Overactive bladder syndrome: an underestimated long-term problem after treatment of patients with localized prostate cancer?

Authors


Ruth Kirschner-Hermanns, Continence Clinic/Clinic of Urology, University Clinic RWTH Aachen, Pauwelsstr. 30, 52057 Aachen, Germany. e-mail ruthkirsch@aol.com

Abstract

Study Type – Therapy (case series)

Level of Evidence 4

What's known on the subject? and What does the study add?

In this study we observed courses of micturition symptoms and differentiated degrees of symptoms for each point in time while also considering the impact of bothersomeness. Our data show that not only significantly more patients who have undergone BT suffer from OAB than those who have undergone RP, but also that those affected show significantly higher values for severity of OAB symptoms throughout the whole observation period of 36 months. Our data analysis further shows that variability of OAB symptoms as well as fluctuation of severity of OAB symptoms vary to a significantly higher degree after BT than after RP. Looking only at mean figures at a given point in time clearly underestimates the underlying problem. This fact is not reflected in the literature.

OBJECTIVE

  • • To look at individual courses of postoperative micturition symptoms, especially urgency, in patients treated either with radical prostatectomy (RP) or with brachytherapy (BT).

PATIENTS AND METHODS

  • • In a prospective longitudinal study we investigated individual changes in micturition symptoms before treatment, and 6, 12, 24 and 36 months after treatment.
  • • All patients received the European Organization for the Research and Treatment of Cancer quality-of-life questionnaire, QLQ-C30, and the International Continence Society male questionnaire at each assessment.
  • • We looked at long-term results as well as changes in time using repeated measures analysis of variance. We further analysed fluctuation of symptoms using sum of changes.

RESULTS

  • • Of the 389 patients treated consecutively in our clinic over the last few years, 99 patients with a mean (sd) age of 65 (6.3) years had completed all five questionnaires and thus were further analysed. Of these, 66 (66.7%) were treated with RP and 33 (33.3%) with BT.
  • • With the exception of age, no significant difference was found between the treatment groups either in physical functioning or in prevalence and severity of overactive bladder (OAB) symptoms.
  • • Adjusted for age and pretreatment symptoms in analysis of covariance, we found that there were statistically more symptoms of OAB 36 months after BT compared with those patients treated with RP (P < 0.025). Whereas 30% of patients complained about severe symptoms of urgency after BT, only 11% did so after RP.
  • • Changes of severity of OAB symptoms over the course of time (P < 0.007) using analysis of repeated measures as well as variability of OAB symptoms (P < 0.033) using the two-sided Wilcoxon t-test were significantly higher in patients treated with BT than in patients treated with RP.

CONCLUSIONS

  • • Independently of age and physical functioning, BT is significantly associated with higher rates of long-term urgency symptoms, even after 3 years.
  • • Repeated measurements show that OAB symptoms are highly fluctuating and that in patients treated with BT, severity of symptoms as well as variability of symptoms was significantly higher than in those patients treated with RP.
  • • Persistent OAB seems to be an underestimated problem after treatment for localized prostate cancer, especially in patients treated with BT.
Abbreviations
RP

radical prostatectomy

BT

brachytherapy

OAB

overactive bladder

EAU

European Association of Urologists

I-125

125-iodine

HRQoL

health-related quality of life

RPP

radical perineal prostatectomy

RRP

radical retropubic prostatectomy

Ir-192

192-iridium

EORTC

European Organization for the Research and Treatment of Cancer.

