Trends in quality of life reporting for radical cystectomy and urinary diversion over the last four decades: A systematic review of the literature

ABSTRACT Objective: To report the trends in quality of life (QoL) reporting for radical cystectomy (RC) and urinary diversion (UD) over the last four decades, as RC for bladder cancer is associated with significant morbidity and QoL issues. Material and methods: We searched PubMed, Medical Literature Analysis and Retrieval System Online (MEDLINE), Excerpta Medica dataBASE (EMBASE), Cumulative Index to Nursing and Allied Health Literature (CINAHL), and the Cochrane library for published studies from January 1980 to January 2017 in the English language. We divided the published articles into three time periods: period-1 (1980–1997), period-2 (1998–2007) and period-3 (2008–2017). Results: A total of 85 QoL studies (8417 patients) were identified, of which 3347 (39.8%) patients had an ileal conduit (IC), 1078 (12.8%) had a continent UD (CD), 3264 (38.8%) had a neobladder (NB), and in the remaining 728 (8.6%) the type of UD was not specified. Whilst there were 15, 24 and 41 studies in period-1, period-2 and period-3 respectively, two (13%), 20 (83%) and 37 (90%) used a validated QoL tool; and none, six (25%) and 23 (56%) used a urology specific QoL tool during these three time periods. Similarly, the number of prospective studies increased from one (7%) to four (17%) and 14 (34%) in these three time periods. The proportion of reported IC patients reduced from 65% (784 patients) to 36% (899) and 35% (1664) from period-1 to period-3, whereas the proportion of NB patients increased from 4.5% (54) to 44% (1105) and 44% (2105). Over the last few years there have been QoL studies on laparoscopic and robotic IC and NB UDs. Conclusion: Our review suggests an increasing use of validated, bladder cancer-specific questionnaires with UD-specific constructs. Abbreviations: BCI: Bladder Cancer Index; BDI: Beck Depression Inventory; BIS: Body Image Scale; CD: continent urinary diversion; EORTC QLQ-30C: European Organisation for the Research and Treatment of Cancer Quality of Life 30-item core questionnaire; ERAS: enhanced recovery after surgery; FACT(-BL)(-G)(-VCI): Functional Assessment of Cancer Therapy(-Bladder Cancer)(-General)(-Vanderbilt Cystectomy Index); IC: ileal conduit; NB: neobladder; (HR)QoL: (health-related) quality of life; (RA)RC: (robot-assisted) radical cystectomy; SF-36: 36-item short-form health survey; SIP: Sickness Impact Profile; UD: urinary diversion


Introduction
Radical cystectomy (RC) with urinary diversion (UD) is associated with significant morbidity. Once patients recover from this surgery, quality of life (QoL) becomes an important priority having a significant role in their future psychological and emotional well-being [1][2][3]. UD impacts QoL and there are different types of UD to choose from, including ileal conduit (IC) to continent cutaneous UD (CD) and neobladder (NB) [3][4][5][6][7][8].
Measuring QoL can help assess the impact of RC and UD, identify patient preference, help in staff training, and be useful for audit and clinical governance [1]. The choice of UD depends on patient suitability and preference, with a possible surgical bias related to the surgical expertise available in the centre. Whilst enthusiasts favour NB, there is little evidence to support that one UD type is better than another [9][10][11][12]. It seems that for now, the choice of UD should be individualised and based on patient counselling and expectations, with an active but unbiased surgical input. Measuring QoL in these patients has changed from self-designed to nonvalidated and now validated tools, including generic and disease-specific measures. Over time the shift has been to use bladder cancer-specific health-related QoL (HRQoL) tools supplemented by patient-reported outcome measures [8,[13][14][15][16][17][18].
Publication trends reflect clinical practice [19]. Trends in the type of UD offered could help patients in their choice of UD type, improve counselling and allocation of healthcare resources. The QoL aspect seems to be the most important element in UD once patients have recovered from their initial surgery. There is no bibliometric study looking at the publication trends of reporting QoL in UD patients.
We therefore assessed the trend in QoL reporting after RC and UD over the last four decades.

Inclusion criteria
(1) All studies reporting on QoL after UD, irrespective of the type of UD. (2) Studies published in English language over the last four decades.

