BJU International (1999), 83, 57–65
Urinary incontinence after radical prostatectomy: a
randomized controlled trial comparing pelvic muscle exercises
with or without electrical stimulation
K.N. M OORE , D. GR IFFI THS* and A. HUGHTON†
King’s College London, Waterloo Road, London, *Urodynamics & Northern Alberta Continence Service, Caritas Health, Misericordia
Hospital, Edmonton, Alberta, and †Northtown Physiotherapy, Edmonton, Alberta, Canada
Objectives To assess the eCectiveness of intensive conservative treatment on and the impact of urinary
incontinence after radical retropubic prostatectomy.
Patients and methods Sixty-three men with urinary
incontinence 8 weeks after radical prostatectomy
were randomized to one of three groups; group 1,
standard treatment (control); group 2, intensive pelvic
muscle exercises (PME); or group 3, PME plus electrical
stimulation (PME+ES). Group 1 received verbal and
written instructions about postoperative PME from
their urologist and from the nurses at the preadmission clinic. Groups 2 and 3 were treated by a
physiotherapist for 30 min twice a week for 12 weeks
and carried out home exercises three times a day on
the days when they were not treated. Outcome was
assessed using the 24-h pad test, two validated qualityof-life questionnaires and a urine symptom inventory,
all obtained at baseline, 12, 16 and 24 weeks after
enrolment. The final pad test was carried out
#8 months after surgery.
Results Fifty-eight patients completed the study, 21 in
group 1, 18 in group 2 and 19 in group 3; five
discontinued, three with bladder neck contractures
requiring dilatation, one with rectal pain when doing
the exercises and one unable to complete therapy
Introduction
Urinary incontinence after radical prostatectomy (RP) is
a common and potentially devastating problem aCecting
between 6% [1] and 63% [2] of men 6 months after
surgery. Discrepancies in the reported incidence occur
because of variation in outcome measures, various
follow-up periods and varying definitions of continence.
Discrepancies also occur in the reported eBcacy of
conservative treatment of incontinence after RP [3–5]
because trials are often not randomized or controlled.
Electrical stimulation (ES) is one method which may
Accepted for publication 18 August 1998
© 1999 BJU International
while on vacation. The mean (median) time elapsed
from surgery to entry into the study was 19 (8) weeks.
At 12 weeks from baseline, the mean overall urine
loss had decreased from 463 g to 115 g but there
were no diCerences among groups, nor were there
significant diCerences in urine loss at 16 and 24 weeks
(F=0.16, P=0.69). There was a significant impact
on quality of life during the early recovery. Despite
preoperative instructions, many patients revealed little
or no knowledge about catheter care, bladder spasms,
rectal pain, incontinence and erectile dysfunction.
Little of the preoperative education was retained
because of the overwhelming nature of the diagnosis.
Conclusions From the initial assessment to the final padtest at #8 months after surgery, incontinence
improved greatly in all three groups. This rapid
improvement may have masked any treatment benefit.
Further research should address incontinence in men
whose urine loss has stabilized and who underwent
surgery >8 months previously. Moreover, a telephone-based follow-up soon after discharge may alleviate many of the concerns expressed.
Keywords Incontinence, radical prostatectomy, quality
of life
enhance the success of pelvic muscle exercise (PME) in
patients with incontinence after RP [6–8]. While PME
represent ordinary muscle contraction, ES artificially
stimulates the pudendal nerve and its branches to cause
direct and reflex responses of the urethral and periurethral striated muscles in women [9,10]. Similar results
are not available for men. Although the Agency for
Health Care Policy and Research [11] recommends
behavioural interventions as first-line treatment for
incontinence, pelvic muscle stimulation for men is not
yet considered a standard treatment for incontinence
because of the absence of supportive research.
To determine whether PME and ES are eCective therapies for incontinence after RP, a randomized, controlled
57
58
K.N. M OOR E et al.
trial was conducted comparing intensive PME with or
without ES in men 8 weeks after surgery. The intent
was to study the objective and subjective changes in
patients’ urinary symptoms before and after 12 weeks
of a twice weekly intensive treatment. Two hypotheses
were tested: (i) men who receive treatment would achieve
continence more quickly by 12 weeks from baseline than
their untreated counterparts; and (ii) men who received
PME+ES would achieve continence more quickly by
12 weeks of therapy than men receiving PME alone.
Patients and methods
A total of 180 radical retropubic prostatectomies were
performed at three university-aBliated hospitals in
Edmonton, Canada during the study period of December
1995 to February 1997. Sixty-three men 8 weeks
after RP met the eligibility criteria (Table 1) and were
randomized to control (group 1), intensive PME (group
2) or intensive PME+ES (group 3). All therapy was
delivered by one experienced physiotherapist. Treatment
options were reviewed and all patients signed informed
consent. The study protocol was approved by the
University of Alberta and Caritas Health Group ethics
review boards.
