criteria. The included study compared combined abdominal paracentesis and intraperitoneal infusion of catumaxomab versus abdominal paracentesis alone. We also identified one ongoing study (NCT04032600). The following findings have to be kept in mind when considering the completeness and applicability of the evidence presented in this review.

• The study reported only data regarding the time to first deterioration of QoL and severe adverse events (Heiss 2010). However, data on QoL provided by one of the reports (Wimberger 2012, in: Heiss 2010) was not suitable to analyse the relative measure of the effect of the intervention. In addition, there was no reported comparison of adverse event between the treatment groups (Heiss 2010). We were unable to obtain additional data by contacting the trial authors.

• Neither of the two reports of included study provided data regarding the improvement of ascites‐related symptoms, satisfaction of participants and caregiver(s), and cost‐effectiveness, which are considered relevant outcomes for this clinical issue.

• Although combined abdominal paracentesis and intraperitoneal infusion of catumaxomab may delay the deterioration of QoL among malignant ascites in adults compared to abdominal paracentesis alone, catumaxomab is currently not available as it has been withdrawn from the market due to the insolvency of the manufacturer of the drug substance.

• There is one ongoing trial identified which is comparing the safety of full paracentesis versus partial paracentesis among women with histologically confirmed ovarian cancer, peritoneal cancer, or fallopian tube cancer who experienced symptomatic malignant ascites. The recruitment of participants started in August 2017 (NCT04032600, Characteristics of ongoing studies).

At present, the available body of evidence that addresses the question of this review, therefore, is insufficient to produce a conclusive result.

Quality of the evidence

We planned to assess the quality of the of evidence of seven relevant outcomes including QoL, an improvement of ascites‐related symptom, rate of severe adverse events, participant satisfaction, caregiver satisfaction, and cost‐effectiveness. However, the included study reported only time to the first deterioration of QoL and severe adverse events. There was no information on improvement of ascites‐related symptoms, rate of severe adverse events, satisfaction of participants and caregiver satisfaction, and cost‐effectiveness (Heiss 2010).

Based on the concerns regarding the serious risk of bias due to lack of blinding, small sample size of the included RCT, and the potential conflict of interest of trial authors, we downgraded the evidence to very low‐certainty for the time to the first deterioration of QoL. We also downgraded the evidence to very low‐certainty for the severe adverse event because of concerns regarding the serious risk of bias and small sample size (see Table 1).

Potential biases in the review process

There were some potential biases of note in the review process. First, with the assistance from the Information Specialist, Cochrane Gynaecological, Neuro‐oncology & Orphan Cancer Group, we made every attempt to find global studies including a thorough search of the grey literature, conference proceedings, and ongoing trials. However, the greatest threat to the validity of the review is likely to be publication bias (i.e. studies that did not find the treatment to have been effective may not have been published). We were unable to assess this possibility as we found only one included study that met the inclusion criteria. Second, available information from either report of the included study was not suitable to measure the difference in the effects of interventions. Therefore, we applied the narrative review approach to report the effects of interventions.

None of the review authors had any links to drug companies or financial interest in the prescription of the drug under assessment, nor were they involved in the conduct of the included study. Thus, there were no issues associated with bias secondary to conflicts of interest in this review.

Agreements and disagreements with other studies or reviews

Becker 2006 conducted a systematic review to assess the intervention for managing malignant ascites in people with various types of primary cancer regardless of the design of the included studies. The review concluded that there was no convincing evidence indicating the effective intervention for alleviating the symptoms of malignant ascites, although paracentesis, diuretics, and shunting are commonly used procedures.

Malignant ascites develops as a consequence of peritoneal carcinomatosis, elevated portal venous pressure, and lymphatic invasion (Tamsma 2007). In view of the difference in the pathophysiology of malignant ascites across the different types of cancer, the appropriate management of drainage for malignant ascites focusing on a specific type of cancer is required (Harding 2012).

The original 2010 review was undertaken to assess the effective intervention for drainage of malignant ascites among women with gynaecological cancer, and identified no relevant studies (Keen 2010). In this updated review, we found one included RCT (two reports) that met the review inclusion criteria (Heiss 2010). The findings of this updated review were of very low certainty of evidence which thus precluded the ability to draw any convincing conclusions. In addition, although abdominal paracentesis combined with intraperitoneal infusion of catumaxomab may delay the deterioration of QoL compared to abdominal paracentesis alone, catumaxomab is currently not available which, therefore, limits the applicability of the results of this updated review.

Authors' conclusions

Implications for practice.

Currently, there is not enough evidence to make recommendations regarding the management of drainage for malignant ascites among women with gynaecological cancer. Although combined abdominal paracentesis and intraperitoneal infusion of catumaxomab may delay the deterioration of QoL among adults with malignant ascites compared to abdominal paracentesis alone, the certainty of this evidence is very low. In addition, catumaxomab is currently unavailable.

