Systematic review of quality of life following pleurectomy decortication and extrapleural pneumonectomy for malignant pleural mesothelioma

Malignant pleural mesothelioma (MPM) is a fatal malignancy that remains a global concern. The prominent risk factor for MPM is occupational or environmental exposure to asbestos [1‐6]. It is estimated that MPM occurs in 2000 to 3000 people annually in the United States [1], while a study of 19,134 cases from the U.S. National Cancer Database from 2004 to 2013 reported the total number of mesothelioma cases in the U.S. ranged from 1800 to 2000 cases per year [7]. The global incidence of mesothelioma is predicted to increase; it was estimated that 9000 cases will occur Western Europe in 2018 [8], with peak incidence occurring in 2025 in Japan [9], leading to hundreds of thousands of deaths. Several population-based studies on MPM patients have reported a median survival of 8–11 months [10, 11], 8 months [12], 9.5 months [13], 9.2 months [13], and 10.7 months [13] under various treatments. Current treatment options include combinations of surgical resection, radiation, and chemotherapy, with chemotherapy administered as neoadjuvant and adjuvant therapy [1, 14‐16]. A previous analysis of 14,228 MPM patients using the Surveillance, Epidemiology, and End Results (SEER) database reported that surgical treatment was an independent determinant of extended survival [17]. An additional SEER analysis of 5937 MPM patients from 1990 to 2004 reported that surgical treatment was performed in 22% of cases [18].

Although the role of surgery in MPM patients remains controversial [1, 7], extrapleural pneumonectomy (EPP) and pleurectomy (P/D) are the most common surgical procedures performed. Our group has previously shown that differences in survival between the two procedures are modest, but favor P/D in both short- and long-term survival [19]. A separate meta-analysis reported that EPP patients had significantly higher 30 day mortality compared to P/D patients as well as higher postoperative complications [20]. Given the mortality and morbidity associated with surgical resection for MPM, it is important to determine if there is a difference in QoL following EPP and P/D in order to inform patients and guide treatment choices. Here we build on previous work [21] and compare published QoL results for patients undergoing EPP or P/D for MPM at 6 months following surgery.

The initial search yielded 96 potential publications. All abstracts were reviewed and 78 were excluded because the publication was a case report (n = 2), review/commentary (n = 25), did not include mesothelioma cases (n = 25) or quality of life data (n = 14), or reported on cases not treated with surgery (n = 12). The 18 articles remaining were reviewed in detail, and 1 was further excluded because the text was not in English [28]. This left 17 articles and 14 distinct datasets for a total of 659 pleural mesothelioma patients with QoL information (Fig. 1; Table 1). Of the 14 datasets, there were 2 RCTs, one comparing EPP to no surgery and the other comparing partial pleurectomy to talc pleurodesis. Data from these two RCTs were not comparing EPP and P/D, and only data from one arm of each RCT was used. The remaining data were from observational studies, both prospective (n = 10) and retrospective (n = 2). While most studies resulted in a good quality rating from the NIH’s quality assessment tool (Table 2), scrutiny of the included studies indicated the risk of bias. The majority of studies did not measure or account for the impact of confounders influencing surgical procedure choice and QoL outcomes. Additionally most studies did not provide a justification for the sample size used and the resulting sample sizes were small. All EPP studies had less than 50 patients per sample, while only 4/8 P/D articles had more than 50 cases, while the EPP vs P/D studies had less than 25 patients or 40 patients in each group (Table 1). These results should be interpreted in light of the risk of bias stemming from the observational studies included here.

This review of the current literature on QoL after surgical resection for MPM suggests that symptoms, lung function parameters, and physical and social functions remain compromised for 6 months following surgery. However, from the limited and low quality data available comparing the two surgical procedures, it would appear that P/D patients fared better than EPP patients across all QoL measures. This result likely reflects the increased morbidity of EPP compared to P/D. These conclusions, however, need to be viewed as preliminary given the lack of published RCTs and the heavy reliance on observational studies for this review.