Initial assessment
All patients underwent a physical examination and had
a complete history taken, including previous urological
symptoms such as frequency, urgency, nocturia, or
incontinence. The physical examination included a
neurological assessment of perianal sensation, anal
sphincter tone and control, bulbocavernosus reflex and
a brief screening for any neurological factors such as
diabetes, Parkinson’s disease, multiple sclerosis or stroke.
Patients completed a general urology symptom inventory
(adapted from Herr [12]) and two validated quality-oflife (QoL) questionnaires (the Incontinence Impact
Questionnaire, IIQ-7, and the EORTC QLQ C30, version
Table 1 Inclusion and exclusion criteria
Inclusion
Exclusion
4 weeks after RP
Urinary incontinence
(>2 g of urine loss on pad test)
Neurologically normal
Within 2 h drive of study centre
Able to speak and read English
Willing to comply with protocol
No current treatment
Not seeking other treatment
Demand pacemaker
Previous pelvic muscle
stimulation
Active rectal lesions or
infections
Known detrusor instability
2), and were instructed in the completion of the 24-h
micturition chart and pad-test.
Patients were randomized, using a computergenerated random-number list placed in sealed envelopes, by the researcher at the end of the assessment
visit, with patient and researcher opening the sealed
envelope. The study was designed to have an 80% power
to detect a 35% diCerence in improvement rates for a
two-tailed test, with 5% type I error. Adherence to
treatment protocol was monitored using patient-recorded
diaries provided at the initial visit with the
physiotherapist.
Outcome assessment
Data were obtained at baseline, and again at 12, 16 and
24 weeks after enrolment. Primary outcome variables
selected a priori were the urine lost during the 24-h padtest, the number of incontinent episodes, the number
and volume of voids and the number of pads used.
Continence was defined as a loss of ∏2 g of urine;
socially acceptable continence was considered as ∏10 g.
Objective QoL measures (IIQ-7 and EORTC QLQ C-30)
and subjective impressions were also recorded.
The 24-h pad-test is a standardized test using preweighed pads changed and weighed every 2 h, with a
chart of the frequency, volume and time of voids and
the amount and type of fluid intake. The pad test is
capable of detecting a 1 g urine loss if the pads are
weighed within 36 h of their placement in a sealed
plastic bag [2].
The patients’ verbal responses were compared with
the objective assessments self-recorded on the IIQ-7 and
the EORTC QLQ C30. The IIQ-7 addresses the impact of
symptoms associated with urinary incontinence and has
four subscales, i.e. physical activity, travel, social/
relationships and emotional health. It is a single, summary scale of seven items with a score ranging from 0
to 100, where 0 denotes ‘No impact’ and 100 ‘Significant
impact’ of incontinence [13]. Although tested on community dwelling women, face validity suggests that the
questions may be appropriate for men. The EORTC QLQ
C30 was designed to record ‘quality of life factors’ which
may assess the impact of cancer and its treatment [14];
28 items are designed to evaluate overall quality of life,
fatigue, pain, other physical symptoms, cognitive abilities, emotional distress and interference with family or
social life. The answers to individual questions are scored
on a four-point Likert Scale (‘Not at all’, ‘A little’, ‘Quite
a bit’, ‘Very much’). Two separate questions asked about
health and global quality of life. Patients were asked to
rate these two questions on a seven-point scale with ‘1’
being ‘Very poor’ and ‘7’ being ‘Excellent’. The QLQ C30
has shown high levels of reliability and validity among
© 1999 BJU International 83, 57–65
URINARY INCONTINENCE AFTER RADICAL PROSTAT ECTOMY
cancer groups, but has not been used extensively on
patients after RP.
Group assignment
Patients randomized to group 1 (standard treatment)
received the usual instructions to conduct PME after RP,
which included simple written and brief verbal instructions on PME by both the nurses in the pre-admission
clinic and the patient’s urologist at postoperative visits.
After the initial assessment, Group 1 had no further
contact with the investigator until the follow-up visits
at 12, 16 and 24 weeks. Patients in group 2 received
the standard treatment (as above) plus intensive physiotherapy 30 min twice a week for 12 weeks. For this,
patients were supine with the knees bent, and the
physiotherapist seated at the patient’s right side. Four
components of muscle function were exercised, i.e.
strength, endurance, speed, and control. Strength exercises focused on recruiting as many muscle fibres as
possible, maintaining the contraction at an optimum
hold, while keeping the accessory muscles relaxed. Initial
contractions were of 5–10 s with a 10–20 s rest, with
12–20 repetitions. Endurance exercises focused on maintaining a muscle contraction at 65–75% of maximum
strength. The ‘hold’ time was 20–30 s with an equal
rest time, with 8–10 repetitions. Speed was achieved by
sets of quick repetitive contractions in a 10-s span with
a 20-s rest. Finally, purposeful control involved gradual
recruitment to a maximal contraction and occurred in
three stages, with a 5-s hold at each stage and a slow
release, with a rest period of 15–30 s.
