- Email: [email protected]
Influence of Preoperative and Postoperative Pelvic Floor Muscle Training (PFMT) Compared with Postoperative PFMT on Urinary Incontinence After Radical Prostatectomy: A Randomized Controlled Trial
EUROPEAN UROLOGY 64 (2013) 766–772
available at www.sciencedirect.com journal homepage: www.europeanurology.com
Platinum Priority – Incontinence Editorial by David F. Penson on pp. 773–775 of this issue
Influence of Preoperative and Postoperative Pelvic Floor Muscle Training (PFMT) Compared with Postoperative PFMT on Urinary Incontinence After Radical Prostatectomy: A Randomized Controlled Trial Inge Geraerts a,*, Hendrik Van Poppel b, Nele Devoogdt a, Steven Joniau b, Ben Van Cleynenbreugel b, An De Groef a, Marijke Van Kampen a a
KU Leuven, Department of Rehabilitation Science, Leuven, Belgium; b UZ Leuven, Department of Urology, Leuven, Belgium
Article info
Abstract
Article history: Accepted January 14, 2013 Published online ahead of print on January 21, 2013
Background: The efficacy of preoperative pelvic floor muscle training (PFMT) for urinary incontinence (UI) after open radical prostatectomy (ORP) and robot-assisted laparoscopic radical prostatectomy (RARP) is still unclear. Objective: To determine whether patients with additional preoperative PFMT regain urinary continence earlier than patients with only postoperative PFMT after ORP and RARP. Design, setting, and participants: A randomized controlled trial enrolled 180 men who planned to undergo ORP/RARP. Intervention: The experimental group (E, n = 91) started PFMT 3 wk before surgery and continued after surgery. The control group (C, n = 89) started PFMT after catheter removal. Outcome measurements and statistical analysis: The primary end point was time to continence. Patients measured urine loss daily (24-h pad test) until total continence (three consecutive days of 0 g of urine loss) was achieved. Secondary end points were 1-h pad test, visual analog scale (VAS), International Prostate Symptom Score (IPSS), and quality of life (King’s Health Questionnaire [KHQ]). Kaplan-Meier analysis and Cox regression with correction for two strata (age and type of surgery) compared time and continence. The Fisher exact test was applied for the 1-h pad test and VAS; the Mann-Whitney U test was applied for IPSS and KHQ. Results and limitations: Patients with additional preoperative PFMT had no shorter duration of postoperative UI compared with patients with only postoperative PFMT ( p = 0.878). Median time to continence was 30 and 31 d, and median amount of first-day incontinence was 108 g and 124 g for groups E and C, respectively. Cox regression did not indicate a significant difference between groups E and C ( p = 0.773; hazard ratio: 1.047 [0.768–1.425]). The 1-h pad test, VAS, and IPSS were comparable between both groups. However, ‘‘incontinence impact’’ (KHQ) was in favor of group E at 3 mo and 6 mo after surgery. Conclusions: Three preoperative sessions of PFMT did not improve postoperative duration of incontinence. Trial registration: Netherlands Trial Register No. NTR 1953. # 2013 European Association of Urology. Published by Elsevier B.V. All rights reserved.
Keywords: Urinary incontinence Pelvic floor muscle training Radical prostatectomy
Please visit www.eu-acme.org/ europeanurology to read and answer questions on-line. The EU-ACME credits will then be attributed automatically.
* Corresponding author. KU Leuven, Department of Rehabilitation Science, Tervuursevest 101, Post Office Box 1501, 3000 Leuven, Belgium. Tel. +32 16 329120; Fax: +32 16 329197. E-mail address: [email protected] (I. Geraerts). 0302-2838/$ – see back matter # 2013 European Association of Urology. Published by Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.eururo.2013.01.013
EUROPEAN UROLOGY 64 (2013) 766–772
1.
Introduction
Urinary incontinence (UI) remains a common postoperative consequence after open radical prostatectomy (ORP) and robot-assisted laparoscopic radical prostatectomy (RARP) [1–4]. UI has a spontaneous recovery in most men, but it may take as long as 1–2 yr after radical prostatectomy (RP) [5]. Investigators have proved that continence can be achieved faster with pelvic floor muscle training (PFMT) [1,2,6]. Additionally, six studies attempted to investigate the effect of preoperative PFMT on the duration and severity of UI after RP [3,7–11]. Five of these studies found positive results of preoperative PFMT [3,7,8,10,11]. However, because of the multitude of existing bias, results must be interpreted with caution. Burgio et al. [3], Parekh et al. [8], and Tienforti et al. [11] altered both preoperative and postoperative treatment, which made defining the effect of preoperative PFMT impossible. Bales et al. [9] compared the effect of two different preoperative treatments, and Sueppel et al. [7] compared only one preoperative session with a control group who completed PFMT 6 wk after surgery and included only 16 patients. Follow-up was usually only 3 mo or 6 mo [3,9–11] Finally, a wide range of continence criteria were used among studies, which made it difficult to compare results. The aim of this study was to determine whether patients who performed PFMT before and after surgery regained urinary continence earlier than patients who performed PFMT only after catheter removal.
2.3.
