The quality of life in men after radical cystectomy with a continent cutaneous diversion or orthotopic bladder substitution: is there a difference?

Authors


Abstract

Objective  To compare the quality of life (QoL) in men after radical cystectomy who had either a continent cutaneous diversion or orthotopic bladder substitution.

Patients and methods  Eighty men with at least 6 months of follow-up and with no signs of recurrent disease after radical cystectomy for bladder carcinoma, and who had either a continent cutaneous diversion or orthotopic bladder substitution, were sent two types of questionnaire, the Functional Assessment of Cancer Therapy–Bladder Cancer (FACT-Bl) and the Hospital Anxiety and Depression Scale (HADS), to determine their QoL.

Results  The FACT-Bl and HADS questionnaires were returned by 90% and 71% of the patients, respectively ( P  < 0.05). In the replies to the generic version of FACT-Bl there were no differences between the groups in any domain, the scores being high in all. In questions covering intestinal, urinary and sexual items, patients with continent diversion had less trouble controlling urine ( P  < 0.0001), had to empty less often ( P  < 0.001), and had fewer symptoms when emptying ( P  < 0.05). Patients with neobladders had a better appreciation of appearance and better preserved erectile function (both P  < 0.05). In the answers to the HADS, the mean scores were low (within the normal ranges) and did not differ between the groups.

Conclusion  Using two instruments tested for validity and reliability, there were no differences between patients with continent diversion and those with orthotopic substitution.

Introduction

New methods for reconstructive surgery are usually introduced with the goal of increasing the quality of life (QoL) of patients. Patients with malignant diseases should be cured with the least possible detriment to their QoL. When continent cutaneous diversion and orthotopic bladder substitution became realistic alternatives in clinical urology it was thus a priori assumed that these measures should considerably improve the patient's adjustment after radical cystectomy for bladder carcinoma, compared with conduit diversion. Over the past 10 years numerous reports have been published on QoL after cystectomy using different methods for reconstructing the urinary tract; it is with some surprise and disappointment that the conclusion was that there is no clear evidence that the new methods give a better QoL [1–3].

The urological community has rather uncritically adopted the concept of QoL and applied it in many diseases and conditions. There is still no consensus among groups working with the concept of QoL, mainly psychologists, about how to define it, although it is commonly agreed that it is multidimensional and subjective. As the definition of QoL varies, the number of instruments in use intended to measure QoL is overwhelming and consequently the results are difficult to evaluate and compare. One reason for the wide range in results obtained in studies of patients after cystectomy could be that different QoL instruments have been used; there are few published studies which use similar measures. Most of the questionnaires have been ad hoc instruments which have not been tested for validity and reliability.

In the present study we used two instruments that are accepted and well known to test whether there were differences between patients with a continent urinary diversion and those with orthotopic bladder substitution, the latter method generally being considered better than diversion in terms of QoL after radical cystectomy.

Patients and methods

All men who had undergone radical cystectomy and continent cutaneous diversion or orthotopic bladder substitution with presently used surgical techniques, and who had at least 6 months of follow-up with no signs of recurrence, were asked to participate. In all, 80 men agreed (39 with a continent diversion and 41 with an orthotopic substitution). They had undergone surgery between December 1987 and March 2000. They were mailed two types of QoL instruments, with a stamped return envelope. If these were not returned within 4 weeks, a reminder letter was sent; 35 and 29 patients, respectively, returned the questionnaires, i.e. 90% and 70% (P < 0.05). These patients form the basis of this report and their characteristics are given in Table 1.

Table 1.  The patients’ characteristics
 DiversionSubstitution
  • *

    P  < 0.05.

No. of patients3529
Mean (range):
age at study, years70 (47–82)66 (48–82)
follow-up, years  4.1 (0.9–9.6)  6.1 (0.7–12.8)*
No. receiving:
Preop irradiation (20 Gy)  1 
Preop chemotherapy (MVAC)  1 
Neoadjuvant chemotherapy  1 
Adjuvant chemo (MVAC)  4 
Outlet revision for urine  leakage  1 
Stomal revision for stenosis  2 
Implantation of AUS   2
CISC for residual urine  and/or urine leakage 11
Penile implant   1

Contacting the 16 men not responding by telephone showed that the reasons for not participating were myocardial infarction in one, other malignancies in two, Alzheimer's disease in one, renal insufficiency caused by Goodpasture's syndrome in one, tiredness in three and unwillingness in eight. The unwilling patients had difficulties in comprehending questions or found it emotionally difficult to confront them. Of these patients, two had undergone continent diversion and six orthotopic bladder substitution.