Patients in group 3 also met with the same physiotherapist twice a week for 30 min. Electrical stimulation
with a surface anal electrode (InCareTM) was alternated
with PME, carried out as in group 2. The intensity was
adequate to induce visual lifting of the levator ani and
pubococcygeus muscle, considering the level of comfort
of the patients [9]. Stimulation parameters were 50 Hz,
a biphasic pulse shape with 1-s bursts, a 1 s pulse width
and 1 s pulse trains.
During treatment patients were monitored manually
for abdominal and gluteal muscle relaxation, and were
instructed to monitor themselves at home when they
repeated the series of exercises. They were also instructed
to carry out a penile lift, by contracting the bulbocavernosus muscle, to watch for a visible dip at the angle
at the base of the penis and to palpate for a contraction
of the perineum. Adherence to the protocol was
reinforced by a check-list exercise reminder completed
by the patient.
The results were analysed statistically using multivariate techniques with natural log-transformed data. There
were no estimates for missing values. Data were analysed
© 1999 BJU International 83, 57–65
59
for the 58 patients with complete data. Descriptive
statistics, frequencies, and t-tests were used to compare
diCerences between untreated and treated groups at
12 weeks from baseline. A one-way repeated-measures
anova using a general linear model was computed to
test the diCerence between and within groups, as well
as the change over time at 12, 16 and 24 weeks.
Multiple analysis of covariance was used to explore the
eCect of time after RP and the age of patients on urinary
continence.
Results
The mean age of the patients was 67 years (range
49–77); all were white, 75% were retired, 95% were
married and 26% were from small towns or rural
communities. Sixteen (28%) had < Grade 8 education.
There were no significant group diCerences at baseline
in education, past medical or surgical history, or medication use. All eligible patients agreed to participate; five
were dropped (three in group 3 and two in group 2,
three because of bladder neck contractures, one because
of rectal pain when he did the exercises, and one because
he went on vacation for 4 months and could not
continue therapy). Data were available for 58 patients;
their complete demographic details are given in Table 2.
Previous surgical procedures included coronary artery
bypass surgery in four (7%), TURP in four, TURBT in
one, and unrelated surgery in four. None of the men
had undergone back or pelvic surgery. Preoperative
LUTS of >3 months duration were reported by 30 of
56 patients (54%, missing data on two). Of these 30
patients, 14 had nocturia alone, 12 had urgency, frequency and nocturia, and four had preoperative retention. The remaining 26 stated they had no presenting
symptoms at the time of diagnosis of prostate cancer.
In terms of incontinence at 24 weeks, of the men with
nocturia alone, 11 of 14 had <10 g urine loss in 24 h,
whereas only four of 12 men with frequency, urgency
and nocturia, and one of four with retention had <10 g
of loss. By comparison, of the 26 men who stated they
had no presenting symptoms, 16 of 25 (64%) had a
urine loss of <10 g at 24 weeks. Only the four men
with retention had sought help directly related to their
urological problems. The others had an abnormal DRE
or raised level of PSA after seeing the physician for an
unrelated problem or routine check-up. Six men had a
family history of prostate cancer and had been followed
with an annual PSA and DRE; a rise in PSA resulted in
a urological consultation.
At enrolment, the mean (median, range) time from
surgery was 18.9 (8.0, 4–241) weeks; four patients
were enrolled >2 years after RP and had urodynamically
confirmed stress urinary incontinence. The three groups
60
K.N. M OOR E et al.
Table 2 Demographic characteristics of the
patients
Group
Variable
Number
Mean age, years
Annual household income ($), n
<20 000
20 000–40 000
40 000–50 000
>50 000
Education grade
<8
8–12
12 completion
College/University
Employment
Full-time
Part-time
Retired
Description of current health
Very healthy
Mainly healthy
Somewhat healthy
Mainly unhealthy
1
(control)
2
(PME)
3
(PME+ES)
All
n (%)
21
66.8
18
67.4
19
65.7
58
6
9
4
2
3
9
2
5
2
12
3
1
11
30
9
8
(19)
(52)
(15)
(14)
11
2
5
3
2
4
5
8
2
7
4
5
15
13
14
16
(26)
(22)
(24)
(28)
3
1
17
4
1
14
2
4
12
9 (16)
6 (10)
43 (74)
5
14
1
0
6
8
5
0
5
10
2
1
16
32
8
1
were not homogeneous for time since surgery (Cochran’s
C, [18,3]=0.89, P<0.001; Bartlett-Box, F [2,67]=
43.40, P<0.001); one ‘outlier’ patient was in group 2
and three in group 3. With these patients temporarily
removed from analysis, group homogeneity was acceptable (Cochran’s C=0.53, P=0.07; Bartlett-Box P=
0.08).
The mean (sd, range) time to complete the study
protocol was 13 (3.2, 6–23) weeks; two men in group
2 and two in group 3 did not complete the full 24
sessions because they felt they could easily practise the
exercises at home. One of the men in group 2 complained
of rectal pain when doing the exercises and discontinued.