767
Interventions
The PFMT program consisted of exercises of the pelvic floor manually controlled by the therapist and supplied with electromyography (EMG) biofeedback. Every patient received individual treatment on an outpatient basis once a week. The experimental group started exercising 3 wk before RP. They received one session of 30 min of guided PFMT per week until surgery. Additionally, patients performed a home program of 60 contractions per day and were instructed on contracting the pelvic floor muscles while coughing and sitting down or getting up from a chair. The fourth day after surgery, with the urinary catheter in situ, patients were encouraged to restart PFMT. The control group started exercising the day of catheter removal. Postoperatively, all patients came to the hospital once a week to discuss their bladder diary and to perform an individual guided exercise session with digital or EMG biofeedback control. Further specific exercises were given, along with advice on how to use the pelvic floor muscles to prevent urine loss during particular functional activities indicated by the patient. In both groups, PFMT was continued as long as any degree of UI persisted. All patients in the experimental group were preoperatively treated by the same therapist (who was different from the therapist who performed the postoperative treatments and different from the assessor). After catheter removal, all patients were treated by another therapist. Both therapists had several years of experience in the treatment of UI. Patients were controlled for adherence to the home exercises. When they performed the prescribed 60 contractions per day, they had to color three squares in their diary (each square equaled 20 contractions).
2.4.
Assessments
After catheter withdrawal, urine loss per 24 h was recorded. Patients continued this record at home until continence was achieved. Continence
2.
Methods
2.1.
Patients
was defined as three consecutive days of 0 g of urine loss on the 24-h pad test. Double-checkingof the patients’ self-measurements was performed on a regular basis by weighing the pads they wore when they came to therapy. From time totime, patients had tocollect all diapers over 24 h in a plastic bag
Inclusion criteria were men planning to undergo ORP or RARP in the
and bring the bag to the hospital for a second measurement. No difference
University Hospitals Leuven who accepted ambulatory visits once a
was made in analysis between types of incontinence (stress or urge).
week until total continence was achieved and agreed to perform
All patients were prospectively assessed before RP and 1 mo, 3 mo,
measurements preoperatively and at 1 mo, 3 mo, 6 mo, and 12 mo after
6 mo, and 12 mo after RP in the physiotherapy department. Patients had to
surgery. Patients were excluded if they had cognitive problems, were
perform a 1-h pad test, fill in a visual analog scale (VAS) concerning their
non–Dutch speaking, simultaneously planned other pelvic surgery or a
subjective feelings about UI, and complete the International Prostate
salvage procedure, had transport problems, had a lack of time, had
Symptom Score (IPSS) [17], a questionnaire to assess voiding symptoms.
psychosocial or other medical problems, did not want to participate,
Additionally, the King’s Health Questionnaire (KHQ), to evaluate the
insisted on preoperative PFMT, were not approachable, or did not have
impact of UI on quality of life, was completed [18]. Strength and endurance
enough time between diagnosis and date of planned surgery.
were also measured. Preoperatively, all patients performed a 24-h pad test during 3 d. One blinded and well-trained assessor performed the
2.2.
Procedure
measurements. Patients were explicitly asked not to mention their group allocation to the assessor. No urodynamic measurements were performed
Before surgery, patients were randomized into the experimental group,
in the first postoperative year. Anticholinergics were not prescribed in the
which started PFMT 3 wk before RP and continued after surgery, or the
initial postoperative period.
control group, which started PFMT after catheter removal. Patients were
Primary outcome parameters were cumulative incidence of conti-
recruited at the outpatient urology clinic. PFMT could start only 3 wk
nence and time to continence (24-h pad test). Secondary outcomes were
before surgery, because the waiting list for surgery was only 3 wk.
the point prevalence of continence, measured with the 1-h pad test and
Randomization was performed within each stratum by using permuted blocks (size = 4). Strata were age (<65 yr compared with
the VAS at 1 mo, 3 mo, 6 mo, and 12 mo after surgery. At the same time points, IPSS and KHQ were assessed.
65 yr) and surgical technique (ORP compared with RARP), because these two factors are important risk factors for UI [12–16]. Allocation to
2.5.
Sample size and statistical analysis
the treatment groups was concealed. A computer program carried out the randomization. The sequence of randomization was determined
Sample size was calculated to detect a difference between the control
by the patients’ presence at the outpatient urology clinic. Choice of surgical
and treatment groups in time until continence based on a two-sided
technique depended on tumor characteristics, choice of the patient, and
log-rank test (a = 0.05). An exponential distribution was assumed for the
medical history. Three experienced surgeons each specialized in ORP and/
event times, yielding 60% and 40% of patients being continent after 6 mo in
or RARP and completed all operations blinded to randomization.
the treatment and control groups, respectively. Note that the scenario for
768
EUROPEAN UROLOGY 64 (2013) 766–772
the difference after 6 mo was derived from Burgio et al. [3]. The assumed
Table 1 – Characteristics of patients according to treatment group
distribution and difference after 6 mo corresponded to a median time until
Experimental group, n = 91
continence of 4.5 mo and 8.1 mo in treatment and control groups,
Control group, n = 89
respectively. It also implied that 84% of the subjects (treatment group) and 64% of the subjects (control group) were continent after 1 yr, the time at which the comparison was made. In total, 166 subjects were needed to have 90% power to detect the assumed difference between both survival curves. With drop-outs taken into account, 180 patients had to be included. We compared patient characteristics between excluded and included patients to analyze whether both groups were comparable. Data were analyzed according to the intention-to-treat principle. Kaplan-Meier analyses with the log-rank test were used to compare the time to continence between both groups. Drop-outs were censored at the moment of last follow-up. Afterward, a Cox regression was applied to compare the different groups concerning the time to continence with correction for the two strata. The Fisher exact test was used to compare objective and subjective point prevalence of urinary continence, defined as 0 or 1 g on the 1-h pad test and the VAS, measured at 1 mo, 3 mo, 6 mo, and 12 mo after surgery. For comparison of voiding symptom severity and quality of life at 1 mo, 3 mo, 6 mo, and 12 mo after surgery, the Mann-Whitney U test was used because data were not normally distributed. All data were analyzed with SPSS 19.0. The procedures of the study received ethical approval from the commission on medical ethics of the University Hospitals Leuven responsible for human/animal experimentation (ML5470).
