Article Text
PDF
Abstract
Background Malignant pleural effusion (MPE) results in breathlessness and impairment of health-related quality of life (HRQOL). This study reviews the existing literature on HRQOL following invasive interventions in MPE.
Methods Five electronic databases were systematically searched and assessed three times during the review process and last completed on 15 June 2018. We included all studies evaluating HRQOL outcomes for the following interventions: therapeutic thoracocentesis, talc slurry (TS) pleurodesis, indwelling pleural catheter (IPC) insertion and thoracoscopic talc poudrage (TTP) pleurodesis. Meta-analysis was not performed due to substantial heterogeneity in the published data.
Results 17 studies were included in the review reporting HRQOL outcomes in 2515 patients. TTP, TS and IPC were associated with modest but inconsistent improvements in HRQOL up to 12 weeks. No intervention was significantly different from another in HRQOL outcomes at any time point. The attrition to follow-up was 48.3% (664/1374) at 3 months. The overall quality of studies was inadequate.
Conclusion TTP, TS and IPC seem to improve HRQOL in MPE over 4–12 weeks, but there are insufficient longer term data due to high attrition rates. Evidence on the most effective treatment strategy is limited by the small number of randomised or comparative studies.
Trial registration number CRD42016051003.
- pleural effusion
- quality of life
- talc pleurodesis
- thoracoscopy
Statistics from Altmetric.com
Introduction
Health is ‘a state of complete physical, mental, and social well-being not merely the absence of disease or infirmity’.1 Although this definition was established in 1946, health-related quality of life (HRQOL) is a relatively new notion that has evolved since the 1980s. It focuses on the effects of illness and treatment, while distinguishing it from aspects of life outside the realm of healthcare. This concept considers several self-reported elements, including symptoms, function in multiple life domains, and general perceptions of life satisfaction and quality.
A malignant pleural effusion (MPE) is a collection of fluid in the pleural space secondary to cancer that causes disabling breathlessness, tiredness and loss of function. The median survival of patients with MPE is 3–12 months,2 with an estimated 50 000 new cases of MPE per year in the UK, translating to 1 new case per 1000 population per year. Research to date in MPE has centred on pleurodesis success, which is often defined radiographically, but there is a poor understanding of the impact of MPE on the lives of patients and their carers.3
Patients and physicians have a variety of choices for the management of MPE, including recurrent therapeutic thoracocentesis, indwelling pleural catheter (IPC) insertion with talc pleurodesis, local anaesthetic thoracoscopy or video-assisted thoracoscopic surgery (VATS). The selection of an appropriate intervention relies on multiple patient-specific factors, including non-expandable lung, comorbidities, local expertise and patient/carer preferences. Each of these procedures is likely to impact on patients differently, for example, living with a long-term drain that requires repeated drainage, recovery from an invasive surgical or local anaesthetic procedure, and the burden of repeated healthcare visits. Regardless of the choice, the aim of any interventional procedure in this patient population with limited prognosis is to ultimately improve quality of life. Unfortunately, there is no consensus in the literature as to the ideal procedural approach for MPE. Due to a lack of data to answer this relevant clinical question, we aimed to systematically review the existing literature on the quality of life outcomes for various interventional approaches for patients with MPE.
Methods
The authors devised a systematic review methodology which underwent internal peer review by a specialist in systematic review methodology. This review was registered on the international prospective register of systematic reviews database prior to commencement. The protocol is made available in online appendix 1.
Supplemental material
Data sources and searches
We conducted a comprehensive search of five databases: Cochrane Library (including the Cochrane Central Register of Controlled Trials), Cochrane Databases of Systematic Reviews, Medline (through Ovid interface), Embase and Web of Science. A combination of controlled vocabulary headings and keywords was used, adapted for each database. The full search strategy is available in online appendix 2.
Supplemental material
The electronic search was repeated three times during the review process and last completed on 15 June 2018.
In addition to an electronic search, references in review articles were manually searched. Authors of potentially eligible abstracts were contacted to obtain complete manuscripts or further information (where required). Non-English papers were excluded.
Study selection
Study populations
Published data for all patients with MPE were included (defined as histological or cytological evidence of pleural malignancy or a radiological diagnosis in a patient with histologically proven malignancy elsewhere). Studies including patients with benign pleural effusion were acceptable if the outcomes for MPE were reported separately. If not specifically reported, authors were contacted to obtain the MPE subgroup data. However, if not procurable, the study was excluded.