Implications for research.

This review indicates that the research assessing appropriate management of drainage for malignant ascites among women with gynaecological cancer remains hugely under‐resourced despite the incidence of malignant ascites in gynaecological cancer is disproportionately high compared to other cancer types (Becker 2006; Mackey 1996; Wilailak 1999).

High‐quality randomised controlled trials are needed to assess the effective intervention for drainage for malignant ascites in women with gynaecological cancer. Particular attention should be given to the comparison of repeat abdominal paracentesis, the most commonly used procedure, versus other advanced drainage procedure (i.e. the insertion of permanent tunnelled catheters and peritoneo‐venous shunts). Adding a pharmacological intervention to a mechanical drainage procedure in order to diminish reaccumulation of ascites also merits attention in future research. Any studies should include an assessment of the effect of intervention on quality of life, cost‐effectiveness, adverse events, and satisfaction of participant and caregiver to adequately provision of information for considering the risks and benefits of the interventions to be used among the terminally ill adults and their caregivers.

What's new

Date	Event	Description
11 December 2019	New search has been performed	One small randomised controlled trial (RCT) identified and included. One ongoing trial identified added to Ongoing studies.
11 December 2019	New citation required but conclusions have not changed	New team has updated the review to include one small RCT but conclusions remain unchanged. Searches updated to November 2019.

Appendices

Appendix 1. CENTRAL search strategy

#1   MeSH descriptor Endometrial Neoplasms explode all trees
 #2   MeSH descriptor Uterine Neoplasms explode all trees
 #3   MeSH descriptor Uterine Cervical Neoplasms explode all trees
 #4   MeSH descriptor Ovarian Neoplasms explode all trees
 #5   MeSH descriptor Vaginal Neoplasms explode all trees
 #6   MeSH descriptor Fallopian Tube Neoplasms explode all trees
 #7   MeSH descriptor Vulvar Neoplasms explode all trees
 #8   MeSH descriptor Choriocarcinoma explode all trees
 #9   (endometr* or uter* or cervi* or ovar* or vagin* or fallopian* or vulva* or gynae* or gyne*)    near/5 (cancer* or neoplas* or carcinom* or malignan* or tumor* or tumour*)
 #10  (#1 OR #2 OR #3 OR #4 OR #5 OR #6 OR #7 OR #8 OR #9)
 #11  MeSH descriptor Ascites explode all trees
 #12  MeSH descriptor Ascitic Fluid explode all trees
 #13  peritone* near/5 effusion*
 #14  peritone* near/5 fluid
 #15  hydroperiton*
 #16  ascites
 #17  ascitic fluid
 #18  (#11 OR #12 OR #13 OR #14 OR #15 OR #16 OR #17)
 #19  malignan*
 #20  (#18 AND #19)
 #21  (#10 AND #20)

Appendix 2. MEDLINE Ovid search strategy

1  exp Endometrial Neoplasms/
 2  exp Uterine Neoplasms/
 3  exp Uterine Cervical Neoplasms/
 4  exp Ovarian Neoplasms/
 5  exp Vaginal Neoplasms/
 6  exp Fallopian Tube Neoplasms/
 7  exp Vulvar Neoplasms/
 8  exp Choriocarcinoma/
 9  ((endometr* or uter* or cervi* or ovar* or vagin* or fallopian or vulva* or gynae* or gyne*) adj5 (cancer* or neoplas* or carcinom* or malignan* or tumor* or tumour*)).mp.
 10 1 or 2 or 3 or 4 or 5 or 6 or 7 or 8 or 9
 11 exp Ascites/
 12 exp Ascitic Fluid/
 13 (peritone* adj5 effusion*).mp.
 14 (peritone* adj5 fluid).mp.
 15 hydroperiton*.mp.
 16 ascites.mp.
 17 ascitic fluid.mp.
 18 11 or 12 or 13 or 14 or 15 or 16 or 17
 19 malignan*.mp.
 20 18 and 19
 21 10 and 20
 22 Animals/
 23 Humans/
 24 22 not (22 and 23)
 25 21 not 24

key: mp=title, original title, abstract, name of substance word, subject heading word

