An unrandomized prospective comparison of urinary continence, bowel symptoms and the need for further procedures in patients with and with no adjuvant radiation after radical prostatectomy


Dr T. Hofmann, Department of Urology, St. James's University Hospital, Beckett Street, Leeds LS9 7TF, UK.



To prospectively assess, using a questionnaire-based study, the relative differences and changes in urinary continence and bowel symptoms, and the need for further surgery, within the first year after radical retropubic prostatectomy (RRP) in patients with and with no adjuvant radiotherapy (aRT).


The study included 96 men with clinically organ-confined adenocarcinoma of the prostate who underwent RRP between March 1998 and June 1999. A subset of 36 patients was recommended aRT of the prostatic fossa (median dose 54 Gy) because of positive surgical margins and/or seminal vesicle involvement. Using a mailed questionnaire all patients were prospectively assessed at 4-month intervals for the first year after RRP.


Valid data were analysed from 83 patients (overall response rate 86%), of whom 30 (36%) had received aRT. At 4 months a significantly lower proportion used no pads and significantly more used 1 pad/day in the aRT than in the RRP group (both P < 0.05). Eight and 12 months after RRP there was no statistically significant difference between the groups in urinary incontinence. However, 53% of men in the aRT group had stool urgency and 13% reported fecal incontinence at 4 months, compared with 1.9% and none (both P < 0.01) of the RRP group. At 1 year after RRP bowel symptoms and fecal continence improved in the aRT group and there was no significant difference for these symptoms between the groups. Starting aRT early (≤ 12 weeks after RP) or late (> 12 weeks) had no significant effect on urinary continence, bowel symptoms and fecal incontinence. Apart from dilatation of urethral strictures in one patient in each group, no further procedures were reported during the follow-up.


A moderate dose of aRT after RRP had a temporary effect on subjective urinary continence at 4 months but not at 8 and 12 months. More patients receiving aRT reported significant bowel symptoms at 4 and 8 months than those with RRP only, but at 1 year most of these symptoms had resolved and there were no significant differences between the groups.


The aim of adjuvant radiotherapy (aRT) after radical retropubic prostatectomy (RRP) is to eradicate any remaining cancer cells at the prostatic fossa, thereby improving local control and preventing later systemic tumour dissemination. Several unrandomized studies showed better local control rates and freedom from PSA progression for aRT after RRP in patients with pT3 tumours and/or margin-positive disease [1–5]. However, the exact role of aRT in improving overall survival rates is unclear, because there are no published randomized prospective studies. The South-western Oncology Group (study 9887/INT 0086) and the European Organization for Research and Treatment of Cancer (study 22911) are currently conducting randomized phase III trials to assess the role of aRT after RRP, and these studies are expected to provide useful information. With the survival benefit of aRT still under discussion the morbidity of this therapy should be tolerable. Furthermore, it is essential for fully informed consent that patients after RRP can be given realistic expectations about the likely complications of radiation.

In the present prospective study we used questionnaires to compare the differences and changes in urinary and bowel symptoms, and the need for further treatment in the first year after RRP, between patients with and with no moderate dose of aRT to the prostatic fossa.


The study included 96 patients who underwent RRP between March 1998 and June 1999. Exclusion criteria consisted of refusal to participate, a current history of urinary or fecal incontinence that required pads, and previous TURP or RT of the abdomino-pelvic region. Each patient was sent a questionnaire at 4, 8 and 12 months after RRP, including a covering letter and a prepaid envelope to return the questionnaire. Completed questionnaires had to be returned within 4 weeks and each patient had to complete each questionnaire at 4, 8 and 12 months after RRP to qualify for the analysis.

The aRT of the prostatic fossa was recommended to 36 (38%) men with microscopic positive margins and/or seminal vesicle involvement, the aRT being chosen by the patient with help from the doctors caring for him. Questionnaires from 30 men with aRT (response rate 83%) and 53 with no aRT (response rate 88%) were available for analysis at the end of the first year (overall response rate 86%). The demographic characteristics of all men according to treatment group are presented in Table 1. Patients were staged according to the fifth edition of the TNM classification of malignant tumours [6] and tumour grade determined using the WHO guidelines [7] and Gleason grade [8].

Table 1.  The patients’ characteristics
CharacteristicRRP onlyRRP + aRTTotal
  1. SVI, seminal vesicle invasion.