3.
Age, yr, mean (SD) (range) 61.88 (5.90) (44–73) Type of surgery, no. (%) ORP 60 (66) RARP 31 (34) D’Amico risk group, no. (%) 1 13 (14) 2 45 (49) 3 32 (35) Missing 1 (1) Nerve sparing, no. (%) Non–nerve sparing 13 (14) Unilateral nerve sparing 22 (24) Bilateral nerve sparing 56 (62) Missing 0 (0) Surgical margin status, no. (%) Negative 73 (80) Positive/doubtful 17 (19) Missing 1 (1) Preoperative continence status, no. (%) Continent 63 (69) Incontinent 25 (28) Missing 3 (3) Body mass index, kg/m2, no. (%) <25.0 30 (33) 25.1–30.0 51 (56) >30.0 10 (11)
62.04 (6.33) (41–76) 56 (63) 33 (37) 9 48 32 0
(10) (54) (36) (0)
11 22 56 0
(12) (25) (63) (0)
63 (71) 26 (29) 0 (0) 53 (59) 30 34) 6 (7) 24 (27) 48 (54) 17 (19)
Results
Between September 2009 and July 2011, 275 patients with localized or locally advanced prostate cancer were scheduled for ORP or RARP in the University Hospitals Leuven. A total of 180 patients met the inclusion criteria (Fig. 1) and signed written informed consent. Patients who did not want to participate for several reasons (n = 81) were asked to report their date of continence. Compared with nonparticipating (excluded) patients, included patients had less ORP and more RARP surgery ( p = 0.001), more patients with an intermediate D’Amico risk group and fewer patients with a high D’Amico risk group ( p < 0.001), and fewer patients with non– nerve-sparing surgery and more patients with bilateral nerve-sparing surgery ( p = 0.001). Other characteristics, such as age and surgical margin status, were comparable between both groups (data not shown). One hundred eighty patients were included; 91 patients were randomized to the intervention group and 89 patients to the control group. Ten patients were lost to follow-up. Figure 1 shows the flow of patients and reasons for dropout. Patients were very strict toward their appointments with the physiotherapist and the performance of exercises.
SD = standard deviation; ORP = open radical prostatectomy; RARP = robotassisted laparoscopic radical prostatectomy.
All characteristics of both groups were comparable (Table 1). Median time to continence was 30 and 31 d for the control group and the experimental group, respectively ( p = 0.878). The median amount of first-day UI was 108 and 124 g for the experimental and control groups, respectively ( p = 0.880). 3.1.
Primary outcomes
Figure 2 shows the Kaplan-Meier analyses for time to continence (24-h pad test). Time to continence was comparable between experimental and control groups during the first year after surgery ( p = 0.878). Additionally, Cox regression with correction for the two strata (age and type of surgery) gave comparable results ( p = 0.773; hazard ratio: 1.047 [0.768–1.425]) (Table 2). Four patients in the control group and two patients in the experimental group remained incontinent 12 mo after surgery. Compared with controls, patients in the intervention
Table 2 – Cox regression for time to urinary continence according to type of treatment B
Treatment group Type of surgery Age
0.046 0.430 0.519
SE
0.158 0.164 0.173
Wald
0.083 6.850 9.035
df
1 1 1
p value
0.773 0.009 0.003
B = B coefficient; SE = standard error; df = degrees of freedom; HR = hazard ratio; CI = confidence interval.
HR
1.047 1.537 0.595
95% CI Lower
Upper
0.768 1.114 0.424
1.425 2.122 0.835
EUROPEAN UROLOGY 64 (2013) 766–772
[(Fig._1)TD$IG]
769
275 paents scheduled for a radical prostatectomy
Excluded, because of: - Cognive problems (2) - Non–Dutch speaking (6) - Simultaneous other surgery (6) - Transport problems (26) - Lack of me (4) - Psychosocial/other medical problems (2) - Refused parcipaon (5) - Insisted on preoperave PFMT (12) - Not approachable (6) - Not enough me between diagnosis and date of planned surgery (26)
180 paents randomized
Experimental group (n = 91)
Drop-outs (n = 5): Died (n = 1) Cerebrovascular accident (n = 1) Transport problems (n = 3)
Assessed at 1 mo, 3 mo, and 6 mo (n = 86)
Drop-outs (n = 1): Refused further parcipaon (n = 1)
Assessed at 12 mo (n = 85)
Control group (n = 89)
Drop-outs (n = 2): Transport problems (n = 2)
Assessed at 1 mo and 3 mo (n = 87)
Drop-outs (n = 2): Refused further parcipaon (n = 2)
Assessed at 12 mo (n = 85)
Fig. 1 – Trial profile. PFMT = pelvic floor muscle training.
group had comparable cumulative incidence rates for continence and average amount of urine loss (24-h pad test) at all time points (Table 3). 3.2.
