Manejo del drenaje para la ascitis maligna en el cáncer ginecológico
Resumen
Antecedentes
La ascitis es la acumulación de líquido dentro de la cavidad abdominal. La mayoría de las pacientes con cáncer ovárico avanzado y algunas pacientes con cáncer de endometrio avanzado necesitan drenajes repetidos para la ascitis maligna. Las guías para el asesoramiento de los profesionales a cargo del drenaje de la ascitis habitualmente se diseñan a nivel local y por lo general no están basadas en la evidencia. El manejo de los drenajes que mejora la eficacia y la calidad de este procedimiento es clave para hacer recomendaciones que podrían mejorar la calidad de vida (CdV) de las mujeres en este período crítico de sus vidas.
Objetivos
Evaluar la efectividad y los eventos adversos de diferentes intervenciones para el manejo del drenaje de la ascitis maligna en los cuidados paliativos de las mujeres con cáncer ginecológico.
Métodos de búsqueda
Se realizaron búsquedas en CENTRAL, MEDLINE y EMBASE hasta el 4 de noviembre de 2019. También se verificaron los registros de ensayos clínicos, la literatura gris, los informes de congresos, las listas de referencias de los estudios incluidos y los libros de texto clave para hallar estudios potencialmente relevantes.
Criterios de selección
Se incluyeron ensayos controlados aleatorizados (ECA) de mujeres con ascitis maligna con cáncer ginecológico. Si los estudios también incluyeron mujeres con cáncer no ginecológico, se planificó extraer los datos específicamente para las mujeres con cánceres ginecológicos o solicitar los datos a los autores de los ensayos. Si esto no fue posible, se planificó incluir el estudio solo si al menos el 50% de las participantes tenían un diagnóstico de cáncer ginecológico.
Obtención y análisis de los datos
Dos autores de la revisión de forma independiente seleccionaron los estudios, extrajeron los datos, evaluaron la calidad de los estudios incluidos, compararon los resultados y evaluaron la certeza de la evidencia mediante la metodología Cochrane.
Resultados principales
La revisión original de 2010 no se identificaron estudios relevantes. Esta revisión actualizada incluyó un ECA con 245 participantes que comparó la paracentesis abdominal y la infusión intraperitoneal de catumaxomab versus la paracentesis abdominal sola. El estudio tuvo alto riesgo de sesgo en casi todos los dominios. Los datos no fueron adecuados para el análisis. La mediana del tiempo hasta el primer deterioro de la CdV varió de 19 a 26 días en las participantes que recibieron paracentesis sola, en comparación con 47 a 49 días entre las participantes que recibieron paracentesis con infusión de catumaxomab (evidencia de muy baja certeza). Los eventos adversos solo se informaron en las participantes que recibieron la infusión de catumaxomab. Los eventos adversos graves más frecuentes fueron dolor abdominal y linfopenia (157 participantes; evidencia de muy baja certeza). No hubo datos sobre la mejoría de los síntomas, la satisfacción de las participantes y los cuidadores y la relación entre coste y efectividad.
Conclusiones de los autores
Actualmente no hay evidencia suficiente para recomendar el manejo más apropiado del drenaje para la ascitis maligna en las mujeres con cáncer ginecológico, ya que solo hubo evidencia de muy baja certeza a partir de un ECA pequeño con alto riesgo general de sesgo.
PICO
Resumen en términos sencillos
Métodos para el drenaje del líquido que contiene células cancerosas acumulado en el abdomen en mujeres con cáncer ginecológico
Antecedentes
La ascitis maligna es la acumulación de líquido dentro de la cavidad abdominal causada por el cáncer subyacente. Las mujeres con cáncer de ovario avanzado y algunas mujeres con cáncer uterino avanzado (también conocido como cáncer de la matriz) a menudo necesitan un drenaje para la ascitis maligna para aliviar el malestar. Las guías para los profesionales a cargo del drenaje de la ascitis habitualmente se producen a nivel local y por lo general se basan en la experiencia de los médicos.
Características de los estudios
Se buscaron estudios hasta noviembre de 2019 que compararon diferentes maneras para manejar el drenaje del líquido recolectado en el abdomen de las mujeres con cáncer ginecológico (cáncer que comienza en los órganos reproductivos de la mujer).
Resultados clave y calidad de la evidencia
La revisión original de 2010 no encontró estudios relevantes. Esta revisión actualizada incluyó un ensayo controlado aleatorizado (ECA: un tipo de estudio en el que las personas se asignan al azar a recibir diferentes tratamientos), que incluyó 245 mujeres y comparó el drenaje combinado con catumaxomab (un fármaco utilizado para tratar la ascitis maligna) versus el drenaje solo. Sin embargo, los resultados no fueron suficientes para evaluar la diferencia entre estos tratamientos. Aunque las mujeres que recibieron drenaje combinado con catumaxomab tuvieron una mejor calidad de vida (el bienestar general de una persona) durante más tiempo en comparación con el drenaje solo, existe muy poca certeza en esta evidencia debido al pequeño número de participantes y ensayos. Hubo algunos efectos secundarios en el grupo de drenaje más catumaxomab (p.ej. dolor, recuento bajo de glóbulos blancos), pero no se informaron de manera adecuada. En la actualidad no hay datos suficientes sobre la mejor manera de manejar el drenaje para la ascitis maligna en mujeres con cáncer ginecológico.
La evidencia se actualizó hasta el 4 de noviembre de 2019.
Authors' conclusions
Summary of findings
| 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 | No of participants | Certainty of the evidence | 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) | ⊕⊝⊝⊝ | Heiss 2010 reported only adverse events among 157 participants receiving catumaxomab infusion (see Table 1). | |
| 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 | ||||||
| aDowngraded one level for serious imprecision (small sample sizes). | ||||||
| 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. | |
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.
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).
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
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 summary of findings Table for the main comparison).
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 2; Wimberger 2012, in: Heiss 2010).
| 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 1). 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).
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 summary of findings Table for the main comparison; Table 1.
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 summary of findings Table for the main comparison).
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.
Study flow diagram.
Risk of bias summary: review authors' judgements about each risk of bias item for each included study.
| 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 | No of participants | Certainty of the evidence | 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) | ⊕⊝⊝⊝ | Heiss 2010 reported only adverse events among 157 participants receiving catumaxomab infusion (see Table 1). | |
| 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 | ||||||
| aDowngraded one level for serious imprecision (small sample sizes). | ||||||
| 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. | |
| 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. | |