Exclusion criteria
(1) Animal studies and case reports.
(2) Studies on UD that did not assess QoL.

Search strategy and study selection
We performed a systematic review of the world literature to identify original studies reporting on QoL in UD. It was carried out using Cochrane and Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) methodology. We searched PubMed, Medical Literature Analysis and Retrieval System Online (MEDLINE), Excerpta Medica dataBASE (EMBASE), Cumulative Index to Nursing and Allied Health Literature (CINAHL) and the Cochrane library for published studies from January 1980 to January 2017. We used the following search terms 'urinary diversion', 'quality of life', 'neobladder', 'ileal conduit', 'cutaneous diversion', 'cystectomy', 'health-related quality of life', and 'QoL'. All articles from 1980 to 1997, and only articles directly comparing two or more different UD types from 1998 to 2017 were selected for screening. All full-length articles published in the English language were included in the original search and the two reviewers (K.R., B.K.S.) independently identified all studies that fitted the inclusion criteria ( Figure 1).

Primary outcome measures
(1) Trends of QoL reporting over the last four decades. (2) Number of studies and type of UD done.

Secondary outcome measures
(1) Geographical variation in the reporting of the QoL studies. (2) Journals which published these QoL studies.

Data extraction and analysis
Both reviewers (K.R., B.K.S.) independently identified all studies that appeared to fit the inclusion criteria and any disagreement was resolved with mutual consensus. All data were collected in an Excel spreadsheet and then transferred to a Word document. The study was carried out using the Cochrane and PRISMA methodology. Included studies showed a high level of heterogeneity and bias, and data were not suitable for a meta-analysis, and hence have been presented in a descriptive manner.
The overall proportion of reported IC patients reduced from 65% (784 patients) to 36% (899) and 35% (1664) from period-1 to period-3, whereas the proportion of reported NB patients increased from 4.5% (54) to 44% (1105) and 44% (2105). The reporting of both the UD types was broadly similar over the last two decades. Over the last few years there have also been QoL studies on laparoscopic and robot-assisted IC and NB UDs.

Discussion
Over the last decade, there have been more QoL studies and more validated and prospective studies. Similarly, the proportionate numbers of NBs has also increased over the last two decades, with newer studies now reporting on laparoscopic and robot-assisted UDs (Tables 1 and 2). There has also been a rise in the number of studies reporting on QoL outcomes in these patients, demonstrating the importance placed on QoL in the last decade.

Change in QoL trends over the last four decades
Whilst publication trends show that better reporting of QoL with more validated questionnaires are now being used, it seems that as long as the patient is well counselled and supported in their decision they learn to cope and adjust with their UD type [1].
Compared to previous decades, the past decade has seen an upsurge in focus on QoL outcomes in bladder cancer. This has occurred in tandem with the development of new and specific HRQoL instruments used in bladder cancer [1,8]. There has also been considerable variability in the use of QoL assessment tools, with a progressive uptake of validated assessment methods in the last decade. Our literature review revealed 13 of the 15 studies between 1987 and 1997 used a non-validated ad hoc (self-designed) instrument to assess QoL outcomes compared to only four in the 2009-2017 period. This suggests that the process of HRQoL measurement is becoming increasingly popular and perhaps clinically responsive.
The ad hoc instruments previously used were potentially poor measures of reliability and qualitative outcomes, and subject to bias due to their inherent non-validated nature [1,8]. In addition, there has been a gradual rise in the globalisation of quality assessment in patients after RC, given that 18 different countries are represented across the studies in 2009-2017 compared to only six in 1987-1997. Only one study across the last three decades accounted for sociocultural influences in health perception and QoL evaluation [11]. Perhaps there is a role for crosscultural testing of these QoL instruments to ensure the validity and reliability of these tools across patients from other countries and cultures [12].