Secondary outcomes
The point prevalence of urinary continence, defined as 0 or 1 g on the 1-h pad test and the VAS, was comparable for both groups at 1 mo, 3 mo, 6 mo, and 12 mo after RP (Table 4). Voiding symptoms (IPSS) did not differ between the experimental and control groups at any time point. Only one aspect of the KHQ, incontinence impact, was in favor of the experimental group at 3 mo and 6 mo after surgery ( p = 0.008 and p = 0.024, respectively).
Strength and endurance did not significantly differ between the two groups at any time point. However patients with stronger pelvic floor muscles needed less time to become continent ( p = 0.015). Of 180 patients, 23 (13%) received additional radiotherapy. For all patients except three, radiotherapy was started only after continence was achieved. In these three patients, mean urine loss at the start of radiotherapy was 7 g/d, 8 g/d, and 64 g/d, respectively. Sixty-five percent of patients were completely continent during 3 d in the preoperative period. Thirty percent had a small amount of urine loss (range: 1–10 g/d). However, Kaplan-Meier analysis for time to continence indicated that patients without preoperative urine loss achieved continence
770
EUROPEAN UROLOGY 64 (2013) 766–772
Table 3 – Cumulative incidence of continence and average urine loss of patients at 1 mo, 3 mo, 6 mo, and 12 mo after radical prostatectomy based on the 24-h pad test Time since catheter removal, mo
Continent patients, no. (%) Experimental group, n = 85
1 3 6 12
44 67 80 83
Average urine loss, g
Control group, n = 85
(52) (79) (94) (98)
44 71 80 81
(52) (84) (94) (95)
Experimental group, n = 85
Control group, n = 85
90 17 12 2
85 13 3 3
Table 4 – Comparison of the point prevalence of continence after radical prostatectomy at 1 mo, 3 mo, 6 mo, and 12 mo after surgery, according to treatment group Experimental group, no. (%)
Control group, no. (%)
Point prevalence of continence, VAS scale, defined as 1/10 At 1 mo (89/88) 35 (39.3) At 3 mo (88/87) 64 (70.7) At 6 mo (88/86) 73 (82.9) At 12 mo (84/84) 72 (85.7) Point prevalence of continence, 1-h pad test, defined as 0 g At 1 mo (85/86) 42 (49.4) At 3 mo (86/86) 63 (73.3) At 6 mo (86/85) 76 (88.4) At 12 mo (81/83) 68 (83.9) Point prevalence of continence, 1-h pad test, defined as 1 g At 1 mo (85/86) 48 (56.5) At 3 mo (86/86) 71 (82.6) At 6 mo (86/85) 78 (90.7) At 12 mo (81/83) 74 (91.4)
Odds ratio (95% CI)
p value*
38 52 65 62
(43.2) (59.8) (75.6) (73.8)
0.853 1.795 1.572 2.129
(0.469–1.552) (0.951–3.388) (0.749–3.302) (0.975–4.649)
0.648 0.080 0.264 0.083
41 61 73 73
(47.7) (70.9) (85.9) (87.9)
1.072 1.123 1.249 0.717
(0.588–1.953) (0.576–2.187) (0.509–3.068) (0.295–1.742)
0.879 0.865 0.655 0.506
51 71 80 76
(59.3) (82.6) (94.1) (91.6)
0.890 1.000 0.609 0.974
(0.485–1.634) (0.455–2.198) (0.191–1.944) (0.326–2.912)
0.758 1.000 0.566 1.000
CI = confidence interval; VAS = visual analog scale. Fisher exact test; statistical significance was defined as p < 0.05.
*
significantly faster than preoperatively incontinent patients ( p = 0.01). Duration of incontinence, based on the patient-reported date of continence, did not differ between included and excluded patients ( p = 0.121).
[(Fig._2)TD$IG]
Fig. 2 – Time to urinary continence according to type of treatment.
4.
Discussion
Patients who performed PFMT before and after surgery had no shorter duration of postoperative UI than patients who performed PFMT only after catheter removal. Only six patients remained incontinent after 1 yr (range: 6–167 g/d). Four patients opted for a male sling procedure and were continent afterward. Two patients refused surgical treatment because of minimal incontinence rate (6 g/d; n = 1) or other comorbidities (130 g/d; n = 1). Secondary analyses revealed that the 1-h pad test and the VAS did not significantly differ between both groups. The median time to urinary continence and the median amount of first-day UI were approximately the same for both groups. Voiding symptoms were comparable between both groups. Quality of life concerning the impact of incontinence was in favor of the experimental group at 3 mo and 6 mo after surgery. Although both groups recovered continence to the same extent, we cannot explain why the experimental group had better quality-of-life scores. Preoperative urine loss had a significant effect on duration of postoperative UI. Our study had several strengths. First, it was a prospective study offering the same postoperative treatment to both groups and only three sessions of preoperative PFMT to the experimental group. This characteristic makes comparison possible. All patients were followed for 12 mo and evaluated at regular time intervals. All patients measured urine loss daily during 24 h by weighing pads
771
EUROPEAN UROLOGY 64 (2013) 766–772
accurately to 1 g. Additionally, patients performed a 1-h pad test, VAS, IPSS, and KHQ at 1 mo, 3 mo, 6 mo, and 12 mo after surgery. Patients were assessed by a blinded independent assessor and treated by another blinded independent therapist. Patients were operated on by three experienced surgeons, using highly standardized surgical procedures. Finally, our sample size was sufficiently large and based on a priori power analysis. Limitations included the fact that the number of patients after ORP and RARP were unequal (116 vs 54 patients). Another limitation was the short waiting list made only three preoperative sessions possible. Every therapy that shortens duration of incontinence is valuable, but patients perhaps would have benefited from a longer preoperative exercise period to enhance strength and endurance. It can be assumed that three PFMT sessions train only awareness of the pelvic floor muscles. Additionally, it would have been interesting to administer the 1-h pad test the day of catheter removal for a second objective incontinence parameter. Patients were encouraged not to stay in bed during the whole day. However, we noticed that several patients tended to be rather inactive to minimalize urine loss. This fact might have caused an underestimate of the first-day UI. Finally, it would have been interesting to add a group without PFMT, but since evidence of the benefits of PFMT exists, adding that group was not allowed [1,2,19]. Twelve months after surgery, 80–97% of patients have regained continence [13,14,20–23]. Although we did not hypothesize about the number of patients with persistent incontinence at 1 yr, results were in line with previous research [1,13,14,23]. A recent review [24] supported the European Association of Urology (EAU) guidelines (page 41) [25] about UI that ‘‘men undergoing some form of PFMT, before or after RP achieved continence more quickly than nontreated men.’’ However, Campbell et al. indicated that findings should be treated with caution, as most trials were of poor to moderate quality [24]. In general, conservative management is advised as the main approach to UI after RP. Several studies consistently demonstrated that PFMT speeded the recovery of UI [2,19,26]. According to the EAU guidelines, approximately 10 trials compared the preventive effect of PFMT prior to RP. However, a very strict analysis of these studies shows that only five studies attempted to investigate the evidence of preoperative PFMT [3,7–10]. Similar to our results, Bales et al. did not find a significant effect of preoperative PFMT, although these authors gave only one preoperative session [9]. In other studies, positive results of preoperative training were mentioned [3,7,8,10,11], although a multitude of existing bias was cited. In our study, both groups recovered continence to the same extent when continence was defined as three consecutive days of 0 g of urine loss (24-h pad test). Additionally, analysis based on a definition of <2 g of urine loss showed no difference in time to continence between the two groups ( p = 0.815). This finding was also confirmed by the 1-h pad test. However, we cannot explain why the experimental group had better quality-of-life scores than the control group. All patients in the experimental group expressed their satisfaction at having received PFMT before surgery.
Patients in the control group were in general rather disappointed in not having had preoperative treatment. Three preoperative sessions of PFMT did not improve postoperative duration of incontinence. 5.
Conclusions
Standard postoperative continence rehabilitation using a strict pelvic floor reeducation scheme could not be improved by adding three preoperative sessions of PFMT. The impact of incontinence on quality of life, however, was less in the experimental group. Author contributions: Inge Geraerts had full access to all the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis. Study concept and design: Geraerts, Van Kampen, Van Poppel. Acquisition
of
data:
Geraerts,
De
Groef,
Van
Poppel,
Joniau,
Van Cleynenbreugel. Analysis and interpretation of data: Geraerts, Van Kampen, De Groef. Drafting of the manuscript: Geraerts, Van Kampen, De Groef. Critical revision of the manuscript for important intellectual content: Van Kampen, Van Poppel, Devoogdt, Joniau, Van Cleynenbreugel. Statistical analysis: Geraerts, De Groef. Obtaining funding: Geraerts, Van Kampen, Devoogdt. Administrative, technical, or material support: Geraerts, Van Kampen. Supervision: Geraerts, Van Kampen, Van Poppel. Other (specify): None. Financial disclosures: Inge Geraerts certifies that all conflicts of interest, including specific financial interests and relationships and affiliations relevant to the subject matter or materials discussed in the manuscript (eg, employment/affiliation, grants or funding, consultancies, honoraria, stock ownership or options, expert testimony, royalties, or patents filed, received, or pending), are the following: None. Funding/Support and role of the sponsor: Agency for Innovation by Science and Technology (Applied Biomedical Research): governmental grant. Acknowledgment statement: The authors acknowledge H. Lemkens, K. Sterckx, and A. Gielen.
References [1] Van Kampen M, de Weerdt W, Van Poppel H, De Ridder D, Feys H, Baert L. Effect of pelvic-floor re-education on duration and degree of incontinence after radical prostatectomy: a randomised controlled trial. Lancet 2000;355:98–102. [2] Manassero F, Traversi C, Ales V, et al. Contribution of early intensive prolonged pelvic floor exercises on urinary continence recovery after bladder neck-sparing radical prostatectomy: results of a prospective controlled randomized trial. Neurourol Urodynamics 2007; 26:985–9. [3] Burgio KL, Goode PS, Urban DA, et al. Preoperative biofeedback assisted behavioral training to decrease post-prostatectomy incontinence: a randomized, controlled trial. J Urol 2006;175:196–201. [4] Ficarra V, Novara G, Rosen RC, et al. Systematic review and metaanalysis of studies reporting urinary continence recovery after robot-assisted radical prostatectomy. Eur Urol 2012;62:405–17. [5] Sanda MG, Dunn RL, Michalski J, et al. Quality of life and satisfaction with outcome among prostate-cancer survivors. N Engl J Med 2008;358:1250–61.
772
EUROPEAN UROLOGY 64 (2013) 766–772
[6] Filocamo MT, Li Marzi V, Popolo GD, et al. Effectiveness of early pelvic floor rehabilitation treatment for post-prostatectomy incontinence. Eur Urol 2005;48:734–8.
counseling. Urol Oncol. In press. http://dx.doi.org/10.1016/j. urolonc.2011.10.003. [17] Barry MJ, Fowler Jr FJ, O’Leary MP, et al. The American Urological
[7] Sueppel C, Kreder K, See W. Improved continence outcomes with
Association symptom index for benign prostatic hyperplasia. The
preoperative pelvic floor muscle strengthening exercises. Urol Nurs
Measurement Committee of the American Urological Association.