Appendix 3. Embase Ovid search strategy

1  exp Endometrium Tumor/
 2  exp Uterus Cancer/
 3  exp Uterine Cervix Tumor/
 4  exp Ovary Tumor/
 5  exp Vagina Tumor/
 6  exp Uterine Tube Tumor/
 7  exp Vulva Tumor/
 8  exp Choriocarcinoma/
 9  ((endometr* or uter* or cervi* or ovar* or vagin* or fallopian or vulva* or gynae* or gyne*) adj5 (cancer* or neoplas* or carcinom* or malignan* or tumor* or tumour*)).mp.
 10 1 or 2 or 3 or 4 or 5 or 6 or 7 or 8 or 9
 11 exp Ascites/
 12 exp Ascites Tumor/
 13 (peritone* adj5 effusion*).mp.
 14 (peritone* adj5 fluid).mp.
 15 hydroperiton*.mp.
 16 ascites.mp.
 17 ascitic fluid.mp.
 18 11 or 12 or 13 or 14 or 15 or 16 or 17
 19 malignan*.mp.
 20 18 and 19
 21 10 and 20
 22 exp Animal/
 23 Human/
 24 22 not (22 and 23)
 25 21 not 24

key: mp=title, abstract, subject headings, heading word, drug trade name, original title, device manufacturer, drug manufacturer name

Appendix 4. Risk of bias' assessment

Assessment of risk of bias was based on Chapter 8 of Higgins 2017 .

• Random sequence generation.

  • Low risk of bias, e.g. participants assigned to treatments on the basis of a computer‐generated random sequence or a table of random numbers.

  • High risk of bias, e.g. participants assigned to treatments on the basis of date of birth, clinic ID number or surname, or no attempt to randomise participants.

  • Unclear risk of bias, e.g. not reported, information not available.

• Allocation concealment.

  • Low risk of bias, e.g. where the allocation sequence could not be foretold.

  • High risk of bias, e.g. allocation sequence could be foretold by participants, investigators, or treatment providers.

  • Unclear risk of bias, e.g. not reported.

• Blinding of participants and personnel.

  • Low risk of bias if participants and personnel were adequately blinded.

  • High risk of bias if participants were not blinded to the intervention that the participant received.

  • Unclear risk of bias if this was not reported or unclear.

• Blinding of outcomes assessors.

  • Low risk of bias if outcome assessors were adequately blinded.

  • High risk of bias if outcome assessors were not blinded to the intervention that the participant received.

  • Unclear risk of bias if this was not reported or is unclear.

• Incomplete outcome data: we recorded the proportion of participants whose outcomes were not reported at the end of the study. We coded a satisfactory level of loss to assessed follow‐up for each outcome.

  • Low risk of bias, e.g. if less than 20% of participants were lost to assessed follow‐up and reasons for loss to follow‐up were similar in both treatment arms.

  • High risk of bias, e.g. if more than 20% of participants were lost to assessed follow‐up or reasons for loss to follow‐up differed between treatment arms.

  • Unclear risk of bias, e.g. if loss to assessed follow‐up was not reported.

• Selective reporting of outcomes.

  • Low risk of bias, e.g. review reported all outcomes specified in the protocol.

  • High risk of bias, e.g. suspected that outcomes were selectively reported.

  • Unclear risk of bias, e.g. unclear whether outcomes were selectively reported.

• Other bias.

  • Low risk of bias, e.g. the review authors did not suspect any other source of bias and the trial appeared to be methodologically sound.

  • High risk of bias, e.g. the review authors suspected that the trial was prone to an additional bias.

  • Unclear risk of bias, e.g. the review authors were uncertain whether an additional bias may have been present.

Appendix 5. Risk Of Bias In Non‐randomized Studies – of Interventions (ROBINS‐I) assessment tool