Types of study included
A scoping search identified a paucity of studies; therefore, prospective and retrospective studies were included. Randomised controlled trials (RCTs), case–control studies, case series of more than 10 patients and observational studies were included. Studies published in the form of conference abstracts were considered if the author was able to provide a completed manuscript or if HRQOL outcomes for MPE were reported in the abstract.
Types of interventions
Invasive interventions considered included thoracocentesis, chest drain and talc slurry (TS) pleurodesis, IPC insertion, thoracoscopic (both general and local anaesthetic) poudrage pleurodesis, as well as thoracoscopy and IPC insertion.
Types of comparison
Due to the paucity of data, those studies that included any other comparator intervention (eg, pleurodesis with an agent other than talc) were included.
Measures of interest
The primary measure of interest was global HRQOL, measured using a validated instrument. Instrument validity was defined as publication of a validation study for general health, cancer or respiratory conditions in a peer-reviewed journal. Studies measuring quality of life as a secondary outcome measure were also included. Papers using novel non-validated scales were excluded from analysis.
Data extraction and quality assessment
One author (PS) performed the electronic search screening of all titles and abstracts. When multiple reports of a study were identified, they were treated as a single study. A data extraction sheet was completed for each paper passing the inclusion and exclusion stage. Two authors (PS and AS) extracted the data from all the full papers identified by PS. Disagreements for all papers were resolved by discussion and consultation with LA to achieve consensus when required.
The two review authors independently assessed the risk of bias in the RCTs using the Cochrane Collaboration’s tool for assessing the risk of bias. This considers seven criteria, including sequence generation, allocation concealment, blinding of participants and personnel, blinding of outcome assessment, incomplete outcome data, selective outcome reporting and ‘other issues’. Risk of bias was not assessed in non-RCTs.
Results
After the original database search and removal of duplicates, 696 abstracts and titles were reviewed. Of these, 641 did not meet the specified inclusion criteria. Of the remaining 55 potentially eligible studies, 41 were excluded (figure 1).
PRISMA flow diagram of evidence synthesis. PRISMA, Preferred Reporting Items for Systematic Reviews and Meta-Analyses.
There was uncertainty regarding the eligibility of four studies which were excluded after discussion between the three reviewers (online appendix 3). During the process of review, seven additional articles were identified by repeating the electronic database search and manually searching reference lists. We summarise the design of 17 studies (table 1). A summary of HRQOL outcomes for each study can be found in online appendix 4. Abbreviations used for quality of life instruments are summarised in table 2. Risk of bias in the RCTs is illustrated in figure 2.
Supplemental material
Supplemental material
Risk of bias assessment for randomised controlled trials.
Summary and characteristics of studies included in the final analysis
Abbreviated names for quality of life questionnaires
Studies markedly varied in design, quality, outcome measures and presentation, with a predominance of case series. Formal data synthesis with a meta-analysis was not performed due to the heterogeneity of data. Due to the nature of the available data, we present each intervention and their effect on HRQOL separately.
Thoracocentesis
There were no studies examining quality of life outcomes in thoracocentesis.
Thoracoscopic talc poudrage
A total of 880 patients across studies received thoracoscopic talc poudrage (TTP). Two of these studies were RCTs comparing VATS-thoracoscopic talc poudrage (VATS-TTP) with chest drain and TS.4 5 Both measured HRQOL as a secondary endpoint.
