Prospective comparative study of quality of life in patients with bladder cancer undergoing cystectomy with ileal conduit or bladder preservation

Introduction

Multiple single and multi-institutional studies have shown that bladder preservation (BP) and radical cystectomy (RC) yield similar oncological outcomes in well-selected patients.1 2 RC involves removal of the bladder and surrounding organs (prostate and seminal vesicles, surrounding adipose tissue and peritoneum in males and urethra, uterus, fallopian tubes, ovaries, anterior vaginal wall in females) with bilateral pelvic node dissection and urinary diversion surgery. It is a major surgery with a 90-day postoperative mortality of 2.5%–5% even in high-volume centres and has a complication rate of 25%–97%, which can impact the quality of life (QOL) in survivors.3–5 BP involves maximal safe transurethral resection of bladder tumour followed by concurrent chemoradiotherapy with or without neoadjuvant chemotherapy (NACT). It provides an advantage of organ preservation over RC. Although it is a non-invasive treatment modality, 20%–25% of patients experience late bowel and bladder toxicity impacting QOL.6 Salvage cystectomy is reserved for local failure.

QOL may be decisive when different treatments yield comparable survival outcomes. Although oncological outcomes are similar with RC and BP, treatment-related toxicity and health-related QOL (HRQOL) can be different with both these treatment modalities. Both treatment modalities can impact urinary, bowel and sexual function predominantly in long-term survivors. There is limited literature comparing HRQOL after RC and BP. The majority of studies are retrospective and include either RC or BP alone.

This prospective study compared the HRQOL of patients who underwent RC or BP as a curative treatment for their bladder cancer using the Bladder Cancer Index (BCI)7 QOL Questionnaire (QOLQ) and European Organisation for Research and Treatment of Cancer Quality of Life Questionnaire (EORTC QLQ-C30) scores.8

The primary objective was to study the impact of RC and BP on HRQOL using different domains of the BCI QOLQ. The secondary objective was to study HRQOL using the EORTC QLQ-C30 in two groups. The inclusion criteria were age between 18 and 85 years, histopathological diagnosis of bladder cancer, The American Joint Committee on Cancer (AJCC) stage cT1–cT4, cN0–N1, cM0 of bladder cancer, no evidence of recurrence and minimum 6 months of follow-up after the primary curative treatment (RC vs BP), patients ability to understand and answer the QOL instrument. Patients treated with palliative intent, those who did not complete planned treatment or who could not comprehend the QOL instruments and who had a history of other cancer except incidentally detected prostate cancer in the cystectomy specimen were excluded. RC involved the removal of the bladder with surrounding organs, bilateral pelvic node dissection and creation of ileal conduit diversion. BP patients received radiotherapy (RT) to the entire bladder to a dose of 64 Gy/32#/6–7 weeks and elective pelvic irradiation to a dose of 55 Gy/32#/6–7 weeks using plan of the day technique. NACT or adjuvant chemotherapy in RC group and concurrent chemotherapy in BP group were offered in chemotherapy fit patients. After written informed consent, two HRQOL instruments were served to eligible patients on follow-up, that is, the BCI for bladder cancer-specific HRQOL and the EORTC QLQ-C30.

BCI is a validated bladder cancer-specific HRQOL instrument developed by researchers at the University of Michigan. It is a disease-specific, multidimensional, reliable, robust and responsive HRQOL instrument used in bladder cancer.7 It consists of 36 items in 3 principal domains (urinary, bowel and sexual) and 2 subdomains of ‘function’ and ‘bother’. After approval from the University of Michigan, BCI QOLQ was translated into two native languages (Hindi and Marathi). Pilot testing of translated Hindi and Marathi QOLQ was done in the first ten patients in order to understand any trouble in responding, confusion while responding, trouble in comprehending the questions or whether the questions were improper. None of the pilot testers reported difficulty understanding and responding to the questions asked; hence, translated versions of Hindi and Marathi BCI QOLQ were accepted for further study.