Risk of bias judgements in ROBINS‐I: preintervention and at‐intervention domains
Risk judgement		Bias due to confounding		Bias in selection of participants into the study		Bias in classification of interventions
Low		No confounding expected		All participants who would have been eligible for the target trial were included in the study and start of follow‐up and start of intervention coincide for all participants.		Intervention status is well defined and based solely on information collected at the time of intervention.
Moderate		Confounding expected, all known important confounding domains appropriately measured and controlled for and reliability and validity of measurement of important domains were sufficient, such that we do not expect serious residual confounding.		Selection into the study may have been related to intervention and outcome, but the authors used appropriate methods to adjust for the selection bias or start of follow‐up and start of intervention do not coincide for all participants, but 1. the proportion of participants for which this was the case was too low to induce important bias; 2. the authors used appropriate methods to adjust for the selection bias; or 3. the review authors are confident that the rate (hazard) ratio for the effect of intervention remains constant over time.		Intervention status is well defined but some aspects of the assignments of intervention status were determined retrospectively.
Serious		Switches in treatment, cointerventions, or problems with implementation fidelity are apparent and are not adjusted for in the analyses.		Proportions of missing participants differ substantially across interventions or reasons for missingness differ substantially across interventions and missing data were addressed inappropriately in the analysis or the nature of the missing data means that the risk of bias cannot be removed through appropriate analysis.		The methods of outcome assessment were not comparable across intervention groups or the outcome measure was subjective (i.e. likely to be influenced by knowledge of the intervention received by study participants) and was assessed by outcome assessors aware of the intervention received by study participants or error in measuring the outcome was related to intervention status.
Critical		Substantial deviations from the intended intervention are present and are not adjusted for in the analysis.		There were critical differences between interventions in participants with missing data that were not, or could not, be addressed through appropriate analysis.		The methods of outcome assessment were so different that they cannot reasonably be compared across intervention groups.
No information		No information is reported on whether there is deviation from the intended intervention.		No information is reported about missing data or the potential for data to be missing.		No information is reported about the methods of outcome assessment.
Risk of bias judgements in ROBINS‐I: postintervention domains
Judgement	Bias due to deviations from intended intervention		Bias due to missing data		Bias in measurement of outcomes		Bias in selection of the reported result
Low	No bias due to deviation from the intended intervention is expected, e.g. if both the intervention and comparator are implemented over a short time, and subsequent interventions are part of routine medical care, or if the specified comparison relates to initiation of intervention regardless of whether it is continued.		Data were reasonably complete or proportions of, and reasons for, missing participants were similar across intervention groups or analyses that addressed missing data are likely to have removed any risk of bias.		The methods of outcome assessment were comparable across intervention groups and the outcome measure was unlikely to be influenced by knowledge of the intervention received by study participants or the outcome assessors were unaware of the intervention received by participants and any error in measuring the outcome is unrelated to intervention status.		There is clear evidence (usually through examination of a pre‐registered protocol or statistical analysis plan) that all reported results correspond to all intended outcomes, analyses, and sub‐cohorts.
Moderate	Bias due to deviation from the intended intervention is expected, and switches, cointerventions, and some problems with intervention fidelity are appropriately measured and adjusted for in the analyses. Alternatively, most (but not all) deviations from intended intervention reflect the natural course of events after initiation of intervention.		Proportions of missing participants differ across interventions or reasons for missingness differ minimally across interventions and missing data were not addressed in the analysis.		The methods of outcome assessment were comparable across intervention groups and the outcome measure is only minimally influenced by knowledge of the intervention received by study participants and any error in measuring the outcome is only minimally related to intervention status.		The outcome measurements and analyses are consistent with an a priori plan; or are clearly defined and both internally and externally consistent and there is no indication of selection of the reported analysis from among multiple analyses and there is no indication of selection of the cohort or subgroups for analysis and reporting on the basis of the results.
Serious	Switches in treatment, cointerventions, or problems with implementation fidelity are apparent and are not adjusted for in the analyses.		Proportions of missing participants differ substantially across interventions or reasons for missing differ substantially across interventions and missing data were addressed inappropriately in the analysis or the nature of the missing data means that the risk of bias cannot be removed through appropriate analysis.		The methods of outcome assessment were not comparable across intervention groups or the outcome measure was subjective (i.e. likely to be influenced by knowledge of the intervention received by study participants) and was assessed by outcome assessors aware of the intervention received by study participants or error in measuring the outcome was related to intervention status.		Outcome measurements or analyses are internally or externally inconsistent or there is a high risk of selective reporting from among multiple analyses or the cohort or subgroup is selected from a larger study for analysis and appears to be reported on the basis of the results.
Critical	Substantial deviations from the intended intervention are present and are not adjusted for in the analysis.		There were critical differences between interventions in participants with missing data that were not, or could not, be addressed through appropriate analysis.		The methods of outcome assessment were so different that they cannot reasonably be compared across intervention groups.		There is evidence or strong suspicion of selective reporting of results, and the unreported results are likely to be substantially different from the reported results.
No information	No information is reported on whether there is deviation from the intended intervention.		No information is reported about missing data or the potential for data to be missing.		No information is reported about the methods of outcome assessment.		There is too little information to make a judgement (e.g. if only an abstract is available for the study).
Source: Stern 2016.

Characteristics of studies

Characteristics of included studies [ordered by study ID]

Heiss 2010.