2001;21:201–10.
J Urol 1992;148:1549–57, discussion 1564.
[8] Parekh AR, Feng MI, Kirages D, Bremner H, Kaswick J, Aboseif S. The
[18] Okamura K, Nojiri Y, Osuga Y. Reliability and validity of the King’s
role of pelvic floor exercises on post-prostatectomy incontinence.
Health Questionnaire for lower urinary tract symptoms in both
J Urol 2003;170:130–3.
genders. BJU Int 2009;103:1673–8.
[9] Bales GT, Gerber GS, Minor TX, et al. Effect of preoperative
[19] Ribeiro LH, Prota C, Gomes CM, et al. Long-term effect of early
biofeedback/pelvic floor training on continence in men undergoing
postoperative pelvic floor biofeedback on continence in men un-
radical prostatectomy. Urology 2000;56:627–30.
dergoing radical prostatectomy: a prospective, randomized, con-
[10] Centemero A, Rigatti L, Giraudo D, et al. Preoperative pelvic floor muscle exercise for early continence after radical prostatectomy: a randomised controlled study. Eur Urol 2010;57:1039–44. [11] Tienforti D, Sacco E, Marangi F, et al. Efficacy of an assisted
trolled trial. J Urol 2010;184:1034–9. [20] Menon M, Shrivastava A, Kaul S, et al. Vattikuti Institute prostatectomy: contemporary technique and analysis of results. Eur Urol 2007;51:648–58.
low-intensity programme of perioperative pelvic floor muscle
[21] Di Pierro GB, Baumeister P, Stucki P, Beatrice J, Danuser H, Mattei A.
training in improving the recovery of continence after radical
A prospective trial comparing consecutive series of open retropubic
prostatectomy: a randomized controlled trial. BJU Int 2012;
and robot-assisted laparoscopic radical prostatectomy in a centre
110:1004–10.
with a limited caseload. Eur Urol 2011;59:1–6.
[12] Tewari A, Srivasatava A, Menon M, Members of the VIPT. A
[22] Lepor H, Kaci L, Xue X. Continence following radical retropubic
prospective comparison of radical retropubic and robot-assisted
prostatectomy using self-reporting instruments. J Urol 2004;171:
prostatectomy: experience in one institution. BJU Int 2003;92: 205–10. [13] Ficarra V, Novara G, Fracalanza S, et al. A prospective, non-randomized trial comparing robot-assisted laparoscopic and retropubic radical
1212–5. [23] Krambeck AE, DiMarco DS, Rangel LJ, et al. Radical prostatectomy for prostatic adenocarcinoma: a matched comparison of open retropubic and robot-assisted techniques. BJU Int 2009;103:448–53.
prostatectomy in one European institution. BJU Int 2009;104: 534–9.
[24] Campbell SE, Glazener CM, Hunter KF, Cody JD, Moore KN. Conser-
[14] Rocco B, Matei D-V, Melegari S, et al. Robotic vs open prostatectomy
vative management for postprostatectomy urinary incontinence.
in a laparoscopically naive centre: a matched-pair analysis. BJU Int 2009;104:991–5. [15] Kim SC, Song C, Kim W, et al. Factors determining functional outcomes after radical prostatectomy: robot-assisted versus retropubic. Eur Urol 2011;60:413–9.
Cochrane Database Syst Rev 2012:CD001843. [25] Lucas MG, Bosch JLHR, Cruz FR, et al. Guidelines on urinary incontinence. Arnhem, The Netherlands: European Association of Urology; 2012. p. 41. [26] Marchiori D, Bertaccini A, Manferrari F, Ferri C, Martorana G. Pelvic
[16] Khoder WY, Trottmann M, Stuber A, Stief CG, Becker AJ. Early
floor rehabilitation for continence recovery after radical prostatec-
incontinence after radical prostatectomy: a community based ret-
tomy: role of a personal training re-educational program. Anticancer
rospective analysis in 911 men and implications for preoperative
Res 2010;30:553–6.