The EORTC QLQ-C30 V.3 was also used for an overall assessment of QOL. The EORTC QLQ-C30 is the most commonly used instrument in patients with cancer and includes 30 items under three domains—functional, symptom and global health. It has already been translated and validated in various native languages, including Hindi and Marathi. The EORTC QLQ-C30 consists of five functional scales, three symptom scales, a global health status scale and six single items.

Based on previously published results9 for patients with bladder cancer, a sample size of 51 patients in each group was required to detect a 10-point difference in the mean scores with 80% power using a two-sided t-test at the 5% significance level. A clinically meaningful difference was defined as a 10-point difference in the mean scores of the two groups for each questionnaire, and a p<0.05 was considered statistically significant.10

For BCI scoring, the response to each item is standardised by a Likert scale item to 0–100. The standardised values for all items within a group are averaged to create the domain score. If 20% of items that comprise a domain are missing a response, the corresponding domain summary cannot be calculated. Otherwise, the domain score was calculated from non-missing items. Higher mean scores represented better health states. In addition to the domain score, the difference between mean scores for individual questionnaires to get in-depth knowledge about HRQOL was evaluated.

For EORTC QLQ-C30 scoring, raw score was calculated for each domain using linear transformation standardised to a score of 0–100. A high score for a functional scale represents a high level of functioning, a high score for the global health status represents a high QOL, but a high score for a symptom scale represents a high level of symptom burden. The difference between mean scores for individual questions was estimated.

The mean QOL scores across various domains and specific questions were compared between the two treatment groups using the independent t-test. The χ² test was used to compare categorical responses between the two groups. The Mann-Whitney U test was used to compare the medians between the two groups. Patients or the public were not involved in the design, or conduct, or reporting, or dissemination plans of our research.

Results

104 patients were enrolled, of whom 56 underwent RC and 48 opted for BP. 95 (91.3%) patients were male. The median age was higher in BP than in RC by 6 years. The median time from treatment completion to QOL assessment was longer in the RC group (31 vs 18 months p=0.1). One-third of patients in both groups received NACT. In addition, 7 (12.5%) RC patients received adjuvant chemotherapy and 29 (60%) BP patients received concurrent chemotherapy. The clinical characteristics of the study population are summarised in table 1.

The response rate to questions about urinary (93/104, 88.5%) and bowel (93/104, 89.4%) function, as well as the EORTC QLQ-C30 questionnaire (101/104, 97.11%), was high. However, the response rate to sexual QOLQ was only 53%. In the RC group, 25/56 (44.6%) responded to sexual QOLQ while in the BP group, 30/49 (62.5%) responded to sexual QOLQ. The mean BCI urinary scores were BP_mean 89.7, RC_mean 85.1, p=0.12. and mean bowel scores were BP_mean 88.9, RC_mean 87.4, p=0.55. There was no significant difference in function or bother subdomains between the two groups (table 2, figure 1). Overall, BCI sexual scores were low in both groups but significantly better after BP (BP_mean 56.9, RC_mean 41.5, p=0.01). Mean scores for sexual function were 38.4 in BP and 25 in RC (p=0.07) and for sexual bother, 81 in BP and 62 in RC (p=0.02).

Request permission

On question-wise assessment, for urinary function, the ability to perform an exercise (BP_mean 95, RC_mean 85,p=0.06) and urinary leakage at night time (BP_mean 91.7, RC_mean 77.6, p=0.01) were better in the BP group while scores for blood in the urine (BP_mean 87, RC_mean 97, p=0.05) were better in the RC group compared with BP. There was no significant difference in any of the bowel-related questionnaires in either group. On question-wise assessment for sexual function, mean scores for frequency of sexual activity (BP_mean 23.53, RC_mean 3.23 p=0.005) and ability to perform sexual function (BP_mean 43.5, RC_mean 25, p=0.02) were significantly better in the BP group. For sexual bother, mean scores for level of sexual desire (BP_mean 84.68, RC_mean 54.84, p=0.003), ability to have intercourse (BP_mean 77.50, RC_mean 50.89, p=0.018) and ability to reach orgasm (BP_mean 79.17, RC_mean 50, p=0.010) were significantly higher in the BP group compared with RC group.