Response rate, %888386
No. with valid data533083
Mean (range) age, years65 (50–74)61 (45–73)63 (45–74)
Pathological stage, n (%)
pT246 (87)  4 (13)50
pT3  7 (13)24 (80)31
pT4  2 (7)  2
SVI (pT3b)12 (40)12
Positive margin24 (80)24
Gleason score
2–750 (94)25 (83)75
8–10  3 (6)  5 (17)  8
bilateral11 (21)  3 (10)14
unilateral15 (28)  8 (27)23
none27 (51)19 (63)46


All 96 patients underwent a bilateral dissection of the obturator lymph nodes followed by an anatomical RRP using a modified Walsh technique [9]. All operations were performed at the Department of Urology, Krankenhaus Muenchen-Bogenhausen. A nerve-sparing technique was attempted in patients who were potent before surgery, when possible without compromising the chance of surgical cure. The indwelling catheter was routinely removed at 10 days after RRP and patients given advice about pelvic floor exercises.


All patients in the aRT group were assessed using CT to assist in treatment planning; aRT was started at a median (range) of 12 (6–26) weeks after RRP, contingent on the recovery of urinary continence. The aRT was delivered at three different centres using linear accelerator-generated photon beams (10 or 18 MV photons) using a four-field (‘box’) technique. The pelvic lymph nodes were not irradiated. The median (mean, range) total dose was 54 (53.4, 49–61) Gy; all patients were treated for 5 days per week at 1.8–2.0 Gy per fraction.


Each patient was questioned about their urinary incontinence and bowel symptoms before RRP, and the results recorded. At 4, 8 and 12 months after RRP all patients were sent an anonymous questionnaire (in German). The section of the questionnaire that assessed urinary symptoms and subjective urinary incontinence was mainly based on questions reported by Herr [10], but modified, e.g. the question ‘In which situations do you leak?’ was changed to more specific questions (patients had to tick the appropriate answer: yes/no or complete the numbers): ‘Does urine leak when you cough or sneeze?’ (stress incontinence, SI, grade I); ‘Does urine leak when you stand up or walk around?’ (SI grade II); ‘Does urine leak when you are lying down?’ (SI grade III).

Further, two questions about symptoms of urinary urge incontinence and urinary frequency were added to the questionnaire, i.e. ‘Does urine leak as a result of a sudden, irresistible urge to pass water? ‘Do you have an increased urge to pass urine?’ (Urinary frequency was considered abnormal when patients had increased urge to pass water more then seven times in 24 h); if yes, ‘how often do you have this sensation per day and per night?’.

Questions to assess subjective bowel symptoms were: ‘Urgency to pass stool?,’‘Presence of stool incontinence?’, ‘Presence of soft stool (diarrhoea)?’ (soft stool was considered as abnormal when combined with a stool frequency of more than three times in 24 h and if symptoms persisted for ≥ 7 days); and if yes, ‘over which period of time (. . . days, . . . weeks, . . . months)?’ or if yes, ‘give stool frequency per day?’.

The questions to assess operations/procedures after RRP or RRP and aRT were: ‘Did you have to undergo any operation or manipulation because of problems (associated with your earlier prostate surgery or radiotherapy)?’; if yes, ‘dilatation of the urethra, urethrotomy, implantation of an artificial urinary sphincter, other (please specify)’; if no, ‘is one of the following procedures planned?’‘dilatation of the urethra, urethrotomy, implantation of an artificial urinary sphincter, other (please specify)’.

The aRT and RRP-only groups were compared using the chi-square test or Fisher's exact test for small samples.


Four (8%) patients in the RRP and one (3%) in the aRT group reported mild occasional urge incontinence before RRP that required no pads. The remaining 78 (94%) patients were subjectively dry before surgery. At 4 months after RRP, when patients in the aRT group were having or had just finished treatment, 28 (34%) regarded themselves as continent of urine, with no significant difference between the groups (34% vs 33%). As shown in Table 2, the percentage of patients who were continent or had symptoms of urinary SI grade I or II (no patient had grade III) did not differ significantly in the two groups at 4 months. However, men in the RRP group were less likely to need pads at 4 months than those undergoing aRT (P < 0.05) and in the aRT group a significantly higher proportion needed 1 pad/day than in the RRP group. In contrast the proportion of men using two or more pads per day was very similar for both groups. At 8 and 12 months after treatment urinary continence in both groups had improved and the differences between the aRT and RRP was not significant for type of urinary continence and the number of pads used. In association with the higher proportion of urge and combined SI/urge in irradiated men there was a trend towards a higher incidence of increased urinary urgency combined with frequent urge to pass urine (> 7/24 h) for the aRT than the RRP group at 4, 8 and 12 months (P > 0.05).