Surgery

For continent urinary diversion, the ‘Lundiana’ Pouch [4] was used. This is a modification of the Indiana Pouch, the main differences being that the ileocaecal valve is also stapled and tethered to the caecal wall, forming a small flap-valve. The stoma was in the right lower quadrant or in the umbilicus. All these patients (except three) had undergone urethrectomy at time of cystectomy because they had widespread carcinoma in situ, tumour of the bladder neck, or tumour in the prostatic urethra/prostate, or because they felt that the possibility of urine leakage after neobladder construction was repugnant.

The neobladders were also constructed from detubularized right colonic segments, essentially according to the technique described by Goldwasser et al.[5]. Before surgery the oncological status and the different modes of urinary tract reconstruction applicable were carefully described to the patient, and the decision on the method for reconstruction used was thus based on oncological principles, the fitness of the patient and his request.

QoL instruments

The Functional Assessment of Cancer Therapy – Bladder Cancer (FACT-Bl) is a questionnaire specially designed for patients with bladder cancer and consists of a general version (FACT-G) which has been validated and tested in several types of cancers [6]. Well-being in four domains is assessed, with patients responding to statements on a five-level ordinal Likert scale, ranging from ‘not at all’ to ‘very much’. The domains are physical, social/family, emotional, and functional. The additional concerns listed in the FACT-Bl concern urinary tract symptoms, intestinal symptoms and sexual symptoms, with a total of 10 statements, with two more for those with a stoma.

The Hospital Anxiety and Depression Scale (HADS), also designed for self-assessment, comprises seven items referring to anxiety and seven dealing with depression [7]. Statements are affirmed or rejected on a Likert scale with four options (0–3), giving a possible maximum score of 21 for depression and for anxiety. A threshold of 8 is recommended to identify all possible cases of anxiety and depression, and a threshold of 11 to detect all probable cases of these disorders. The effectiveness of HADS for detecting mode disorders has been assessed in various cancer populations [8,9].

The study was approved by the Ethics Committee of the Lund University. The results were analysed statistically using the chi-square test, Fisher's exact test, the Mann–Whitney U-test and Pearson's R-test, with P < 0.05 taken to indicate statistical significance.

Results

There were no differences between the groups in the replies to the FACT-G (Table 2). In all domains the patients had high scores, which for physical and emotional well-being were close to the maximum scores obtainable. For the additional concerns expressed in the FACT-Bl, patients with continent diversion had less trouble controlling urine (P < 0.0001). Thus, 28 of 35 men reported no trouble at all, while this was reported by only six of 29 with orthotopic substitution. Likewise, the former group had to ‘urinate less often’ (P < 0.001; Table 3) and had less ‘burning’ (P < 0.05). While interest in sex was equal, being affirmed to varying degrees by 19 of 33 with diversion and 17 of 27 with neobladders, only one of 31 men with diversion could maintain an erection to varying degrees, while this was possible by six of 26 with a neobladder (P < 0.005). The statement ‘I like the appearance of my body’ was confirmed to varying degrees by 29 of 34 patients with a reservoir and by 27 of 28 with a neobladder (P < 0.05).

Table 2.  The median (interquartile range) scores in the FACT-G and HADS instruments
Function (maximum)DiversionSubstitution
  1. HADS; 0–7, normal; 8–10, mild; 11–14, moderate; 15–21, severe.