available at www.sciencedirect.com journal homepage: www.europeanurology.com
Platinum Priority – Incontinence Editorial by David F. Penson on pp. 773–775 of this issue
Influence of Preoperative and Postoperative Pelvic Floor Muscle Training (PFMT) Compared with Postoperative PFMT on Urinary Incontinence After Radical Prostatectomy: A Randomized Controlled Trial Inge Geraerts a,*, Hendrik Van Poppel b, Nele Devoogdt a, Steven Joniau b, Ben Van Cleynenbreugel b, An De Groef a, Marijke Van Kampen a a
KU Leuven, Department of Rehabilitation Science, Leuven, Belgium; b UZ Leuven, Department of Urology, Leuven, Belgium
Article info
Abstract
Article history: Accepted January 14, 2013 Published online ahead of print on January 21, 2013
Background: The efficacy of preoperative pelvic floor muscle training (PFMT) for urinary incontinence (UI) after open radical prostatectomy (ORP) and robot-assisted laparoscopic radical prostatectomy (RARP) is still unclear. Objective: To determine whether patients with additional preoperative PFMT regain urinary continence earlier than patients with only postoperative PFMT after ORP and RARP. Design, setting, and participants: A randomized controlled trial enrolled 180 men who planned to undergo ORP/RARP. Intervention: The experimental group (E, n = 91) started PFMT 3 wk before surgery and continued after surgery. The control group (C, n = 89) started PFMT after catheter removal. Outcome measurements and statistical analysis: The primary end point was time to continence. Patients measured urine loss daily (24-h pad test) until total continence (three consecutive days of 0 g of urine loss) was achieved. Secondary end points were 1-h pad test, visual analog scale (VAS), International Prostate Symptom Score (IPSS), and quality of life (King’s Health Questionnaire [KHQ]). Kaplan-Meier analysis and Cox regression with correction for two strata (age and type of surgery) compared time and continence. The Fisher exact test was applied for the 1-h pad test and VAS; the Mann-Whitney U test was applied for IPSS and KHQ. Results and limitations: Patients with additional preoperative PFMT had no shorter duration of postoperative UI compared with patients with only postoperative PFMT ( p = 0.878). Median time to continence was 30 and 31 d, and median amount of first-day incontinence was 108 g and 124 g for groups E and C, respectively. Cox regression did not indicate a significant difference between groups E and C ( p = 0.773; hazard ratio: 1.047 [0.768–1.425]). The 1-h pad test, VAS, and IPSS were comparable between both groups. However, ‘‘incontinence impact’’ (KHQ) was in favor of group E at 3 mo and 6 mo after surgery. Conclusions: Three preoperative sessions of PFMT did not improve postoperative duration of incontinence. Trial registration: Netherlands Trial Register No. NTR 1953. # 2013 European Association of Urology. Published by Elsevier B.V. All rights reserved.
Keywords: Urinary incontinence Pelvic floor muscle training Radical prostatectomy
Please visit www.eu-acme.org/ europeanurology to read and answer questions on-line. The EU-ACME credits will then be attributed automatically.
* Corresponding author. KU Leuven, Department of Rehabilitation Science, Tervuursevest 101, Post Office Box 1501, 3000 Leuven, Belgium. Tel. +32 16 329120; Fax: +32 16 329197. E-mail address: [email protected] (I. Geraerts). 0302-2838/$ – see back matter # 2013 European Association of Urology. Published by Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.eururo.2013.01.013
EUROPEAN UROLOGY 64 (2013) 766–772
1.
Introduction
Urinary incontinence (UI) remains a common postoperative consequence after open radical prostatectomy (ORP) and robot-assisted laparoscopic radical prostatectomy (RARP) [1–4]. UI has a spontaneous recovery in most men, but it may take as long as 1–2 yr after radical prostatectomy (RP) [5]. Investigators have proved that continence can be achieved faster with pelvic floor muscle training (PFMT) [1,2,6]. Additionally, six studies attempted to investigate the effect of preoperative PFMT on the duration and severity of UI after RP [3,7–11]. Five of these studies found positive results of preoperative PFMT [3,7,8,10,11]. However, because of the multitude of existing bias, results must be interpreted with caution. Burgio et al. [3], Parekh et al. [8], and Tienforti et al. [11] altered both preoperative and postoperative treatment, which made defining the effect of preoperative PFMT impossible. Bales et al. [9] compared the effect of two different preoperative treatments, and Sueppel et al. [7] compared only one preoperative session with a control group who completed PFMT 6 wk after surgery and included only 16 patients. Follow-up was usually only 3 mo or 6 mo [3,9–11] Finally, a wide range of continence criteria were used among studies, which made it difficult to compare results. The aim of this study was to determine whether patients who performed PFMT before and after surgery regained urinary continence earlier than patients who performed PFMT only after catheter removal.
2.3.
767
Interventions
The PFMT program consisted of exercises of the pelvic floor manually controlled by the therapist and supplied with electromyography (EMG) biofeedback. Every patient received individual treatment on an outpatient basis once a week. The experimental group started exercising 3 wk before RP. They received one session of 30 min of guided PFMT per week until surgery. Additionally, patients performed a home program of 60 contractions per day and were instructed on contracting the pelvic floor muscles while coughing and sitting down or getting up from a chair. The fourth day after surgery, with the urinary catheter in situ, patients were encouraged to restart PFMT. The control group started exercising the day of catheter removal. Postoperatively, all patients came to the hospital once a week to discuss their bladder diary and to perform an individual guided exercise session with digital or EMG biofeedback control. Further specific exercises were given, along with advice on how to use the pelvic floor muscles to prevent urine loss during particular functional activities indicated by the patient. In both groups, PFMT was continued as long as any degree of UI persisted. All patients in the experimental group were preoperatively treated by the same therapist (who was different from the therapist who performed the postoperative treatments and different from the assessor). After catheter removal, all patients were treated by another therapist. Both therapists had several years of experience in the treatment of UI. Patients were controlled for adherence to the home exercises. When they performed the prescribed 60 contractions per day, they had to color three squares in their diary (each square equaled 20 contractions).
2.4.
Assessments
After catheter withdrawal, urine loss per 24 h was recorded. Patients continued this record at home until continence was achieved. Continence
2.
Methods
2.1.