Methods	Randomised, multicentre, phase II/III study of women with malignant ascites due to epithelial cell adhesion molecule‐positive cancer. Women were categorised as ovarian cancer and non‐ovarian cancer cases
Participants	Number of women: 258 randomised to 2 groups in a 2:1 ratio; 170 in the paracentesis plus catumaxomab group and 88 in the paracentesis alone group. 129 women with ovarian cancer and 129 women with non‐ovarian cancer Setting: 13 European countries: Austria, France, Germany, the Netherlands, and the UK for Western Europe, and Czech Republic, Estonia, Latvia, Lithuania, Poland, Romania, Russian Federation, and Ukraine for Eastern Europe. Inclusion criteria: women aged ≥ 18 years with histologically confirmed epithelial cancer and EpCAM tumour cells in the ascites fluid; Karnofsky performance status 60; life expectancy > 8 weeks; estimated ascites volume > 1 L by CT scan; ≥ 1 symptomatic paracentesis within 5 weeks as well as an objectively verified, clinical need for a second paracentesis; refractory or resistant to chemotherapy or standard chemotherapy was no longer feasible. Exclusion criteria: exposure to cancer chemotherapy or radiotherapy within the previous 28 days; previous treatment with murine monoclonal antibodies; enteral feeding at study entry or ileus within the previous 30 days and > 70% tumour infiltration of the liver, or portal vein obstruction, or thrombosis.
Interventions	Comparison: paracentesis plus catumaxomab (catumaxomab group) vs paracentesis alone (control group) Intervention: 1 therapeutic paracentesis on day 0 plus catumaxomab by intraperitoneal infusion on day 0 at 10 μg, day 3 at 20 μg, day 7 at 50 μg, and day 10 at 150 µg. Control: 1 therapeutic paracentesis on day 0. In both groups, repuncture was performed if women required relief of ascites symptoms. Investigators had a clear algorithm to determine when a therapeutic paracentesis should be performed; ascites volume > 1 L based on CT scan; symptomatic ascites based on signs and symptoms assessed by the investigator. Follow‐up: 5 follow‐up visits at 8 days and 1, 3, 5, and 7 months (end of study) after the last infusion (catumaxomab group) or therapeutic paracentesis (day 0, control group). In the control group, the end of the study was reached when the participant required the next paracentesis or died, whichever occurred first.
Outcomes	In the first report of this RCT(Heiss 2010), the primary efficacy endpoint was puncture‐free survival which was defined as the time to first need for therapeutic puncture or death after treatment, whichever occurred first. Secondary efficacy parameters included time to next paracentesis, ascites signs and symptoms, and overall survival. In the second report of this study (Wimberger 2012, in: Heiss 2010), outcomes of interest included paracentesis‐free interval, health‐related QoL, and time to first deterioration in health‐related QoL. The QoL was evaluated using the European Organisation for Research and Treatment of Cancer Quality of Life Questionnaire‐Core 30 items (EORTC QLQ‐C30) questionnaire at screening; 1, 3, and 7 months after treatment; and in the case of repuncture on the day of paracentesis.
Notes	Funded by a grant from Fresenius Biotech GmbH. Trial registration number: NCT00836654
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Unclear risk	No statement regarding random sequence generation.
Allocation concealment (selection bias)	High risk	Open‐label study.
Blinding of participants and personnel (performance bias) All outcomes	High risk	Open‐label study.
Blinding of outcome assessment (detection bias) All outcomes	High risk	Open‐label study.
Incomplete outcome data (attrition bias) All outcomes	Low risk	Approximately 15.3% of participants with ovarian cancer in catumaxomab group were excluded from the analysis in the first report due to major protocol deviation. Approximately 5% of participants did not complete ≥ 1 of EORTC QLQ‐C30 items at screening and then were excluded from the analysis of second report.
Selective reporting (reporting bias)	High risk	There was no reported comparison of adverse events in the control group. There were adverse events among participants receiving paracentesis plus intraperitoneal infusion of catumaxomab.
Other bias	High risk	Funded by a grant from Fresenius Biotech GmbH. Several authors had served as consultant and advisory board member for this company.

QoL: quality of life.

Characteristics of excluded studies [ordered by study ID]