FACT-G
Physical well-being (28)27.0 (24.0–28.0)26.0 (24.0–27.0)
Social/family (28)22.5 (20.0–24.5)24.0 (20.0–25.0)
Emotional (24)22.0 (20.0–24.0)22.0 (20.0–24.0)
Functional (28)23.0 (20.0–26.0)22.0 (20.0–25.0)
Total score (108)95.0 (87.5–100.0)94.0 (86.5–98.0)
HADS
Anxiety2.0 (0–5.0)1.0 (1.0–3.0)
Depression2.0 (0–5.0)2.0 (0–4.0)
Table 3.  Urine leakage and emptying after cystectomy
Response ‘trouble controlling
my urine’
‘urinate more
frequently than usual’
DiversionSubstitn.DiversionSubstitn.
Not at all28  62711
A little bit  314  0  5
Somewhat  1  4  4  4
Quite a bit  0  3  0  6
Very much  1  2  1  1
Not answering  2   –  3  2
P< 0.0001 < 0.001 

Two statements were eligible only for the 35 patients with a reservoir; ‘I am embarrassed by my stoma’ and ‘Caring for my stoma is difficult’. Of the 31 patients who answered the questions, 26 replied ‘not at all’ to both.

The HADS showed no differences between the modes of reconstruction; mean scores were low for anxiety and depression, falling within the normal ranges (Table 2). Six and two of the 35 patients with a reservoir scored > 7 for anxiety and depression, respectively, while the corresponding values among the 29 patients with an orthotopic substitute were 2 and 1 (not significant). Age and duration of follow-up did not influence the results obtained in the FACT-Bl or in the HADS.

Discussion

This study showed that patients with orthotopic bladder substitutes had less control over storage and emptying of urine than had patients with a continent cutaneous diversion, while the latter group, in which the vast majority of patients had undergone urethrectomy, had less erectile function. These functional differences did not translate into differences in QoL as measured by the questionnaires used. Psychiatric distress was low in both groups of patients.

Several explanations must be considered for these results: (i) There may have been bias in the sample of patients assessed; (ii) The instruments used were not sufficiently sensitive to detect differences; (iii) The measures used were unsuitable; (iv) There is no real difference in QoL between the groups, i.e. the result obtained is valid.

For the first item, it is undoubtedly a source of concern that only 70% of patients with a neobladder returned the questionnaires, while 91% of those with a continent stoma did so. Six and two patients, respectively, did not want to participate, because they found the questions difficult or felt uneasy when confronting them. Missing data are a problem in QoL studies, with some reporting very low participation rate [3]. As those not responding tend to have poorer functional results than responders in clinical trials [10], had they answered it could possibly have worsened the results in the neobladder group. Indeed, the sample of patients in both groups was relatively small and a larger series might have given results of statistical significance. However, all patients operated upon and alive were contacted. In addition, it is possible that pre-existing differences in QoL between the groups are important, e.g. if one group had a much better QoL before the procedure but was more adversely affected, then the outcome would show no differences.

It is possible that the instruments used were insufficiently sensitive to pertinent differences between the groups, especially if the effects were small. Evidence of validity and reliability does not guarantee responsiveness; QoL is a highly individual judgement and some of the items on the FACT-Bl are very general; ‘I feel close to my friends’, ‘I get emotional support from my family’. It is unlikely that such issues would be affected by differences in surgical procedures, especially when the patient has no knowledge of how he might have fared had he undergone the alternative procedure. Although the questionnaires have been previously used in QoL studies of patients with bladder cancer this does not necessarily mean that they are measures of QoL. FACT-G (in the present study, FACT-Bl) has been tested for validity and reliability, translated into several languages and used in another study of patients after cystectomy, comparing those with conduit diversion and those with continent reconstruction [11]. Those authors were unable to detect any difference in QoL between incontinent and continent reconstruction (most with orthotopic substitution), with both groups having high scores despite urinary incontinence being more common after continent reconstruction. Some other groups have used established instruments, e.g. the Sickness Impact Profile [12], the Short-Form-36 [13,14], and the Beck Depressive Inventory and Profile of Mood States [3,15], and found no differences between patients with different modes of reconstruction after cystectomy. However, a notable difference is a recent report comparing patients with an ileal conduit and those with orthotopic neobladders, in which the latter scored significantly better than the former in all domains in the EORTC QLQ C-30 [16]. However, these different instruments do not measure the same attributes; thus it was recently reported from a study comparing FACT-G and the QLQ C-30 that neither of the instruments can be replaced by the other, and direct comparison of the results obtained with the two instruments is impossible [17].