No others complained of adverse eCects from the therapy.
Sixteen men were older and 42 younger than 70 years;
there were no diCerences between these age groups in
total urine loss at 12 weeks (t=−1.32, P=0.09) or at
24 weeks (t=0.61, P=0.27). DiCerences were also not
significant when patients were categorized as >65 years
(n=48) or <65 years old (n=10), at 12 weeks (t=
0.50, P=0.31) or 24 weeks (t=−0.92, P=0.18).
The 24-h pad-tests were performed by all patients at
the four sample times; the mean urine loss is shown in
Table 3. There were no diCerences among the groups
(F=0.23, P=0.80) at any of the measurements; by 24
weeks, 28 (49%) had <10 g urine loss. Mean (range)
pad usage (adult diapers) decreased from 4 (1–8) at 12
weeks to 1.5 (0–5) at 24 weeks, with a concomitant
(28)
(56)
(14)
(2)
increase in volume and number of voids (mean 100 mL,
2.1 voids at 12 weeks, to a mean of 275 mL, 4.5 voids
at 24 weeks). Thus the first hypothesis, that men who
received early intensive intervention would improve
more quickly than those who did not participate in
active therapy, was not supported.
The second hypothesis, that men who received
PME+ES would improve more quickly than those who
received PME alone, also was not supported. There were
no diCerences at 12 weeks (t=1.42, d.f.=35, P=0.16),
or at any other sample time (anova F=0.16, P=0.69).
Although there were no diCerences among the groups,
time was highly significant (F=85.23, P=0.001;
Table 3).
Quality of life and the impact of incontinence
Objective measures of the impact of urinary incontinence
showed a moderate correlation between the IIQ-7 and
urine loss (P0.05). Pearson’s correlation coeBcient
ranged from 0.34 (P=0.003) at 12 weeks to 0.51 (P=
0.001) at 24 weeks. The QLQ C30 scores did not change
significantly over time on any of the five categories
(physical, emotional, role, cognitive or social functioning) including the two questions combined to make
a single score (‘How would you rate your overall health
during the past week?’ and ‘How would you rate your
overall quality of life during the past week?’). There were
© 1999 BJU International 83, 57–65
URINARY INCONTINENCE AFTER RADICAL PROSTAT ECTOMY
61
Table 3 Urine loss in 24 h with group and time after treatment
Mean (median) [sd, range] urine loss (g)
Group
Baseline
12 weeks
16 weeks
24 weeks
1 (control)
n=21
2 (PME)
n=18
3 (PME+ES)
n=19
Total
n=58
385.9 (395.5)
[256.9,6.3–921.5]
565.6 (513.9)
[403.3,21.5–1538.6]
452.5 (492.1)
[385.1,5.3–1344.8]
463.5 (419.8)
[352.2,5.3–1538.6]
103.8 (23.8)
[176.3,1.0–702.4]
86.9 (32.50)
[123.0,2.2–385.9]
155.5 (87.5)
[168.1,1.0–509.3]
115.5 (27.2)
[158.7,1.0–702.4]
67.3 (11.5)
[137.4,2.0–530.3]
73.5 (10.35)
[131.4,1.0–494.6]
202.2 (85.7)
[242.23,1.0–753.4]
114.2 (14.1)
[185.6,1.0–595.7]
54.1 (6.9)
[103.1,1.0–277.3]
69.9 (8.7)
[113.5,1.0–362.8]
98.2 (8.95)
[132.1,1.0–424.2]
72.5 (7.5)
[115.7,1.0–424.2]
no significant group diCerences in either the IIQ-7 or the
QLQ C30.
Four questions of the Symptom Inventory addressed
the bother of urinary incontinence and erectile dysfunction. Question 1 asked ‘Does urine leakage aCect your
life?’. At 12 weeks, as expected, 72% of patients reported
that their lives were aCected ‘moderately’ or ‘very much’.
By 24 weeks, with improved continence, only a quarter
of patients reported that incontinence aCected their lives
(three in group 1, three in group 2 and six in group 3).
Question 2 asked ‘Would it bother you if you had to
spend the rest of your life with urine control the way it
is now?’. At baseline, 84% replied that they would be
‘very bothered’; however, at 24 weeks, 61% of patients
were still ‘moderately’ or ‘very bothered’ with their
current bladder control. Questions 3 and 4 asked about
erectile dysfunction, ‘Does erectile dysfunction (impotence) aCect your life?’ and ‘Would it bother you if you
had to spend the rest of your life with erections the way
they are now?’. At 12 weeks, to both questions, 53% of
patients responded ‘moderately’ or ‘very much’; by 24
weeks, there was an increase, with 61% reporting that
the eCect of and bother from erectile dysfunction was
‘moderate’ or ‘very much’. There were no significant
group diCerences for any of the questions (F=0.40,
P=0.68).