As most studies in the literature are retrospective, it is possible that patients in each cohort were treated with the appropriate surgical procedure that considered their clinical characteristic and/or comorbidities, with these same factors potentially independently impacting QoL. This type of selection bias would be accounted for in a RCT comparing the two procedures. In one of the two studies comparing the two procedures in an observational retrospective design [45], lung function was comparable at baseline, but was significantly worse at 6 months among EPP patients as compared to P/D patients. In the other [44], P/D and EPP patients were very similar at baseline, but the physical, social functioning and global health scales were worse after EPP in comparison to P/D. Therefore, even when one considers how selection bias might affect the results of retrospective studies, the conclusions may still show some advantages for P/D. Comorbidities that impact choice of surgical resection for an MPM patient generally associate sicker patients with P/D (as they would not likely tolerate EPP) and any impact of these comorbidities on QoL should favor EPP. With regard to extent of disease, critics of P/D may suggest that patients undergoing EPP have higher tumor burden and this might impact QoL. However, most studies in the review stratified by stage or otherwise excluded patients with advanced disease. The reviewed literature clearly suggests that there is a delicate balance between maximum prolongation of survival and preservation of quality of life in mesothelioma surgical patients, and that the relative priorities of each patient should be considered when deciding on the extent of surgery. Patients should also be made aware that it is unrealistic to expect this extensive surgery to improve QoL, but rather just preserve the preoperative values.

This review highlights serval gaps and limitations in the existing literature and identifies the low quality of available evidence to compare QoL after P/D and EPP. The number of datasets that included QoL measures was relatively small (14 datasets) especially given the extensive literature on MPM surgery. There was only data from one arm of two RCTs, with the remaining data from observational studies. Despite high scores on the NIH Quality Assessment Tool (Table 2), low quality evidence exists to perform the QoL comparison, consistently stemming from low sample size, lack of randomization, and the presence of confounders. Each dataset involved a small number of patients, from 12 to a maximum of 114, which is a key limitation in data stemming from predominately observational study designs. There was also variability in the years of data collection and the length of the studies, which may make it difficult to compare the effects of surgical procedures that could have changed in technicality and invasiveness over time. Similarly this long time span introduces confounders which influence QoL and limit the ability to perform a comparison.

The instruments used to measure QoL were also highly variable and often non-comparable with each other, thus making it difficult to quantify the effect of each surgical approach on QoL. Another source of variability was that QoL measurement was often performed at baseline and then after surgery at different time points, from 1 to 6 months, and occasionally at 1 year. The use of different QoL measures at different times limited the amount of studies included in comparisons. Other treatments, such as chemotherapy and radiation, were often administered, but their effect on QoL was not accounted for in the publications. Additionally these adjuvant treatments varied by study, sample size, and administration (pre-op, post-op, both, neither), which accounts for variability in QoL determinations. Patients included in the QoL studies were very heterogeneous in age, stage, and comorbidities. While most radical resections are performed via thoracotomy, it is also possible that Video-Assisted Thoracoscopy (VATS) approaches were used, and while this may have had a differential impact on QoL, these details were missing in the publications. There were more QoL data on patients who underwent P/D than EPP, and it is possible that if more EPP patients were included, the results might differ. Furthermore, whenever QoL questionnaires are used, one must consider that the subset of patients who respond may exclude the most ill (and lowest QoL) patients, who are unable to respond, or those with better performance status who prefer to continue with their daily activities rather than remain involved in clinical studies.

Accounting for the net direction of these biases would allow for a more accurate quantification of change in QoL after MPM surgery. Statistical aggregation of the individual data was not performed due to the heterogeneity across studies and the lack of abundant and comparable data. Future studies on MPM treatment and outcomes should include QoL measurements acquired at baseline and multiple time intervals, stratified according to treatment, including multimodal therapies. Since no RCTs compared EPP and P/D, data was gathered from observational studies, or in two instances, one arm of an RCT. Further RCTs are needed to directly compare EPP and P/D; the ongoing MARS-2 trial may be able to add unbiased information on the role of P/D on QoL of mesothelioma patients [46]. QoL results are needed to inform patients and treating clinicians to guide treatment choices in MPM.

This literature review shows that there is a limited amount of studies on QoL changes after P/D and EPP for MPM, and that these studies use different methodologies for inclusion criteria, QoL measures, and methods for reporting results. Although the existing evidence is limited and of low quality, it suggests that QoL is better for patients undergoing P/D compared to EPP for an extended period following surgery. QoL outcomes should be strongly considered when choosing type of surgery for MPM, and possible effects on lung function and QoL should be discussed with patients when presenting surgical treatment options.