Table 2.  The type of urinary incontinence, use of pads and bowel symptoms before and at 4, 8 and 12 months after RRP and aRT
Variable, n (%) of patientsRRP onlyRRP/aRT
  • *

    P  < 0.01 vs RRP

Continent/SI GI49 (92)37 (70)42 (79)43 (81)29 (97)19 (63)24 (80)23 (77)
SI GII  9 (17)  7 (13)  7 (13)3 (10)2 (7)4 (13)
Urge or SI/urge  4 (8)  7 (13)4 (8)  3 (6)  1 (3)8 (27)4 (13)3 (10)
Use of pads/day
053 (100)36 (68)42 (79)42 (79)30 (100)11 (37)*21 (70)20 (67)
1  8 (15)  5 (9)  6 (11)14 (47)*  7 (23)  5 (17)
≥  2  9 (17)  6 (11)  5 (9)  5 (17)  2 (7)  5 (17)
Bowel symptoms and fecal incontinence
Stool urgency  1 (2)  1 (2)  2 (4)  2 (4)16 (53)*  8 (27)*  3 (10)
Soft stool  1 (2)  6 (20)*  2 (7)  1 (3)
Stool incontinence  4 (13)*  2 (7)  1 (3)


One patient in the RRP group had stool urgency before surgery; none of the patients assessed reported fecal incontinence or a relevant history of prolonged (> 1 week) soft stool consistency before surgery (Table 2). All the RRP patients were continent of stool during the year of follow-up. The reported incidence of soft stool consistency (combined with stool frequency of > 3/day for at least 1 week) and stool urgency in the latter group was 0–4% for the period assessed. In contrast, 4 months after RRP, when patients in the aRT group were receiving irradiation or the treatment had recently been completed, there was a considerable increase in the incidence of stool urgency (53%), fecal incontinence (13%) and soft stool consistency (20%), and the differences between the treatment groups for all three were significant (Table 2). At 8 and 12 months after surgery there was an important decrease in bowel symptoms in the aRT group, with no significant differences in the three symptoms between the groups at 1 year.

One patient each in the RRP (2%) and aRT group (3%) reported a urethral dilatation at 8–12 months after RRP. No further operations or procedures (planned or undertaken) related to the effects of RRP and/or aRT were reported.

To determine the effect of an early or late start of aRT on urinary and bowel symptoms we compared men who had their radiation started within 12 weeks of RRP (14) with those starting later (> 12 weeks; 16 men). There was no significant difference in subjective urinary incontinence, fecal incontinence and bowel symptoms between these groups in the first year.


Of men undergoing RRP 30–60% will have tumour penetrating the prostatic capsule, margin-positive disease or seminal vesicle involvement. Although many of these men are likely to relapse biochemically and/or locally, there is no uniform adjuvant therapeutic strategy for patients with adverse prognostic indicators after RRP. Thus aRT offers an effective approach to reduce local recurrence rates and to improve the rate of biochemical freedom from relapse in these patients. However, any potential effect of aRT on disease-free survival remains to be confirmed, and the routine use of aRT in every patient with adverse prognostic factors after RRP could expose a significant number to over-treatment, as surgery alone may cure a significant proportion of them, particularly in the presence of low-risk factors for tumour recurrence (e.g. PSA < 10 ng/mL, Gleason sum < 7 and limited capsular perforation [11]. While there are no indications for aRT based on the results of major prospective studies, knowing the possible side-effects of aRT is of particular interest when deciding about this therapy. Although the incidence of urinary incontinence, impotence and bowel complications were reported by other, mainly retrospective, studies on patients after RRP and aRT, the reported rates of these functional problems are inconsistent [3,5,12–14]. Apart from differences in the total dose (45–70 Gy) and in the daily fraction (1.8–3.1 Gy) applied to the prostatic fossa, these variations could be a result of the different ways that these symptoms and disorders were assessed (e.g. telephone interview, questionnaire, clinical examination) in the studies. Furthermore, none of these studies compared groups of irradiated and unirradiated men after RRP to assess the potential effect of aRT on patients’ symptoms. To address these limitations the present prospective study of urinary and bowel symptoms, including irradiated and unirradiated men after RRP, used self-reported questionnaires at 4-month intervals in the first year. The present radiation technique was based on a moderate dose of aRT, as described previously [4,12,15]. A study with 622 patients showed that a radiation dose of 45–54 Gy to the prostatic fossa of patients with pT3N0 prostate cancer reduced the incidence of local recurrences and improved disease-free survival to equal that of a lower-risk group of patients treated with RRP alone [4].