Subjective reports of the impact of urinary incontinence revealed several unexpected postoperative concerns which had an impact during the early months of
recovery. From the day of discharge until #4 months
after surgery, these concerns fell into four main groups,
coping with catheter care, pain control, incontinence
after catheter removal and erectile dysfunction.
Preoperative teaching by both the urologist and preadmission clinic nurses covered these categories, but
patients stated that their thoughts focused on the word
‘cancer’ and all that it entailed. For patients, the sideeCects or sequelae of surgery had been a distant issue at
the time of diagnosis and surgery. As one man candidly
© 1999 BJU International 83, 57–65
reported, ‘Even though the physician spent a long time
with me and answered all my questions before surgery,
the only thing I heard was cancer.’
Catheter pain and pain control
All patients were discharged with an indwelling catheter
which was removed #21 days after surgery. Bladder
spasms, by-passing and haematuria were all problematic
and, despite written information being provided on catheter care, most stated that they were prepared for neither
the discomfort nor the care of the catheter and drainage
bag. Rectal pain was graphically described; ‘felt like I
had been kicked in the rear end with a cowboy boot’.
Visits to the emergency department occurred because of
bladder spasms, bypassing or haematuria. Issues raised
during these phases were expressed by all patients,
although the 15 men and their spouses who received
visits by the home-care nurses expressed fewer concerns
than those who did not receive such visits.
Catheter removal
The degree of incontinence after catheter removal was
a shock to all patients and devastating to some; no
patient felt adequately prepared. On a material level,
patients had little knowledge about incontinence products or where to obtain them, nor did they understand
that gross incontinence, lasting for a few days to several
weeks, was to be expected after RP. Patients stated that
the urologist may have told them about incontinence,
but they had been unable to conceptualize what the
word incontinent really meant. Patients expressed feelings
of loss of control, fear of odour, self-consciousness,
embarrassment and shame.
By 24 weeks after surgery, the men who had not
regained most of their bladder control were upset, disappointed and despondent. However, those who had only
a small amount of leakage and who required a pad ‘just
62
K.N. M OOR E et al.
in case’ had adjusted to the inconvenience of leakage.
These men were ready to continue with life, but erectile
dysfunction interfered qualitatively with life as they
perceived it. More than one claimed specifically ‘I’ve lost
my manhood’. Although they wanted to be treated and
privately hoped that the urologist would raise the topic
of erectile dysfunction, they were self-conscious and
reluctant to ask for help, rationalizing the satisfaction of
curative cancer surgery as adequate in itself. As a result,
they suCered erectile dysfunction in dissatisfied silence.
Despite these concerns, at 24 weeks (#8 months after
surgery) when asked ‘If you had to do it again, would
you choose radical prostatectomy as treatment for your
prostate cancer?’ (1=No to 4=Yes), all patients
responded positively (mean response 3.4); no patient
said ‘No’. Moreover, there were no diCerences between
continent and incontinent men.
Discussion
Young et al. [15] were the first urologists to recommend
exercising the ‘sphincter muscles by cutting oC the flow
of urine and retaining urine as long as possible’ for men
with incontinence after RP. To date, the treatment of
incontinence after RP continues to focus on exercising
the pubococcygeus and levator ani muscles, therapy
which has worked well in women with stress urinary
incontinence. However, it is unclear whether the theoretical basis of PME in women can be applied directly to
men. The normal male continence mechanism diCers
from that in the female, notwithstanding the impairment
which occurs after RP. Nevertheless, there are few
alternatives and PME remain the mainstay of treatment
for incontinence after RP.
The first hypothesis proposed that early intervention
with treatment would accelerate the return to continence
after RP. Both treatment and control groups improved
similarly, with improvement unaCected by such factors
as time after RP or age at the time of surgery. This lack
of treatment eCect has not been reported previously and
was unexpected. Based on the work of Burgio et al. [16],
who successfully treated men after prostatectomy using
biofeedback and PME, it was anticipated that patients
who were treated would improve by at least half over
those in the control group. However, the study by Burgio
et al. diCered from the present study in some important
ways, e.g. advice on urge suppression and fluid management was part of the therapy, there was no control
group and men were treated 6 months after surgery.
Furthermore, Burgio et al. used biofeedback-assisted PME.
It is possible that these behavioural interventions
improved the outcome of therapy. Such factors should
be considered in future research on incontinence after
prostatectomy.
The second hypothesis tested whether ES and PME
together would enhance treatment eBcacy. Stimulation
did not change the degree of incontinence after 12 weeks
of treatment, nor did it change the rate of improvement
over time. Similar preliminary results have recently been
presented on a randomized, blinded, sham-control study
of ES in men after prostatectomy [17]. Thirty-nine men
were randomized to either treatment or sham stimulation
for 15 min daily for 12 weeks at home, using a portable
home unit. Thirty-two of the men improved, including
those using the sham device.
To control for the eCect of time and gross incontinence,
and to examine diCerences between patients for whom
the variations were smaller, two subsets were analysed
separately; patients who were treated 12 weeks after
surgery and those with >100 g and <800 g urine loss
at baseline. Group diCerences were not significant for
either of these subsets.