There was no clinically meaningful difference in EORTC QLQ-C30 scores for functional, symptom and global health scales in both groups (table 2 and figure 2).

Request permission

Discussion

This prospective, cross-sectional study compared the long-term patient-reported QOL outcomes in bladder cancer survivors treated in a tertiary care centre. Assessment using a validated bladder cancer-specific QOL using BCI and general QOL assessment using EORTC QLQ-C30 showed that overall, the QOL outcomes were good irrespective of treatment modality. Sexual QOL was significantly better in the BP group compared with the RC group, with an absolute difference of 15.4 points (BP_mean 56.9, RC_mean 41.5, p=0.01). There was no clinically meaningful difference between urinary and bowel BCI score or EORTC QLQ-C30 domains.

Results of better sexual QOL with RT than surgery have been reported previously.9 11–13 In a study by Caffo et al, four ad hoc questionnaires related to sexual QOL were administered. In the sexually active group, 20% from BP and 70% from the RC group experienced some limitations in sexual activity. Erectile dysfunction was reported in 25% of BP and 80% of RC group.11 Mak et al compared sexual QOL in both groups using EORTC QLQ-BLM30 which included eight questions related to sexual activity. The mean scores for erectile dysfunction, ejaculation problems, discomfort during intercourse and fear of contaminating a partner during sexual contact were significantly higher in the RC compared with the BP group. Our study showed a 32-point difference in erectile dysfunction between both groups. We used a gender non-specific BCI questionnaire and showed a 30-point difference in the level of sexual desire, a 29-point difference in the ability to reach orgasm and 27-point difference in the ability to have intercourse. Both RC and BP affect sexual function for males and females in different ways.

Sexual dysfunction after treatment for bladder cancer is multifactorial.14 In males, RC may cause damage to the cavernous sinus located within the neurovascular bundle (NVB), direct nerve injury, inflammation and fibrosis which leads to erectile dysfunction.15 RC in females includes removal of genital organs that reduce vaginal capacity either from vaginal foreshortening or narrowing. In female, parasympathetic fibres mediate the release of vasoactive intestinal peptide in the vagina, resulting in vaginal lubrication. After cystectomy injury to NVB leads to vaginal dryness resulting in dyspareunia. Damage to genital organs, devascularisation of the clitoris, total urethrectomy and the need for complete anterior pelvic exenteration may lead to further deterioration in sexual function.16 Impaired physical condition, body alteration due to urinary diversion and urinary incontinence can impact sexual function after surgery. Damage to the NVB in male and vaginal mucosal fibrosis leading to dryness and atrophy are the main iatrogenic causes of sexual dysfunction after pelvic RT.17–19 Irrespective of treatment, impaired physical condition, emotional and psychological responses of both patients and their partners psychological symptoms including anxiety, depression, low self-esteem, age, comorbidity, pretreatment sexual function, body image and age-related changes in libido and sexual interest are other causes of sexual dysfunction. Sexual QOL after newer surgical strategies such as nerve-sparing or urethral or vaginal sparing cystectomy after RC or NVB-sparing RT needs to be explored in future studies.20 21

Intuitively, one would believe that patients with an intact and functioning bladder would have a better urinary QOL than a patient with a urinary diversion and a stoma. Surprisingly, QOL with respect to the urinary domain of BCI showed no significant difference in overall scores, function scores or bother scores between the two groups. Our findings are in line with previous studies by Mak et al and Catto et al.9 12 The possible reason for these findings may be the use of generic questions related only to frequency, urgency and incontinence alone, which is unlikely to capture the nuances that are specific to treatment modality such as psychological impact on body image.22 Other reasons could be cross-sectional study design,23 24 small sample size and gender non-specific questions,25 as well as good anatomical and psychosocial adjustment to the ileal conduit over a period of time.26 27 When individuals undergo changes in their health status, they may adjust their internal standards, values or perception of QOL. These ‘response shifts’ can also impact or skew QOL results.28