Generic vs cancer-specific QoL assessment
In the past more generic QoL assessment tools were used. Previously, the Beck Depression Inventory (BDI) and Sickness Impact Profile (SIP) [13] were used to assess HRQoL across a wide range of medical conditions, and therefore these were not responsive to       Retro. The mean BIS score in ileal ONB group patients was significantly better than that in IC group patients at the 1-year follow-up, but there was no significant difference at the long-term follow-up. 75 Urol Oncol Pros. There were no clinical or statistical differences between the two arms in QoL change from baseline to 3 month or from 3 to 6 months in any of the evaluated domains (Continued)  [15] and Functional Assessment of Cancer Therapy-General (FACT-G) [16]] failed to address specific domains of importance to patients with bladder cancer. The importance of developing instruments that measure specific outcomes for patients with bladder cancer is slowly being addressed as demonstrated by the fact that 23 studies from 2009 onwards used a bladder-specific QoL tool. Tools such as the FACT-Bladder Cancer (FACT-BL), Bladder Cancer Index (BCI) [17], Body Image Scale (BIS), FACT-Vanderbilt Cystectomy Index (FACT-VCI) [18], suggest a greater appreciation for having a responsive tool that can identify specific concerns in post-RC patients and hopefully act as a framework to compare outcomes and validate more specific tools.
Despite several retrospective studies reporting no clear superiority for NB surgery [17,[20][21][22][23], recent data suggest that NB is increasingly being offered to patients. Although the type of UD should be individualised, the surgeon or centre should be able to offer both types of UDs for surgical equipoise, based on patient preference. Table 1 suggests a progressive increase in the number of prospective studies being performed in the last decade, along with a rise in the reported numbers of NB UDs. We can perhaps postulate that the shift towards NBs has predominantly been driven by improved surgical training in a more complex procedure and better patient counselling, offering a choice of UD rather than the QoL outcomes [24].

Role of laparoscopic and robot-assisted surgery in post-RC QoL
The advent of minimally invasive surgery, such as robot-assisted RC (RARC) and laparoscopic surgery, has led to decreased length of stay and morbidity, and faster recovery. Multicentre data from the USA has suggested a significant rise in its use from 0.6% in 2004 to 12.8% in 2010 [2,3,25]. Poch et al. [26] assessed QoL before and after RARC and reported no significant QoL advantage for RARC. However, the authors found quicker return of urinary function and better body image postoperatively with intracorporeal vs extracorporeal UDs. Studies have failed to show a QoL benefit of RARC compared to open RC [2,27]. A recent meta-analysis also suggests postoperative HRQoL to be similar in patients undergoing RARC and open RC [28]. With the advent of enhanced recovery after surgery (ERAS) protocols, there is now a reduction in postoperative morbidity and hospital stay, with a recent study reporting higher emotional well-being in patients who underwent ERAS [29].

Strengths and limitations of the review
Despite the current trend of QoL studies moving in the right direction with the increased use of validated and specific HRQoL measures; the fact remains that there are still significant challenges in measuring QoL in UD patients. Sexual dysfunction although common is perhaps poorly captured. Conversely, although diseasespecific questionnaires are more responsive than generic questionnaires to subtle changes within disease-specific domains, the high disease-specific sensitivity of these questionnaires may limit the ability to account for unexpected events. For example, an unanticipated neurological adverse event may not be addressed in the disease-specific instrument's questions and as a result this may not be reflected in an accurate QoL change. Of the newer QoL tools, the Bladder Utility Symptom Scale (BUSS) seems to be a novel patientreported outcome instrument and measures HRQoL for all patients with bladder cancer regardless of treatment received or stage of the disease [30]. Various studies have been published investigating QoL after RC and UD. However, there is an extensive deal of heterogeneity amongst these studies with regards to methodology, the use of nonvalidated QoL instruments, and the underpowered and retrospective nature of the majority of data make interpretation difficult. Based on the present studies, QoL has not shown to be significantly variable across the different types of UD. As the majority of them are retrospective in nature, there also remains the risk of inherent selection bias. Furthermore, QoL is only measured postoperatively in most of these studies, and in the absence of preoperative QoL data, it is not truly possible to determine the effect of UD. With different approaches to UD and in the absence of any randomised trials, results from the ongoing prospective, multicentre, randomised trial of open vs robotic radical cystectomy (RAZOR) trial may provide an answer in the near future [31].

Conclusion
The last four decades has seen gradual but significant improvements in the way QoL assessment is conducted in RC patients, with the implementation of several validated, bladder cancer-specific questionnaires and UD-specific constructs. The emergence of more prospective studies with validated QoL instruments has improved our ability to identify their QoL and to understand the differences between various UD types.

Disclosure statement
No potential conflict of interest was reported by the authors.