Of note was the rapid improvement in bladder control,
particularly from baseline to 12 weeks (median 20 weeks
after RP), a finding also reported by others [18]. By
2 weeks after catheter removal, only a third of the 180
men who were contacted for inclusion in the study
considered themselves incontinent enough to be potential
subjects. By 12 weeks, of the 63 men recruited, 34% of
the sample had <10 g of urine loss. At the final pad test
(#8 months after RP) 49% of the sample had <10 g of
loss. Thus, in the sample under study, at a mean of
8 months after RP, only 15% had urine leakage of >10 g.
It is this relatively small proportion at >8 months after
RP to whom further research should be directed.
The early recovery period was decidedly stressful
because of a perceived lack of knowledge about normal
postoperative events and a perceived lack of professional
healthcare support. The objective measures of QoL, the
IIQ-7 and the QLQ C30, did not provide an assessment
of the individual concerns after surgery. While the IIQ-7
was modestly correlated with urine loss, the QLQ C30
scores showed little relationship with urine loss or with
time. Similar findings of a weak correlation between a
generic QoL instrument (the Sickness Impact Profile) and
amount of urine leakage has been reported in females
[19]. Those authors proposed that it was the presence
of incontinence per se which aCected quality of life, and
that only cure of the problem would improve quality of
life, not just a reduction in the number of leaks.
In the present study, the timing of the initial questionnaire may have aCected the responses. At baseline, most
of the respondents were <8 weeks after surgery and
were still recovering physical and role functioning. Also,
patients believed they had undergone curative cancer
surgery. Despite gross incontinence, their perceived relief
at being cured of a life-threatening illness may have
lessened the impact of urinary incontinence on their
© 1999 BJU International 83, 57–65
URINARY INCONTINENCE AFTER RADICAL PROSTAT ECTOMY
reported quality of life. As time progressed, the patients’
attitudes toward being incontinent may have changed.
At 8 months after surgery, with less urine loss but still
requiring a pad, men may have rated quality of life as
relatively poor. Herr [12] provided some evidence for a
change of attitude over time in men with incontinence
after RP. Despite significant incontinence, 83% of subjects at ∏3 years after surgery reported that they would
choose RP again, whereas only 47% of those >5 years
after RP would choose surgery. Further follow-up of the
present patients at 1 and 2 years may add support to
Herr’s finding.
One of the most poignant findings in this study and
reported in more detail elsewhere [20] was the degree
of distress experienced by patients in the early recovery
period. Early discharge (3 or 4 days after RP) and sameday admission provide limited opportunities for nurses
and physicians to prepare the patient and his family for
postoperative events. However, the most frequently
stated concerns involved those related specifically to the
indwelling catheter, bladder spasms, rectal pain and
fatigue. After catheter removal, concerns focused on the
inevitable, albeit gradually resolving, urinary incontinence. The patients’ concerns related to the timing of
and inability to process the information conveyed to them
during preoperative discussions with both urologists and
nurses. Notably, patients randomized to therapy all
expressed appreciation for the support and encouragement which the physiotherapist provided, even when the
therapy itself did not significantly improve the incontinence, and those who were visited by the home-care
nurse had fewer questions about catheter care.
Clearly, supportive intervention is worthwhile but the
form of that support requires further investigation. The
possibilities immediately after surgery are several. The
present patients would have benefited greatly from written information which described the normal events after
RP and from nursing follow-up immediately after discharge. Telephone follow-up by a nurse presents an
inexpensive and eCective method of providing patients
with information and assistance. Improved ability to
communicate with the physician, an improved understanding of treatment-related stressors, a better understanding about the risks and side-eCects of treatment are
all positive outcomes from telephone follow-up in
oncology patients [21,22].
Several risk factors have been suggested for incontinence after RP, most significantly age and detrusor
dysfunction [23]. Age as a risk factor [24] may be
explained partly because detrusor instability, decreased
compliance and decreased contractility are all more
common in the elderly [25–28]. However, age was not
a significant factor in the present patients; no diCerence
was detected in urine loss at baseline and any of the
© 1999 BJU International 83, 57–65
63
four sampling times between men older or younger than
65 years or 70 years.
Detrusor instability, poor compliance and decreased
contractility have also been implicated in incontinence
after RP [29,30]. In the present study, over half of the
patients (30 of 58) described preoperative LUTS
(urgency, frequency, and/or nocturia). Notably, only
three of the 14 men with nocturia alone had persistent
incontinence, whereas eight of 12 of those with preoperative frequency, urgency and nocturia had persistent
incontinence. Including a systematic evaluation of preoperative symptoms with a standardized scale such as the
AUA symptom score may help identify men with LUTS.
If preoperative symptoms correlate with a higher risk of
urinary incontinence after surgery, then patients could
be counselled accordingly.