Bowel QOL was expected to be worse with RT due to the pelvic irradiation, including small bowel within RT portals. There was no significant difference in the mean bowel scores between the two groups in this study. Mak et al also reported no clinically meaningful change in the bowel domain.9 Even though pelvic RT can impact bowel function, recent advances in RT techniques allow greater dose escalation on the treatment targets, with lower doses to the surrounding tissues, which might result in less toxicity and consequently better QOL. All participants in our study underwent RT using advanced adaptive image-guided technique3 which may have contributed to minimal late bowel side effects and consequently QOL. Additionally, the small sample size and cross-sectional study design may have contributed to the similar QOL outcomes in both groups.

We noticed a poor response rate for sexual subdomain questions compared with bladder and bowel-related subdomains. The response to sexual questions is a complex and multifactorial issue. Contrary to expectation, response to sexual questions was more in the BP (63%) compared with the RC (45%) group even after the age difference. In elderly individuals with bladder cancer, particularly in the BP group with a median age of 65, the poor response could be linked to sexual inactivity even before treatment. The poor response rate in the young surgical group could be due to non-response bias, as patients with problems are less willing to answer. It is difficult to frame sexual QOLQ in a culturally and socially acceptable way, which can be one of the reasons for poor response. A similar poor response rate of 35%–85% for sexual QOLQ has been reported in the published literature.9 11 12 Even though BCI sexual QOLQ are gender-friendly and gender neutral a poor response rate was seen in females. Only two females from the RC group answered the sexual questionnaire while two reported as sexually inactive. Zietman et al have also reported a similar poor response rate for sexual QOLQ in females.29

There was an excellent response rate of 97.11% for EORTC QLQ-C30. Overall QOL was good in both groups, indicating good adaptation post-treatment in both groups. We could not find any clinically meaningful difference in functional, symptom or global health scale in both groups. On subdomain analysis role scale, social scale, pain, nausea and vomiting subscales indicated statistically significant differences but were not considered clinically meaningful change (online supplemental table 1). A study by Mak et al demonstrated a 9.7-point difference in EuroQOL EQ-5D scores indicating better general QOL in the BP group compared with RC. This study also showed higher physical (5.7), role (8.2), social (9.9), emotional (6.6) and cognitive functioning (8.1) scores of EORTC QLQ-C30 in the BP group.9 Another study compared general QOL using EORTC QLQ-C30 (GHS 69 in BP vs 72 in RC) and EuroQOL EQ-5D scores showed that there was no difference between RC and RT group. However, this study also showed that general QOL was poor in bladder cancer survivors as compared with the general population and other patients with pelvic cancer.12

The cross-sectional design is one of the limitations of this study. Some studies report an initial drop in QOL immediately after treatment, but then it reaches a plateau over time.23 24 30 A cross-sectional study design is inadequate to capture these dynamic changes in QOL. Prospective cohort studies to evaluate dynamic changes in QOL are ongoing.31 32 Another limitation of this study is the different median follow-up times in both groups (BP 18 months vs RC 31 months) which may not be representative of QOL at the same time points with two different treatment modalities. Also, we could not compare QOL outcomes with the general population or assess the impact of age, gender and comorbidities in this study. There were some differences in patient and treatment-related factors between the BP and RC groups, most of which were statistically not significant and are reflections of clinical practice. Since all our RC patients underwent ileal conduit urinary diversion, we could not assess the impact on the QOL of patients who had neobladder reconstruction. Self-selection bias is another limitation of this study as we did not maintain a screening log for this cross-sectional study.

Conclusion

Survivors of bladder cancer have good general, urinary and bowel-related QOL regardless of the type of local treatment. Sexual function is low but better in BP as compared with the RC group. In the modern era of personalised medicine, QOL outcomes should be discussed with patients during treatment decisions.