Study	Reason for exclusion
Abdel Ghaffar 2014	Wrong study design. Retrospective study evaluating the safety and efficiency of percutaneous implantable peritoneal port in minimally invasive treatment of intractable malignant ascites.
Adam 2004	Wrong study design. Review of treatment options for malignant ascites.
Akinci 2011	Wrong study design. Purpose was to evaluate retrospectively the safety and effectiveness of radiologically placed tunnelled peritoneal catheter in palliation of malignant ascites. No comparisons were made in this study.
Amiel 1984	Wrong study design. Case series evaluating the use of diuretics for malignant ascites.
Appelqvist 1982	Wrong study design. Study reviewed a series of 100 consecutive cases of malignancy treated for ascites with abdominal paracentesis.
Ayantunde 2007	Wrong study design. Retrospective study undertaken to assess the pattern of ascites and prognostic factors.
Becker 2006	Wrong study design. Systematic review of management of malignant ascites.
Becker 2012	Wrong study design. Commentary article of another published study.
Brooks 2006	Wrong study design. Authors reported off‐label semi‐permanent catheter placement in a person for treatment of malignant ascites.
Campioni 1986	Wrong study design. Authors retrospectively reviewed 42 people with malignant ascites treated with a peritoneal venous shunt over a 5‐year period to assess the incidence of surgical and postsurgical complications.
Chung 2008	Wrong study design. Review of treatment of ascites.
Courtney 2008	Wrong study design. Evaluated long‐term use of Pleurx drains.
Decruze 2010	Wrong study design. Evaluated physiological changes during paracentesis for treating malignant ascites among women with ovarian cancer.
Easson 2007	Wrong study design. Study assessed the ability of existing questionnaires to measure symptom change after paracentesis for symptomatic ascites.
Edney 1989	Wrong population. Study compared use of Denver and Leveen shunts for malignant ascites. Women with gynaecological cancers accounted for < 50% of the cases.
Eskander 2012	Wrong study design. Review article.
Fleming 2009	Wrong study design. Systematic review.
Garrison 1986	Wrong study design. Study evaluated ascitic protein concentration and its relationship to survival.
Gough 1993	Wrong patient population. Women with gynaecological cancer accounted for < 50% of the cases.
Gu 2016	Wrong patient population. Aimed to document the efficacy of symptom relief and conduct survival analysis of people with non‐ovarian cancer with malignant ascites who received paracentesis and indwelling catheter drainage.
Harding 2012	Wrong study design. Authors assessed the safety of free drainage and the feasibility and cost‐effectiveness of day case paracentesis.
Husain 2010	Wrong study design. Study assessed the symptoms of people with malignant ascites.
Kaya 2019	Wrong patient population. Women with gynaecological cancers accounted for < 50% of the cases.
Leblanc 2010	Wrong study design. Short review for the incidence, pathophysiology, and classification of malignant ascites.
Lungren 2013	Wrong study design. Study assessed the success and safety of tunnelled peritoneal drainage catheters for the management of ascites refractory to medical management. No comparisons were made.
Mackey 1996	Wrong study design. Study did not compare different methods of management of drains.
Malik 1991	Wrong study design. Study assessed clinical outcomes and patterns of treatment among 45 people (33 females, 12 males) with cytologically confirmed malignant ascites.
Morita 2005	Wrong interventions. Multicentre, prospective, observational study of 226 consecutive terminally ill people with abdominal malignancies. Participants were classified into 2 groups: the hydration group (59 people) who received 1 L or more of artificial hydration per day, 1 and 3 weeks before death, and the non‐hydration group (167 people).
O'Neill 2018	Wrong study design. Report reviews the practice and safety of outpatient paracentesis.
Ozkan 2007	Wrong study design. Authors reported their experience with a radiologically placed peritoneal port‐catheter in palliation of 7 people with malignant ascites.
Papacharalabous 2010	Wrong study design. Care report.
Rosenberg 2004	Wrong population. Women with gynaecological cancer accounted for approximately 20% of the cases.
Ross 1989	Wrong study design. Study assessed the results and complications of sonographically guided paracentesis among people with malignant ascites.
Soderlund 1986	Wrong study design. Study assessed the results and complication of Denver peritoneo‐venous shunt insertion in 54 consecutive participants for relief of malignant (24 participants) or cirrhotic (30 participants) refractory ascites.
Stratton 1981	Wrong study design. Study examined cytological profile of malignant ascites in people with advanced gynaecological cancers.
Stukan 2015	Wrong study design. Authors assessed the use of a vascular catheter in the management of symptomatic malignant ascites and lymphocyst.
Sugawara 2011	Wrong study design. Authors conducted a retrospective, multicentre study to evaluate the safety and efficacy of radiologically placed peritoneo‐venous shunts in people with malignant refractory ascites.
Walton 2007	Wrong study design. Commentary of another published article.
Wilailak 1999	Wrong study design. Retrospective study evaluating the primary site of cancer among women with malignant ascites.
Winter 1997	Wrong study design. Letter commenting on another published article.
Woopen 2009	Wrong study design. Review article.
Yong 2008	Wrong study design. Letter presenting a case report.
Yu 2013	Wrong study design. Review article.

Characteristics of ongoing studies [ordered by study ID]

NCT04032600.