The present study had some weaknesses which should
be recognized: foremost was the initial rapid improvement, particularly notable from initial assessment to 12
weeks (#8 weeks to 20 weeks after surgery), thus
masking any treatment eCect. Second, the subjects were
self-selected, so that the incidence of incontinence may
have been higher than actually admitted, despite specific
questions by the researcher at the initial telephone
interview on the number of pads being used, the improvement since catheter removal and the perceived bother
of incontinence. Third, the 24-h pad-test is at best only
an estimate of urine loss. While every attempt was made
to keep activity standardized for each of the four tests, it
is possible that increased activity increased the amount
of leakage. Fourth, the incorporation of behavioural
strategies such as urge suppression and fluid management into the treatment protocol may have enhanced
the results. However, the introduction of several more
variables would have necessitated more groups and a
larger sample size. Finally, incorporation of biofeedbackassisted PME may have improved the outcome, allowing
for objective assessment of muscle strength, endurance
and control while maintaining relaxation of the abdominal and gluteal muscles. As it was, the therapist’s
interpretation may have been prone to subjectivity and
imprecision. However, neither the role of biofeedback in
the treatment of male incontinence nor measurement of
pelvic muscle strength in men have been clearly defined.
To date, no studies in men have been conducted illustrating a clear benefit of biofeedback over PME. The results
are also equivocal in women; while some report eCectiveness [31,32] others have been unable to detect a
significant benefit over PME alone [33,34]. However,
despite these limitations, the present study is the first
randomized, controlled trial evaluating the treatment of
men after RP and should provide the starting point for
further systematic evaluation of this debilitating consequence of surgery.
64
K.N. M OOR E et al.
Acknowledgements
Dr Moore’s studies were funded for 4 years by a Doctoral
Fellowship from the Kidney Foundation of Canada.
Funding for the research project was received from the
Oncology Nurses’ Society, Canadian Nurses’ Foundation,
Caritas Health, Alberta Physiotherapy Association, Edna
Minton Foundation, and the University of Alberta,
Edmonton, Canada. In 1998, Dr Moore was a
Leverhulme Trust postdoctoral fellow at King’s College
London. Statistical advice was provided by Zack Florence,
PhD, Alberta Research Council.
References
1 Walsh PC, Partin AW, Epstein JI. Cancer control and
quality of life following anatomical radical retropubic
prostatectomy: results at 10 years. J Urol 1994; 152:
1831–6
2 Jønler M, Madsen FA, Rhodes PR, Sall M, Messing EM,
Bruskewitz RC. A prospective study of quantification of
urinary incontinence and quality of life in patients
undergoing radical retropubic prostatectomy. Urology
1996; 48: 433–40
3 Jackson J, Emerson L, Johnston B, Wilson J, Morales A.
Biofeedback: a noninvasive treatment for incontinence after
radical prostatectomy. Urol Nursing 1996; 16: 50–4
4 Joseph AC, Chang MK. A bladder behavior clinic for post
prostatectomy patients. Urol Nursing 1989; 9: 15–9
5 Meaglia JP, Joseph AC, Chang M, Schmidt JD. Postprostatectomy urinary incontinence: response to behavioral
training. J Urol 1990; 144: 674–6
6 Hirakawa S, Hassouna M, Deleon R, Elhilali M. The role of
combined pelvic floor stimulation and biofeedback in postprostatectomy urinary incontinence. J Urol 1993; 149:
235A (Abs 87)
7 Salinas CJ, Virseda CM, Salomon MS, Bravo de Rueda C,
Aristizabal JM, Resel EL. Results of electrical stimulation in
the treatment of post-prostatectomy urinary incontinence.
Actas Urolog Espanol 1996; 20: 544–50 (Abs)
8 Sotiropoulos A, Yeaw S, Lattimer JK. Management of
urinary incontinence with electronic stimulation: observations and results. J Urol 1976; 116: 747–50