In numerous publications, patient self-reported questionnaires have been confirmed as a valid method of obtaining information about patients’ symptoms and the results of surgery. Litwin et al.[16] showed that physicians’ ratings of symptoms in patients with prostate cancer of various stages did not correlate well with patient self-assessment, and tended to underestimate the effect on the patient's quality of life. Similarly, others [17,18] confirmed that the patients’ assessment of the side-effects of prostate cancer treatment and its effect on quality of life were generally more accurate than an evaluation by the physician. Furthermore, the role of questionnaires as a sensitive instrument for assessing urinary incontinence is supported by a recent study [19] comparing subjective and objective urinary continence 12 months after RRP. Donnellan et al.[19] found that the proportion of patients regarding themselves as continent (by questionnaire, 80%) and who were also dry according to a pad test (84%), were very similar.

As in the results of a prospective randomized study [20], the present data could not be used to confirm that aRT had a relevant influence on urinary continence after RRP. Van Cangh et al.[20] used both a pad test and a questionnaire to assess urinary continence (bowel symptoms were not assessed) and found that 77% of irradiated (median radiation dose 60 Gy) and 83% unirradiated patients were not using pads and were dry after a mean follow-up of 2 years. Similarly, the present results showed that two-thirds of the irradiated and 79% of the unirradiated patients needed no pads at 1 year after RRP. Although not significantly different, both studies showed a trend to more men having moderate or severe urinary incontinence after aRT than in those with RRP only, after 1 or 2 years. These findings might indicate some intermediate and late toxicity of aRT on the external urethral sphincter. In the present study the increase from 7% to 17% of men needing two pads/day (P > 0.05) between 8 and 12 months after RPP might indicate an effect of radiation-induced toxicity on the urinary sphincter.

Using a moderate dose of aRT, as used here, Petrovich et al.[15] assessed urinary continence by interval history and physical examination, reporting that 87% of 201 irradiated men were continent or had mild SI at 1 year after RRP. The results were similar to the present rate of 77% for complete continence or mild SI in the aRT group at 1 year. As in recent reports [12,20] which contradict earlier findings that early aRT could expose patients to a greater risk of incontinence, the present data showed no significant difference in urinary continence between patients with early or late aRT.

The most important problems after aRT in the present study were fecal incontinence and stool urgency during and some time afterwards. Most studies report a 10–38% rate of mild to moderate (RTOG grade 1/2) and a 0% to 4% rate of significant (RTOG grade 3/4) gastrointestinal complications in patients who underwent aRT [3,5,14,15,20]. As shown by the present data, bothersome gastrointestinal symptoms of all grades tended to be mainly temporary and decreased during the follow-up. A possible explanation for the discrepancy between the present data at 4 months after surgery (53% stool urgency and 13% fecal incontinence) and values in the other studies might include that gastrointestinal symptoms are transient side-effects during and some time after radiotherapy, and therefore underrated by studies with long intervals between aRT and patient assessment. At 1 year after surgery most gastrointestinal symptoms had subsided and only one irradiated patient (3%) had fecal incontinence (RTOG grade 3; Table 2).

As reported [21] the present urethral stricture rate was not significantly affected by radiation; one patient in each group developed a urethral stricture during the follow-up. Apart from urethral dilatation in these two patients no other procedures were needed for treatment-related problems. Despite a prospective design the present study has several deficiencies. First, patients were not randomized; the treatment was chosen by the patient and his physician, and therefore there were differences among the groups in age and pT stage. Second, there were relatively few patients; although there were enough patients in each group (53 and 30) for a sufficient statistical analysis, the ability to detect differences between further subgroups (e.g. side-effects in patients with late vs early aRT) was limited. Third, the lack of questions covering quality-of-life aspects might also be criticised. This point is valid, but must be weighed against our intention to minimize the number of questions to reduce the burden on the participant.

In conclusion, the results of this prospective analysis provide further evidence that a moderate dose of aRT has no relevant effect on urinary continence or urethral stricture rate in the first year after RRP. However, men considered for aRT after RRP should be informed that a transient negative effect on urinary continence and temporary irritative bowel symptoms are likely, and that ≈ 15% of patients will have transient fecal incontinence as a further side-effect of radiation.


radical retropubic prostatectomy


adjuvant radiotherapy


urinary stress incontinence.