INTRODUCTION

Only few studies looking at long-term bladder dysfunction after treatment of localized prostate cancer are prospective studies and include symptoms of an overactive bladder (OAB), which are often more variable, more bothersome and more difficult to treat than symptoms of stress incontinence. For patients with localized T1a-T2a prostate cancer, the European Association of Urologists (EAU) guidelines [1] provide no distinct recommendation as to whether radical prostatectomy (RP) or a form of brachytherapy (BT) is the optimum treatment. The long-term cure rate for these patients is quite high; 15-year survival rates range from 93–96% in well differentiated carcinomas (Gleason 2–4) to 13–40% in poorly differentiated tumours (Gleason 8–10) [2]. Current literature states that in patients with well or moderately differentiated prostate cancer there is no survival difference between RP and BT [3]. Thus BT, in the form of high dose rate iridium BT or low dose rate 125-iodine (I-125) seeds application, has become a common treatment method for localized prostate cancer. BT is often marketed as achieving superior health-related quality of life (HRQoL) [4]. The main side effects observed are associated with bladder and bowel function as well as impairment of sexual function [5–8]. Although incontinence is less frequent after BT than after RP, patients who have undergone BT frequently complain of LUTS and urgency [6,9]. Although, in recent years, data have been collected regarding complications and HRQoL after different treatment methods [4,7], most studies focus on incontinence and neglect symptoms of OAB, especially urgency. More importantly, all data available base their findings on group analysis, calculating means, and thus ignore the fact that there is widespread and individual variability of symptoms. Standard analysis of data might find that patients whose symptoms improve are equated with patients with worsening symptoms. The aim of the present study is to observe courses of micturition symptoms and to differentiate degrees of symptoms for each point in time, while also considering the impact of ‘bothersomeness’.

PATIENTS AND METHODS

Between November 1999 and December 2006 a total of 389 consecutive patients with T1-T2 prostate cancer underwent therapy for localized prostate cancer in our clinic. Of these, 265 patients (68.1%) underwent either radical perineal prostatectomy (RPP) or different forms of radical retropubic prostatectomy (RRP) and 124 patients (31.9%) underwent BT in the form of high dose rate 192-iridium (Ir-192) BT or low dose rate I-125 seeds application. A self-administered 11-page questionnaire containing a European Organization for the Research and Treatment of Cancer (EORTC) quality-of-life questionnaire (the EORTC QLQ-C30 with the prostate cancer module) [10], the IPSS and the ICS male questionnaire for urinary symptoms [11] and the Kelley questionnaire [12] for faecal incontinence was sent to all patients before therapy. Follow-up questionnaires were sent after 6, 12, 24 and 36 months. Complete questionnaire data (questionnaires completed at each of the five timepoints: before treatment, 6, 12, 24 and 36 months after treatment) were obtained from 99 patients whose mean (sd) age was 65 (6.3) years (median 64, range 43–77 years). The epidemiological and oncological parameters of this subgroup of 99 patients were not significantly different (P > 0.05) between the two treatment groups (Table 1). Neither did the epidemiological and oncological data from the entire collective of 389 patients differ significantly from the data from the subgroup patients who completed questionnaires, with the exception of physical functioning and global health status scores (Table 2) Patients who completed questionnaires scored slightly better but the differences in those domains remained below the 10 points identified by the EORTC as clinically significant [10].

Table 1.  Characteristics of study cohort before treatment
CharacteristicsRP, n= 66BT, n= 33Study cohort, N= 99
P
  1. ns, not specified; HEEQ: higher education entrance qualification.

Mean (sd; range) age, years63.8 (6.4; 43–77)67.2 (5.4 (range 57–75)ns, 0.009
Mean (sd; range) urgency score1.8 (0.8; 1–4)1.7 (0.7 (range 1–3)ns, 0.775
Mean (sd; range) bothersome urgency score1.5 (0.6; 1–3)1.5 (0.6 (range 1–3)ns, 0.890
Mean (sd; range) EORTC questionnaire scores   
 Physical functioning94.9 (12.3; 60–100)97 (7.3; 80–100)ns, 0.364
 Role functioning92.4 (21.9; 0–100)98.5 (8.7; 50–100)ns, 0.13
 Emotional functioning73.2 (25.7; 0–100)74.0 (24.2; 25–100)ns, 0.885
 Cognitive functioning87.7 (19.4; 16.7–100)87.9 (19.7; 33.3–100)ns, 0.964
 Social functioning85.1 (25.2; 0–100)83.8 (24.1; 16.7–100)ns, 0.809
 Global health status71.9 (21.5; 16.7–100)71.6 (19.3; 33.3–100)ns, 0.945
 LUTS (%)9/66 (13.6)5/33 (15.2)ns, 1
 Stress incontinence (%)1/66 (1.5)1/33 (3.0)ns, 1
 Urge incontinence (%)1/66 (1.5)0/33 (0)ns, 1
 Urgency (%)8/66 (12.1)4/33 (12.1)ns, 1
 HEEQ (%)26/61 (42.6)16/31 (51.6)ns, 0.508
 Employed (%)17/66 (25.8)5/31 (16.1)ns, 0.291
 Clinical tumour stage (cT) (%)  ns, 0.063
  cT 132/61(52.4)8/24 (33.3)
  cT 224/61(39.3)16/24 (66.7)
  cT 35/61(8.2)0/24 (0)
 PSA ≥10 ng/mL, (%)26/66 (39.4)13/33 (39.4)ns, 1
Table 2.  Characteristics of study cohort before treatment: comparison between the entire collective and patients with complete follow-up
 Characteristics
Study cohort, N= 389Patients who completed questionnaires, n= 99P
  1. ns, not specified; HEEQ: higher education entrance qualification.