Reduced fatigue was demonstrated in the VATS-TTP arm (p=0.016), but no significant differences were seen in the other 14 subscales of the European Organisation for Research and Treatment of Cancer Quality of Life Questionnaire-Core 30 (EORTC QLQ-C30) at 30 days postprocedure.4 However only 44.4% (214/482) completed follow-up at this time point. The other smaller RCT (n=60) using the Short Form 36 (SF-36) and WHO quality of life instrument showed no intergroup differences in HRQOL at 30 days postintervention, with no participants lost to follow-up.5
There were four prospective case series included in the analysis.6–9 Two of these examined HRQOL as a primary outcome following VATS-TTP.6 8 In 46 patients, Karnofsky Index (KI) scores improved significantly at discharge postprocedure (preoperative mean KI 62.1±12.2 SD vs postoperative mean KI 71.3±13.2 SD; p=0.014), although the length of hospitalisation postprocedure was not stated.6 Postoperative global health and functional scales did not significantly improve in 45 patients when measured with the EORTC QLQ-C30 save for emotional function (p=0.035).8 Clinical and statistically significant improvement in global health was observed at 3 months postprocedure with a sustained clinical improvement at 6 and 12 months. Clinical but not statistically significant improvements were seen at 12 months in the physical role and emotional function domains, compared with baseline values.8
One retrospective case-matched study compared the quality of life outcomes in VATS-TTP with local anaesthetic thoracoscopic talc poudrage (LA-TTP) (n=462). Although significant improvements favouring LA-TTP were seen in the physical function, global health and dyspnoea domains, the groups were poorly matched with a significantly higher incidence of comorbidity in this group. Over a 6month period of follow-up, no other significant intergroup differences in any of the HRQOL domains were identified.10 LA-TTP has also been modified to include postprocedure IPC insertion to reduce the length of stay. In this small retrospective series (n=30), postprocedure KI scores increased from a mean of 57.9 to 72.1.11
Chest drain and talc pleurodesis
In seven studies, 546 patients underwent chest drain and TS pleurodesis. Five are RCTs,4 5 12–14 two of which compare TS with VATS-TTP and are outlined above.4 5
The Second Therapeutic Intervention in Malignant Effusion Trial (TIME2) study randomised 106 patients to an IPC or TS. Global health (measured with EORTC QLQ-C30) improved in both groups at 6 weeks, but no significant differences were identified up to 6 months. There was a considerable loss to follow-up primarily driven by mortality, with 65.1% completing questionnaires at 6 weeks and only 38.7% at 6 months.12
In the Australasian Malignant Pleural Effusion (AMPLE) RCT, which randomised 144 patients to TS or IPC, improvements in HRQOL were seen following initial treatment in both groups using the 100 mm Visual Analogue Scale (VAS) and EuroQol 5D (EQ-5D) measures, and these were maintained over a 12-month period in 30% of patients who completed follow-up at this time point.14 No significant differences were seen in HRQOL between the groups at all time points of follow-up.
MesoVATS compared TS with VATS-partial pleurectomy (VATS-PP) in 175 patients with malignant mesothelioma.13 No significant differences in EQ-5D scores were seen at 1 and 3 months; however, the VATS-PP group had significantly better scores at 6 months (mean difference 0.08 (95% CI 0.003 to 0.16), p=0.042) and 12 months (mean difference 0.19 (95% CI 0.05 to 0.32), p=0.006). When measured with the EORTC QLQ-C30, no significant difference in global health was seen at any time point. However, at 3 months there were significant improvements in physical function (mean difference 8.2 (95% CI 1.1 to 15.4), p=0.03) and dyspnoea scores (mean difference −12.9 (95% CI −23.0 to −2.7), p=0.01) in the VATS-PP groups, but this was not maintained at 6 and 12 months. A loss to follow-up was particularly apparent in the months leading up to death, with only 65.5% and 41.5% of those randomised completing the EORTC questionnaire at 6 and 12 months, respectively.
Indwelling pleural catheter
In ten studies, 977 patients underwent IPC insertion. Five of these are RCTs, of which TIME2 and AMPLE are outlined above.6 15 An RCT comparing IPC with doxycycline pleurodesis via a chest drain showed improvements in HRQOL measured with the Guyatt Chronic Respiratory Disease Questionnaire in the IPC arm at 30 days (mean improvement 5.2±7.5 SD), and this was maintained in 34.4% (33/96) completing follow-up at 90 days (mean improvement 7.2±7.1).12 Similar scores were seen in the doxycycline pleurodesis group. In the Impact of Agressive Versus Standard Drainage Regimen Using a Long Term Indewlling Pleural Catheter (ASAP) RCT (n=149), all patients received an IPC and were randomised to either ‘aggressive’ daily drainage or ‘standard’ alternate day drainage.16 No significant improvement was seen in the SF-36 general health scores at 12 weeks in the aggressive care group versus the standard group (difference in mean general health score −12.6 (95% CI −29.3 to 4.1)); however, this is a small sample size with 12week HRQOL scores only available in 27 of 118 surviving patients. No significant differences were seen in the other SF-36 domains (including physical function, energy and emotional well-being) or KI scores between the two groups at baseline and 2 and 12 weeks.