9 Hahn I, Sommar S, Fall M. A comparative study of pelvic
floor training and electrical stimulation for the treatment
of genuine female stress urinary incontinence. Neurourol
Urodyn 1991; 10: 545–54
10 Wise BG, Khullar V, Cardozo L. Bladder neck movement
during pelvic floor contraction and intravaginal electrical
stimulation in women with and without genuine stress
incontinence. Neurourol Urodyn 1992; 11: 309–11
11 Urinary incontinence in adults. Clinical Practice Guidelines.
Rockville, Maryland: US Department of Health and Human
Services; Agency for Health Care Policy and Research
publication no. 96–0682, 1996
12 Herr HW. Quality of life of incontinent men after radical
prostatectomy. J Urol 1994; 151: 652–4
13 Uebersax JS, Wyman JF, Shumaker SA, McClish DK, Fantl
JA and the Continence Program for Women Research
Group. Short forms to assess life quality and symptom
distress for urinary incontinence in women: The
Incontinence Impact Questionnaire and the Urogenital
Distress Inventory. Neurourol Urodyn 1995; 14: 131–9
14 Aaronson NK, Ahmedzai S, Bergman B et al. The European
Organization for Research and Treatment of Cancer QLQC30: a quality-of-life instrument for use in international
clinical trials in oncology. J Nat Canc Inst 1993; 85: 365–76
15 Young H, Davis DM, Johnson FP. Young’s Practice of Urology
(based on a study of 12,500 cases). Vol. I. Philadelphia: WB
Saunders Co, 1926: 495–653
16 Burgio K, Stutzman RE, Engel BT. Behavioral training for
post-prostatectomy urinary incontinence. J Urol 1989;
141: 303–6
17 Bennett JK, Foote JE, Green BG, Killorin EW, Martin SH.
ECectiveness of biofeedback/electrostimulation in treatment
of post-prostatectomy urinary incontinence. Proceedings of
the Annual Urodynamics Society Meeting, New Orleans,
1997 (Abstract)
18 Donnellan SM, Duncan HJ, MacGregor RJ, Russell JM.
Prospective assessment of incontinence after radical prostatectomy: objective and subjective analysis. Urology 1997;
49: 225–30
19 Hunskaar S, Vinsnes A. The quality of life in women with
urinary incontinence as measured by the Sickness Impact
Profile. J Am Ger Soc 1991; 39: 378–82
20 Moore KN. ECect of urinary incontinence on QOL after
radical prostatectomy. Unpublished doctoral dissertation.
University of Alberta, Edmonton, Canada, 1997
21 Aaronson NK, Visser-Pol E, Leenhouts GH et al. Telephonebased nursing intervention improves the eCectiveness of
the informed consent process in cancer clinical trials. J Clin
Oncol 1996; 14: 984–6
22 Alter CL, Fleishman SB, Kornblith AB et al. Supportive
telephone intervention for patients receiving chemotherapy.
A pilot study. Psychosomatics 1996; 37: 425–31
23 Diokno A. Post prostatectomy urinary incontinence.
Proceedings of the First International Conference for the
Prevention of Incontinence, Simon Foundation, USA and The
Continence Foundation UK, Marlow, England, June
25–27, 1997
24 Kerr LA, Zincke H. Radical retropubic prostatectomy for
prostate cancer in the elderly and the young: complications
and prognosis. Eur Urol 1994; 25: 305–12
25 Eastham JA, Kattan MW, Rogers E et al. Risk factors for
urinary incontinence after radical prostatectomy. J Urol
1996; 156: 1707–13
26 Gormley EA, GriBths DJ, McCracken PN, Harrison GM,
McPhee MS. ECect of transurethral resection of the prostate
on detrusor instability and urge incontinence in elderly
males. Neurourol Urodyn 1993; 12: 445–53
27 GriBths DJ, McCracken PN, Harrison GM. Incontinence in
the elderly: objective demonstration and quantitative
assessment. Br J Urol 1991; 67: 467–71
28 Speakman MJ, Sethia KK, Fellows GJ, Smith JC. A study of
the pathogenesis, urodynamic assessment and outcome of
detrusor instability associated with bladder outflow obstruction. Br J Urol 1987; 59: 40–4
© 1999 BJU International 83, 57–65
URINARY INCONTINENCE AFTER RADICAL PROSTAT ECTOMY
29 Foote J, Yun S, Leach GE. Post-prostatectomy incontinence:
pathophysiology, evaluation, and management. Urol Clin
N Am 1991; 18: 229–41
30 Haab F, Yamaguchi R, Leach GE. Postprostatectomy
incontinence. Urol Clin N Am 1996; 23: 447–57
31 Burgio KL, Robinson JC, Engel BT. The role of biofeedback
in Kegel exercise training for stress urinary incontinence.
Am J Obstet Gynecol 1986; 154: 58–64
32 Glavind K, Nohr SB, Walter S. Biofeedback and physiotherapy versus physiotherapy alone in the treatment of
genuine stress urinary incontinence. Int Urogynecol J Pelvic
Floor Dysfunct 1996; 7: 339–43
33 Berghmans LC, Frederiks CM, de Bie RA et al. EBcacy of
biofeedback, when included with pelvic floor muscle
exercise treatment, for genuine stress incontinence.
Neurourol Urodyn 1996; 15: 37–52
© 1999 BJU International 83, 57–65
65
34 Burns PA, PranikoC K, Nochajski TH, Hadley EC, Levy KJ,
Ory MG. A comparison of eCectiveness of biofeedback and
pelvic muscle exercise treatment of stress incontinence in
older community-dwelling women. J Gerontol 1993; 48:
M167–74
Authors
K. Moore, PhD, Assistant Professor.
D. GriBths, PhD, GriBths Urodynamics and Pro-Continence
Consulting, Edmonton.
A. Hughton, BPT, Northtown Physiotherapy, Edmonton.
Correspondence: Dr K. Moore, Faculty of Nursing, 3rd Floor
Clinical Science Building, University of Alberta, Edmonton,
Canada T6G 2G3.