RP, n (%)124 (31.9)33 (33.3)ns, 0.782
BT, n (%)265 (68.1)66 (66.7) 
Mean (sd) age, years65.34 (6.16)64,95 (±6,27)ns, 0.572
Urgency score  ns, 0.206
Median22
Mean (sd)1.97 (1.03)1.77 (0.77)
Bothersome urgency score  ns, 0.712
Median11
Mean (sd)1.62 (0.84)1.50 (0.61)
Mean (sd) physical functioning score91.58 (17.22)95.59 (10.90)0.03
Mean (sd) role functioning score90.47 (24.15)94.44 (18.75)ns, 0.13
Mean (sd) emotional functioning score69.92 (25.06)73.47 (25.11)ns, 0.211
Mean (sd) cognitive functioning score84.94 (20.64)87.76 (16.67)ns, 0.22
Mean (sd) social functioning score79.81 (26.61)84.69 (24.27)ns, 0.101
Mean (sd) global health status score64.81 (21.85)71.82 (20.71)<0.005
LUTS, n (%)236 (63.4)62 (63.3)ns, 0.98
Stress incontinence, n (%)64 (17.4)15 (15.5)ns, 0.652
Urgency, n (%)222 (64)59 (60.2)ns, 0.494
HEEQ, n (%)62 (29)33 (35.5)ns, 0.26
No HEEQ, n (%)152 (71)60 (64.5) 
Employed, n (%)93 (24.9)22 (22.7)ns, 0.87
Retired, n (%)275 (73.5)73 (75.3) 
Others, n (%)6 (1.6)2 (2.0) 

Of the 99 patients in the subgroup, 66 (66.7%) patients underwent RP and 33 (33.3%) underwent BT. Of the 66 patients scheduled for RP, 53 (80%) patients underwent RPP and 13 (20%) patients underwent RRP. An extended RP was carried out in patients for whom the preservation of potency was not desirable [13,14]. Although extended RP was routine at our institution during the study period, unilateral nervesparing was introduced only in recent years, and none of these patients were included in the present study. All surgery was performed by one of the authors (G.J.). Only patients with a prostate volume of <60 mL, a urinary flow rate of >10 mL/s, and no significant residual urine were eligible for BT. Of the 33 patients who were treated with BT, 8 (24%) patients were treated with external Ir-192 irradiation combined with a temporary interstitial high dose rate boost [15] and 25 (76%) patients underwent a permanent low dose rate therapy with I-125 seeds application according to the recommendations of the European Society for Radiotherapy and Oncology/EAU/EORTC [16].

In addition to clinical data, bladder functioning, physical functioning and ‘bothersome’ impact were evaluated. The evaluation of the questionnaires and the database parameters was performed by a urologist and a biometrician (R.K.-H., A.H.) not involved in the care of these patients.

We considered single symptoms of the ICS questionnaire as well as ‘bother’ scores for those affected. Intensity of symptoms was graduated on a five-point ordinal scale: no, sometimes, often, frequent and always symptoms. A symptom is classified as bothersome if the patient describes it as quite a problem or even as a serious problem. OAB and their bothersome impact were our main outcome variables.

Patients’ characteristics were shown using median and range for numerical data and by absolute and relative frequencies for categorical variables. To compare treatment groups, the unpaired Wilcoxon's rank-sum test and Spearman correlation were used. All tests were performed undirected (two-sided) and for any test a P value <0.05 was considered to indicate statistical significance. Whenever significance was based on P < 0.1 this was specifically stated. Calculations were performed using SAS®, Version 9.1.3. and Microsoft Excel. In addition we assessed individual courses with regard to whether the symptoms appeared for the first time or whether they were symptoms that worsened during the time of observation. Sum changes of symptoms between each point in time were conducted and compared using Wilcoxon's rank-sum test. Repeated measures anova using procedure GLM in SAS 9.2. were used for evaluation of time effect on symptoms.