The IPC-PLUS RCT compared instillation of TS versus placebo through an IPC in 154 patients and reported better HRQOL scores in those who received talc, with a significant difference in EORTC QLQ-C30 global health score on day 28 (mean difference 9.2 (95% CI 1.1 to 17.4), p=0.03) and day 42 (mean difference 14.7 (95% CI 5.9 to 23.5), p=0.001), although a significant difference was only seen on day 42 using the EQ-5D questionnaire (0.12 (95% CI 0.01 to 0.22), p=0.03).5 Pleural catheters were removed due to drainage cessation in 38% of patients in the talc arm vs 14.5% in the placebo arm.
Three studies offer prospective observational data following IPC insertion, with duration of follow-up ranging from 14 weeks to 1 year.9 11 17 Over a 14-week period, clinical and significant improvements in global HRQOL were seen at 2 weeks (mean improvement in EORTC QLQ-C30 global health score 12.3, p<0.001). In 35.3% (29/82) of patients completing follow-up at 14 weeks, this improvement was maintained (p<0.05).17 In a series examining 60-day outcomes, 28 of 51 patients who completed follow-up at 30 days demonstrated improvement in global health, but this was not statistically significant (mean improvement in EORTC QLQ-C30 global health score 9.21 (95% CI −1.55 to 19.98), p=0.09). A non-significant deterioration in global health was observed at 60 days compared with 30 days, but only 13 of the 28 patients completed follow-up.9 Similarly, global HRQOL measured with the Short Form 6D failed to show an improvement in utility at 1 month post-IPC insertion (mean difference 0.023 (95% CI −0.004 to 0.05), p=0.10). However multivariate analysis suggests that patients undergoing chemo/radiotherapy post-IPC placement and those more short of breath experienced greater improvement in utility from baseline.11
In a small retrospective study,8 IPC was superior to talc pleurodesis in improving quality of life at 7 days (measured with a 100 mm VAS); however, its retrospective approach and heterogeneous pleurodesis group consisting of TS, LA-TTP and VATS-TTP make it very difficult to draw a firm conclusion from this data set.
Attrition to follow-up
Attrition and incomplete follow-up of HRQOL outcomes, primarily through mortality and loss to follow-up, were apparent in most studies. Attrition data are outlined in table 3. Pooling the data (where available), only 75.9% (346/456) of patients enrolled in studies completed 1-month follow-up, with only 48.3% (664/1374) of patients enrolled in studies completing follow-up at 3 months. This estimate should be treated with caution as the inclusion criteria (such as prognosis and performance status) varied between studies.
Summary of attrition rates during study follow-up (where available)
Recurrent pleural intervention
Ten studies provided data on the rate of pleural effusion recurrence requiring intervention, although the duration of follow-up ranged from 1 week to 12 months following initial management (table 4). Pooled rates of pleurodesis failure (defined as the need for repeat pleural intervention on the ipsilateral side) in TS and TTP were similar—42.9% (146/340) and 39.7% (310/779), respectively. Of patients who underwent IPC insertion, 9.2% (50/542) required a repeat pleural intervention. One study reported the use of intrapleural fibrinolytic therapy to treat symptomatic loculation in five patients.8 This was not classified as a repeat pleural intervention.
Recurrent ipsilateral intervention rate
Adverse events
There was considerable variation in the quality of adverse event reporting between studies. Table 5 summarises the most frequent intervention-related complications. Although the lowest complication rate is seen in LA-TTP (12.2%), data were only obtainable from one single-centre retrospective study10 with a high risk of recall bias.
Pooled adverse event rates (excluding pleurodesis failure where applicable)
Conclusions
Patients and clinicians both seek to improve quality of life in selecting an appropriate interventional approach for the management of MPE. Although clinical research in this area often cites an assessment of ‘quality of life’, these studies often extrapolate outcomes from surrogate endpoints instead of using validated HRQOL instruments. Clinicians recognise the importance of improving quality of life in malignant pleural disease, but there is no consensus on the best treatment approach for their patients.18 To our knowledge, this is the first study to systematically review the evidence for HRQOL outcomes following invasive pleural interventions for MPE.
Based on currently available literature, treatment with VATS-TTP, TS and IPC seems to have a positive impact on quality of life in MPE over 4–12 weeks, but there are insufficient longer term follow-up data on whether this is sustained. There are no data to recommend therapeutic thoracentesis as a strategy to improve quality of life. With the current data that are available, we cannot conclude whether one treatment strategy is superior to another.