Trial name or title	Malignant ascites in ovarian cancer: impact of total paracentesis on haemodynamics (ATLANTIS)
Methods	Study type: interventional (clinical trial) Estimated enrolments: 60 participants Allocation method: randomisation Intervention model: parallel assignment Masking: none
Participants	Participant characteristics: women aged ≥ 18 years with histologically confirmed ovarian cancer, peritoneal cancer, or fallopian tube cancer Inclusion criteria: symptomatic (e.g. abdominal pressure, pain, shortness of breath) malignant ascites with clinical indication for paracentesis and sonographic estimate of > 3 L Exclusion criteria: aged < 18 years; missing written informed consent; lack of sufficient knowledge of German or English language; no willingness to consent to the storage or distribution of anonymous disease‐specific data inside the clinical trial; placement inside a state facility due to judicial order; employee status at Charite‐University Medicine of Berlin; chronic kidney insufficiency (defined as serum creatinine levels > 1.2 g/dL at time of admission); active neurological/psychiatric disorder at time of admission; cardiac insufficiency (defined as > NYHA I at time of admission); manifest ileus at time of admission; manifest chronic arterial hypo‐ or hypertension (defined as chronic baseline systolic pressure < 90 mmHg or > 140 mmHg and diastolic pressure < 70 mmHg and > 90 mmHg); active infection; blood clotting disorder (congenital or acquired); thrombocytopenia (platelets < 80,000/nL); active participation in another clinical intervention trial at time of admission; chronic atrial fibrillation on time of admission; status postcardiac pacer implantation; liver cirrhosis; liver metastases
Interventions	Full paracentesis (all ascites is drained) vs partial paracentesis (3 L of ascites are drained, then the drain is clamped and the rest of the ascites is drained on the next day)
Outcomes	Primary outcomes Change of stroke volume: 20 minutes prior to paracentesis, during paracentesis up to the end of paracentesis, a mean of 50 minutes. Stroke volume measured by advanced haemodynamic monitoring Change of mean arterial pressure: 20 minutes prior to paracentesis, during paracentesis up to the end of paracentesis, a mean of 50 minutes. Mean arterial pressure measured by advanced haemodynamic monitoring Secondary outcomes Incidence of adverse effects in the postparacentesis period: hypotension and impairment of kidney function Incidence of symptoms in the postparacentesis period: hypotension and impairment of kidney function Change of stroke volume: 2 hours after the finish of the paracentesis, measured by advanced haemodynamic monitoring Change of mean arterial pressure: 2 hours after finish of the paracentesis, measured by advanced haemodynamic monitoring Change of laboratory values: measurement in plasma (blood count, creatinine, aspartate transaminase, alanine transaminase, urea, aldosterone, renin, sodium, potassium, albumin, C‐reactive protein, leukocytes) Urine excretion Quality of life pre‐ and postparacentesis using the EuroQuol‐Group 5 Dimensions ‐ Visual Analogue Scale (EQ‐5D VAS) questionnaire, the questionnaire directs to characterise the quality of life by mobility, activities, pain, and fear and subjective scale between 1 (worst) and 100 (best) Ascites‐related symptoms pre‐ and postparacentesis using and Functional Assessment of Chronic Illness Therapy‐Ascites Index (FACIT‐AI) questionnaire, the questionnaire directs to characterise the symptoms associated to ascites regarding 13 factors, e.g. appetite, sleep, activities, symptoms, emotional distress by 0 to 4 points. The maximum score is 52 (best result: 13 × 4), 0 is the minimum score (worst result 13 × 0) Measurement of exact drainage volume in millilitre Vascular endothelial growth factor level in ascites Change of venous return and mean systemic filling pressure, measured by advanced haemodynamic monitoring
Starting date	1 August 2017; first posted: 25 July 2019; last update posted: 25 July 2019
Contact information	Klaus Pietzner, MD Email: klaus.pietzner@charite.de Location: Charite‐University Medicine of Berlin, Department of Gynecology‐Campus Virchow Klinikum, Berlin, Germany
Notes	Recruitment status: recruiting

NYHA: New York Heart Association.

Differences between protocol and review

The following sections were updated or revised (or both).

Objectives

In the original 2010 review (Keen 2010), the objectives were "To evaluate the benefit and harms of different practices in the management of drains for malignant ascites in the care of women with advanced or recurrent gynaecological cancer. The review aimed to evaluate the evidence regarding the following questions; How long should the drain stay in place? Should the volume of fluid drained be replaced intravenously? Should the drain be clamped to regulate the drainage of fluid? Should any particular vital observations be regularly recorded?"

In this updated and revised review, the objectives were "To evaluate the effectiveness and adverse events of different interventions for the management of malignant ascites drainage in the palliative care of women with gynaecological cancer."

Types of studies

In the original 2010 review (Keen 2010), the authors planned to include quasi‐randomised trials, non‐randomised trials, controlled before‐and‐after studies, and interrupted time series. Case‐control studies and case series of fewer than 30 participants were excluded.

In this updated and revised review, we planned to include individual randomised controlled trials (RCTs). If no individual RCTs are identified, we planned to include non‐randomised studies (NRS) with a parallel comparison. We did not restrict the number of participants in the studies to be included (see Types of studies).

Types of participants

We additionally stated that If studies included women with non‐gynaecological cancer (i.e. pancreatic cancer, colon cancer, gastric cancer, or hepatobiliary cancer), we planned to extract data specifically for women with gynaecological cancers or planned to contact the trial authors to retrieve data related to participants with gynaecological cancer. If this is not possible, we planned to include the study only if at least 50% of participants were diagnosed with gynaecologic cancer (see Types of participants).

Types of interventions

In the original 2010 review (Keen 2010), the following comparisons were planned with regard to the management of multiple paracentesis (repeated drainage of fluid): shorter versus longer length of time for drain remaining in place; intravenous replacement of fluid versus no intravenous replacement of fluid; clamping of drain versus no clamping of drain; recording of vital observations versus no recording of vital observations; and multifactorial interventions (which include any combination of any of the above interventions) versus usual practice.