The HADS was developed specifically for use with psychiatric patients; whether or not it is suitable for the assessment of QoL must remain a moot point. To our knowledge the HADS has been used in only one study in patients after cystectomy [18]. In those with a neobladder anxiety and depression was found in only 5%, which did not differ from the incidence in normal controls. In the present study the incidence of anxiety was higher (12.5%) but the incidence of depression was similar. However, a recent study questioned the reliability of the HADS for detecting psychological morbidity [19]. Although that study concerned women treated for early breast cancer, the result adds to the uncertainty about a suitable instrument to use in research and screening for adjustment disorders after radical cystectomy. Perhaps the instruments used are too crude to detect possible differences among groups of patients with different modes of reconstruction. Qualitative methods, e.g. an in-depth interview, are more likely to detect psychological distress [19].

Adaptation to illness and disability is well-documented [20]; thus, after 6 months it is likely that most patients will have adjusted to their changed circumstances. In the present study there was a wide range in the duration of follow-up of the patients (0.5–13 years) but no correlation between the duration and the answers to the questionnaires, indicating that the process of adaptability occurs soon after surgery. This is in agreement with one of our previous studies [2]. Drummond [21] found that patients with chronic conditions ‘protected’ episodes of illness from the social context of their everyday life, and factors such as family, friends and leisure pursuits were more important than health status.

Failure to establish a clear definition of QoL has meant that problems with validity are paramount. It would seem advantageous that the FACT-Bl and the forthcoming BLM 30 from EORTC focus on consequences specific to patients with bladder cancer, but it is becoming increasingly obvious that symptoms and health status are constructs that are clearly separated from the construct ‘QoL’[22,23]. It is important to keep this in mind, as published reports show numerous examples of these ideas being used interchangeably. The use of the expression ‘health-related QoL’ does little to clarify the issue, as it is impossible to separate the effects of the disease or surgical procedure from other influences on a person's QoL [24]. Therefore it was not surprising that the present patients in both groups had high scores for emotional well-being in the FACT-G and showed little or no anxiety or depression in the HADS. This result fits well with studies that have shown that patients after cystectomy do not differ from the population at large in QoL [14,15]. In a recent study decreased well-being was as common in a control population as in patients with an orthotopic bladder substitute [25].

A careful and objective description to the patient preoperatively of how the surgery is undertaken and how the reconstructed urinary tract works is probably of considerable importance, underscored in many studies. This creates realistic expectations and the patient will then more easily adapt. The importance of active participation of the patient in decision making has been indicated [14].

Confusion stems from the fact that there is no definition of QoL; as it is a hypothetical construct it is unlikely that it will ever be possible to define it. Obviously it has very little to do with health status, and the urological community should recognize this. A possible danger in not doing so is that patients with physical consequences from diseases or from ‘older’ surgical procedures are labelled as having a lower QoL than patients with a ‘modern’ type of reconstruction or some average or ‘normal’ individual. Such reasoning could lead to serious ethical consequences in the perception of the disabled.

Care should be exercised when planning ‘QoL’ after surgical procedures for treatment of malignant diseases. Studies seem relevant for specific questions concerned with, e.g. the influence of different cultural and religious backgrounds, spiritual beliefs, the effects of different supportive measures, etc. There has even been a suggestion that measuring QoL is counterproductive and would be better replaced by a more easily understood assessment of common factors in perceived health status [22]. Nevertheless, it is important to continue to study health-related aspects of radical cystectomy with different reconstructive methods for bladder cancer. However, definitive answers would require comparative prospective trials, which are currently rare.

Acknowledgements

This study was supported by grants from The Swedish Cancer Society, GAE Nilsson Cancer Foundation and The Swedish Foundation for Health Care Sciences and Allergy Research.

A. Månsson, Department of Nursing, Lund University, PO Box 157, 22100 Lund, Sweden.
e-mail: asa.mansson@omv.lu.se

Ancillary