Only eight of the 17 studies were RCTs and 4 of these are considered very good quality. However, none are powered for HRQOL as a primary outcome. Nine case series provide some useful data, but the lack of a comparator in six of these series prevents a comparison of the different approaches used. The heterogeneity in study design and the use of 11 different measuring tools for quality of life outcomes also prevent a meaningful statistical analysis when pooling the data. The HRQOL instruments used have not been specifically validated in the MPE population, which may explain the inconsistencies when using multiple measures in the same population; for example, the significant changes seen in EQ-5D scores were not apparent with the EORTC QLQ-C30 questionnaire in the MesoVATS trial.14
A considerable number of the identified abstracts focused on endpoints such as pleurodesis success or breathlessness. Although breathlessness may be a key component of HRQOL, there is limited research into factors such as overall well-being, social and functional activities which may be relevant to a patient with advanced cancer. Defining pleurodesis success also varies in the literature, with many studies using radiological outcomes with quoted success rates of 80%.19 If we focus on the best outcome for patients, the need to avoid a further repeat pleural intervention becomes a priority. From our pooled data set, we demonstrated that over a third of patients receiving talc pleurodesis required a repeat pleural intervention irrespective of the method of insufflation. Given that the biology of pleurodesis failure is poorly understood and its incidence easily underestimated by short periods of follow-up in prospective studies, a patient-centred outcome measure of treatment success is of greater importance in future studies of pleurodesis for MPE.
There is also a variable duration of follow-up with outcomes beyond 6 months poorly represented in our study. More work needs to be done to understand the progression of HRQOL in these long-term survivors as it is likely to change over the course of cancer treatment well beyond the initial pleural intervention. This presents a formidable challenge as we have identified a high risk of attrition bias with many patients not completing follow-up due to mortality and loss to follow-up in the months preceding death. In the future, implementation of prognostic tools such as the LENT score20 prior to patient inclusion may help limit mortality-related attrition bias. Other potential strategies include increasing patient and public involvement in study design and questionnaire selection to develop a protocol that is feasible, non-burdensome and maintains patient interest.
Two clinical trials are due to complete recruitment and publish results (table 6). One examines HRQOL as a primary outcome measure, with the other using similar and validated quality of life instruments, which may offer comparable data and further insight into quality of life outcomes for these patients.
Currently recruiting or yet-to-be published trials measuring HRQOL in MPE (correct as of 2 July 2018)
Our study highlights an opportunity for future research to answer an important question for patients with MPE, especially as the targeted therapy for lung cancer and other malignancies is rapidly evolving with improving outcomes. The ideal research strategy should appreciate the impact of comorbid conditions, molecular and genetic cancer typing, and burden of anticancer treatment on quality of life outcomes in MPE. It is therefore vital that future research incorporates a collaborative approach with oncologists and other specialties to better understand HRQOL outcomes in the different phenotypes of this diverse condition. We recommend the development and validation of a bespoke instrument to measure HRQOL in MPE, and that as a research community we should agree on a core outcome set for use in all future MPE studies as per the Core Outcome Measures in Effectiveness Trials initiative. This has the potential to reduce reporting bias, create homogeneity in outcome reporting, guarantee the quality of future systematic reviews and permit meta-analyses in this field.
Until further data are available, rather than a ‘one size fits all’ approach, decision-making to improve quality of life should be individualised based on the clinical picture and patient preference. Each treatment will have a different impact influenced by the expected outcome, patients’ and carers’ personal views, as well as their social and cultural background. It is therefore vital that the description and presentation of the different options allow patients to make informed decisions that suit their individual circumstances.
Acknowledgments
The authors would like to thank Dr Bronwen Connelly for her valuable advice on study design and literature searching.
References
Footnotes
Contributors All authors included in the paper fulfil the criteria for authorship. PS, AS and LA were involved in project conception, drafting and final approval of the manuscript. PS and AS conducted abstract review and data analysis. LA reviewed the methodology, analysis and presentation of the results. PS is the guarantor.
Funding LA and PS have received funding from CareFusion for the OPTIMUM trial (UKCRN 19615), and PS and LA have received funding from Rocket Medical UK for the SINE Study (UKCRN 34053).
Patient consent for publication Not required.
Provenance and peer review Not commissioned; externally peer reviewed.
Data availability statement All data relevant to the study are included in the article or uploaded as supplementary information.