In this updated and revised review, we planned to include any studies that attempted to compare the following comparisons: drainage techniques (i.e. repeated abdominal paracentesis; insertion of permanent tunnelled catheters; or peritoneo‐venous shunts) versus no drainage; drainage technique versus other different techniques; combinations of intervention (i.e. drainage technique and medication) versus no drainage; combinations of interventions versus single intervention; and combinations of interventions versus other combinations of interventions (see Types of interventions).

Types of outcome measures

In the original 2010 review (Keen 2010), the primary outcome was quality of life (QoL). The secondary outcomes included participant satisfaction and adverse events. In attempts to cover a broad range of outcome measures, we revised the secondary outcomes as: improvement of symptom; adverse event; participant satisfaction; caregiver satisfaction; and cost‐effectiveness. We additionally classified the adverse events as: infectious complication (i.e. intra‐abdominal infection, intra‐abdominal collection, peritonitis, sepsis); digestive complication (i.e. bowel perforation, accidental bowel puncture, gut obstruction); cardiovascular and thromboembolic complications (i.e. hypotension, pulmonary embolism, venous thrombosis, disseminated intravascular coagulopathy); catheter‐related complication (i.e. catheter placement failure, catheter occlusion). We also stated that we planned to apply the Common Terminology Criteria for Adverse Events (CTCAE) to determine the severity of adverse events (see Types of outcome measures).

In addition, we planned to present a 'Summary of findings' tables reporting all outcomes listed below in the updated and revised review.

• QoL.

• Improvement of symptom.

• Rate of overall adverse events.

• Participant satisfaction.

• Caregiver satisfaction.

• Cost‐effectiveness.

Search methods for identification of studies

We added the lists of conferences for handsearching (see Searching other resources).

Data collection and analysis

We stated that we planned to display a PRIMA flow diagram to indicate the section process of the studies and planned to use Covidence for study selection.

Assessment of risk of bias in included studies

We stated that we planned to apply the Cochrane 'Risk of Bias' in non‐randomised Studies of Interventions (ROBINS‐I) tool for assessing the risk of bias in non‐randomised studies (see Assessment of risk of bias in included studies; Appendix 5).

Data synthesis

We added the section of 'Main outcomes of 'Summary of findings' table for assessing the certainty of the evidence' to describe the presentation of the overall certainty of the evidence for each outcome according to the GRADE approach (see Data synthesis; Appendix 6).

Subgroup analysis and investigation of heterogeneity

There was no plan for subgroup analyses in the original 2010 review (Keen 2010). In this updated and revised review, we planned to carry out subgroup analyses to assess the effect of the different types of drainage procedure (see Subgroup analysis and investigation of heterogeneity). However, we did not perform any of the planned subgroup analyses and investigation of heterogeneity because this review included only one trial.

Sensitivity analysis

In the original 2010 review (Keen 2010), the authors intended to perform sensitivity analysis by excluding non‐randomised studies if RCTs had been included, excluding studies at high risk of bias, and using unadjusted results.

In this updated and revised review, we planned to carry out sensitivity analyses by: repeating the analysis excluding RCTs judged at 'high' or 'unclear' risk of bias for allocation concealment (in case of RCTs available); repeating the analysis excluding NRS judged to be 'high risk' of bias according to the Cochrane ROBINS‐I tool (in case of no RCTs available); repeating the analysis excluding studies that reported unadjusted measures; repeating the analysis excluding studies with more than 20% missing data at the time points at which outcomes were reported; and repeating the analysis excluding unpublished studies (see Sensitivity analysis). However, we were unable to conduct sensitivity analysis because this review included only one trial.

Authorship

The previous authors (Alison Keen, Debbie Fitzgerald, Andrew Bryant and Heather Dickinson) were no longer able to update this review so a new team was convened.

Contributions of authors

CK: coordinated, developed, and completed the review.
 SR: developed and completed the review.
 NJ: developed and completed the review.
 PL: edited and had an advisory role.
 JS: edited and had an advisory role.
 All authors approved the final version of the review.

Sources of support

Internal sources

• Department of Obstetrics and Gynaecology, Faculty of Medicine, Khon Kaen University, Thailand.

• Department of Epidemiology and Biostatistics, Faculty of Public Health, Khon Kaen University, Thailand.

• Department of Obstetrics and Gynaecology, Faculty of Medicine, Chiang Mai University, Thailand.

• Cochrane Thailand, Thailand.

External sources

• Long‐term Institutional Development HUBs (LID‐HUBs), the Human Reproduction Programme (HRP) Alliance for Research Capacity Strengthening, Department of Reproductive Health and Research, World Health Organization, Geneva, Switzerland.

Declarations of interest

CK: none known.
 SR: none known.
 NJ: none known.
 PL: none known.
 JS: none known.

New search for studies and content updated (no change to conclusions)