RESULTS

Of the 389 patients treated consecutively in our clinic, 99 had completely filled out all five questionnaires and could be further analysed. Of these, 66 (66.7%) were treated with RP and 33 (33.3%) with BT. Beside age and clinical tumour stage we did not find any significant difference between the treatment groups. We analysed physical functioning, pad usage and LUTS, including OAB symptoms, stress incontinence and urge incontinence, before treatment. In accordance with current literature [17] prevalence of stress incontinence in our clinic was 7.6–9.1% in patients after RP and 0–3.0% after BT, but surprisingly, at 36 months follow-up, 30% of patients who had undergone BT and 11% of patients who had undergone RP still complained of severe (3–5 on a severity scale) symptoms of urgency (Appendix 1, Fig. 1). For severity of symptoms of OAB at 36 months follow-up, including those with no symptoms, the mean score was 2.18 in patients treated with BT vs 1.85 in patients having undergone RP (P= 0.05). Further follow-up throughout the observation period showed consistent differences between the two treatment groups (Fig. 2). Scores for bothersome symptoms in those affected were lower for patients who had undergone RP than in patients who had undergone BT (1.51 vs 1.74, respectively), however, the differences were not significant. Age had no significant influence on development of OAB symptoms after BT (P= 0.07), but pre-existent LUTS symptoms and mode of therapy (BT) were significantly associated with higher rates of OAB (P= 0.001/P= 0.025).

Figure 1.

Frequency of OAB symptoms before treatment and 6, 12, 24 and 36 months after RP compared with BT.

Figure 2.

Severity of OAB symptoms before treatment and 6, 12, 24 and 36 months after RP compared with BT.

Observation over 36 months showed that 6 months after BT, 42.4% had symptoms of urgency and 36 months after BT, 30% still had these symptoms. In patients who underwent RP there was almost no worsening of symptoms above baseline, with a peak of 15.2% after 6 months. Variability of symptoms changed between each timepoint and was significantly higher in patients treated with BT (P < 0.033). Analysis of repeated measures highlighted a significant association between BT and urgency symptoms throughout the whole observation period (F (4.376) = 3.74, P= 0.005).

Furthermore we found a significant overall effect of BT (F (1.94) = 7.58, P= 0.007) on severity of urgency. Treatment method was the determining factor for OAB, regardless of age and physical functioning.

DISCUSSION

As there are limited prospective longitudinal data, the best individual treatment for patients with localized prostate cancer remains unknown [18]. Compared with RP, BT results in lower incidences of urinary incontinence and sexual dysfunction [19,20]. A recent study on HRQoL in patients with low-risk prostate cancer shows the best HRQoL after active surveillance, the next best after BT, then intensity-modulated radiation therapy and lastly RP [4]. Penson et al. [21] and Sanda et al. [5] found that urinary incontinence is a strong predictor of HRQoL after treatment for prostate cancer. Improvement of incontinence rates during the first 2 years after RP is reported by Wei et al. [22]. Although urinary symptoms are common in patients after treatment of localized prostate cancer in the literature, little attention is paid to bladder symptoms such as urgency. There are very few data about long-term development of OAB symptoms and the course of symptom severity after different treatment methods. The present results show that 30% of patients suffer from symptoms of OAB after a period of 36 months after BT and 11% after RP. Our long-term follow-up highlighted a mostly neglected problem that shows little improvement over time.

The present data show that not only significantly more patients suffer from OAB after BT than after RP but also that those affected show significantly higher values for severity of OAB symptoms throughout the whole observation period of 36 months. Although the incidence of urinary incontinence, as defined by loss of urine, is lower in patients after BT than after RP, bladder symptoms, which are known to be often more bothersome than symptoms of urine loss [23–25], are often overlooked. Urinary symptoms are often regarded as an age-related problem as reported in several studies, where multivariate analysis identifies age as a risk factor of urinary incontinence [22,26,27]. Patients who underwent BT in the present study were older than those who underwent RP but comparison of pre-treatment data between these patient groups did not show significant differences in physical functioning, pre-existing LUTS or bothersome LUTS. Two studies have found that age does not affect continence status after treatment of prostate cancer [28,29] although Geary et al. [28] evaluated 481 consecutive patients with a follow-up of only 1 year after RP. Furthermore, even after adjusting our data for age, OAB symptoms were significantly more frequent and patients affected had symptoms of greater severity after BT than patients who underwent RP.

The present data also show that variability of OAB symptoms and fluctuation of severity of OAB symptoms are significantly greater after BT than after RP. Looking only at mean figures at a given timepoint clearly underestimates the underlying problem. This fact is not reflected in the literature. In a previous study we used urodynamic testing in 69 men before treatment and 1 year after treatment to show that, especially after BT, men with pre-existent symptoms of urgency have a higher risk for long-term OAB [30]. These data are confirmed by Wei et al. [22].

The present long-term dataset does not include information about different treatment methods for OAB symptoms such as different antimuscarinergic drugs or intravesical drug therapy; however, one can assume that since all patients see their urologist regularly for tumour control that these incidences of OAB are present despite the treatment administered. The inability to delay micturition can result in more severe restrictions to everyday life with a higher impact on HRQoL than having to cope with symptoms of stress incontinence [31]. Unsurprisingly, the difference between the ‘bother’ score of those affected with urgency after BT did not differ significantly from the scores of those affected with urgency after RP. One also has to keep in mind that stress incontinence can be treated in many cases, sometimes with minimally invasive surgical intervention. Persistent urgency, however, remains a therapeutic challenge.

A limiting factor of the present study is the low response rate. The long observation period and the necessity to respond at all five timepoints resulted in a response rate of 20%, which means a response rate of 72% at each timepoint. Our finding that patients who completed questionnaires scored slightly better in physical functioning and global health status supports a former study where, together with refusal, lower health status is a leading reason for no response [32]. Another reason for the low response rate might be gender. In a study by Hazell et al. [33] responders were more likely to be female. In that study, response rates of ≈ 70% were similar to ours at each point in time. We realize that non-response bias is a common problem of postal surveys but, because of limited funding, we could not enhance response rates by use of reminders or monetary incentives.

However, we found no significance difference in epidemiological and oncological data between patients with a completed set of questionnaires and non- or non-complete responders. The strengths of our study are its prospective design and long-term-follow up.

In conclusion, the greater risk for OAB symptoms after BT compared with RP should be discussed in detail with the patient to find the best treatment for the individual. The present findings facilitate patient counselling regarding post-treatment bladder symptoms and highlight the need for prospective studies to evaluate urinary symptoms in addition to incontinence and HRQoL.

ACKNOWLEDGEMENTS

This study was funded in part by the Förderverein zur Kontinenzforschung und Kontinenzaufklärung e.V., Amtsgericht Aachen, Vereinsregister-Nr: 43 57, Steuernummer: 201/5953/1387.

CONFLICT OF INTEREST

None declared.

Appendix

APPENDIX 1

DEFINITIONS

Intensity of symptoms was measured using 5 categories:

  • • 1 = never
  • • 2 = sometimes
  • • 3 = often
  • • 4 = frequent
  • • 5 = always

LUTS was defined as a positive answer to at least one of the following questions, disregarding severity of the symptom:

  • • Do you have to rush to the toilet to urinate?
  • • Does urine leak before you can get to the toilet?
  • • Does urine leak when you cough or sneeze?
  • • Do you ever leak for no obvious reason and without feeling that you want to go?
  • • Do you leak urine when you are asleep?

Urgency is defined as a positive answer to the following question, disregarding severity of the symptom:

  • • Do you have to rush to the toilet to urinate?

Urge incontinence is defined as a positive answer to the following question, disregarding severity of the symptom:

  • • Does urine leak before you can get to the toilet?
    • OAB is defined as either a combination of urgency and urge incontinence or one of those symptoms.

Stress incontinence is defined as a positive answer to the following question, disregarding severity of the symptom:

  • • Does urine leak when you cough or sneeze?
    • A symptom is classified as ‘bothersome’ if the patient states that the symptom is quite a problem or even a serious problem.

    • A symptom is considered to be a de novo symptom if the symptom did not exist at the previous dates of observation.

    • A symptom is considered to have become worse or better if, between two observational dates, a difference has occurred in the severity scoring, whereby each grade on the five-point scale is considered to be one point.

Ancillary