Patients
was defined as three consecutive days of 0 g of urine loss on the 24-h pad test. Double-checkingof the patients’ self-measurements was performed on a regular basis by weighing the pads they wore when they came to therapy. From time totime, patients had tocollect all diapers over 24 h in a plastic bag
Inclusion criteria were men planning to undergo ORP or RARP in the
and bring the bag to the hospital for a second measurement. No difference
University Hospitals Leuven who accepted ambulatory visits once a
was made in analysis between types of incontinence (stress or urge).
week until total continence was achieved and agreed to perform
All patients were prospectively assessed before RP and 1 mo, 3 mo,
measurements preoperatively and at 1 mo, 3 mo, 6 mo, and 12 mo after
6 mo, and 12 mo after RP in the physiotherapy department. Patients had to
surgery. Patients were excluded if they had cognitive problems, were
perform a 1-h pad test, fill in a visual analog scale (VAS) concerning their
non–Dutch speaking, simultaneously planned other pelvic surgery or a
subjective feelings about UI, and complete the International Prostate
salvage procedure, had transport problems, had a lack of time, had
Symptom Score (IPSS) [17], a questionnaire to assess voiding symptoms.
psychosocial or other medical problems, did not want to participate,
Additionally, the King’s Health Questionnaire (KHQ), to evaluate the
insisted on preoperative PFMT, were not approachable, or did not have
impact of UI on quality of life, was completed [18]. Strength and endurance
enough time between diagnosis and date of planned surgery.
were also measured. Preoperatively, all patients performed a 24-h pad test during 3 d. One blinded and well-trained assessor performed the
2.2.
Procedure
measurements. Patients were explicitly asked not to mention their group allocation to the assessor. No urodynamic measurements were performed
Before surgery, patients were randomized into the experimental group,
in the first postoperative year. Anticholinergics were not prescribed in the
which started PFMT 3 wk before RP and continued after surgery, or the
initial postoperative period.
control group, which started PFMT after catheter removal. Patients were
Primary outcome parameters were cumulative incidence of conti-
recruited at the outpatient urology clinic. PFMT could start only 3 wk
nence and time to continence (24-h pad test). Secondary outcomes were
before surgery, because the waiting list for surgery was only 3 wk.
the point prevalence of continence, measured with the 1-h pad test and
Randomization was performed within each stratum by using permuted blocks (size = 4). Strata were age (<65 yr compared with
the VAS at 1 mo, 3 mo, 6 mo, and 12 mo after surgery. At the same time points, IPSS and KHQ were assessed.
65 yr) and surgical technique (ORP compared with RARP), because these two factors are important risk factors for UI [12–16]. Allocation to
2.5.
Sample size and statistical analysis
the treatment groups was concealed. A computer program carried out the randomization. The sequence of randomization was determined
Sample size was calculated to detect a difference between the control
by the patients’ presence at the outpatient urology clinic. Choice of surgical
and treatment groups in time until continence based on a two-sided
technique depended on tumor characteristics, choice of the patient, and
log-rank test (a = 0.05). An exponential distribution was assumed for the
medical history. Three experienced surgeons each specialized in ORP and/
event times, yielding 60% and 40% of patients being continent after 6 mo in
or RARP and completed all operations blinded to randomization.
the treatment and control groups, respectively. Note that the scenario for
768
EUROPEAN UROLOGY 64 (2013) 766–772
the difference after 6 mo was derived from Burgio et al. [3]. The assumed
Table 1 – Characteristics of patients according to treatment group
distribution and difference after 6 mo corresponded to a median time until
Experimental group, n = 91
continence of 4.5 mo and 8.1 mo in treatment and control groups,
Control group, n = 89
respectively. It also implied that 84% of the subjects (treatment group) and 64% of the subjects (control group) were continent after 1 yr, the time at which the comparison was made. In total, 166 subjects were needed to have 90% power to detect the assumed difference between both survival curves. With drop-outs taken into account, 180 patients had to be included. We compared patient characteristics between excluded and included patients to analyze whether both groups were comparable. Data were analyzed according to the intention-to-treat principle. Kaplan-Meier analyses with the log-rank test were used to compare the time to continence between both groups. Drop-outs were censored at the moment of last follow-up. Afterward, a Cox regression was applied to compare the different groups concerning the time to continence with correction for the two strata. The Fisher exact test was used to compare objective and subjective point prevalence of urinary continence, defined as 0 or 1 g on the 1-h pad test and the VAS, measured at 1 mo, 3 mo, 6 mo, and 12 mo after surgery. For comparison of voiding symptom severity and quality of life at 1 mo, 3 mo, 6 mo, and 12 mo after surgery, the Mann-Whitney U test was used because data were not normally distributed. All data were analyzed with SPSS 19.0. The procedures of the study received ethical approval from the commission on medical ethics of the University Hospitals Leuven responsible for human/animal experimentation (ML5470).
3.
Age, yr, mean (SD) (range) 61.88 (5.90) (44–73) Type of surgery, no. (%) ORP 60 (66) RARP 31 (34) D’Amico risk group, no. (%) 1 13 (14) 2 45 (49) 3 32 (35) Missing 1 (1) Nerve sparing, no. (%) Non–nerve sparing 13 (14) Unilateral nerve sparing 22 (24) Bilateral nerve sparing 56 (62) Missing 0 (0) Surgical margin status, no. (%) Negative 73 (80) Positive/doubtful 17 (19) Missing 1 (1) Preoperative continence status, no. (%) Continent 63 (69) Incontinent 25 (28) Missing 3 (3) Body mass index, kg/m2, no. (%) <25.0 30 (33) 25.1–30.0 51 (56) >30.0 10 (11)
62.04 (6.33) (41–76) 56 (63) 33 (37) 9 48 32 0
(10) (54) (36) (0)
11 22 56 0
(12) (25) (63) (0)
63 (71) 26 (29) 0 (0) 53 (59) 30 34) 6 (7) 24 (27) 48 (54) 17 (19)
Results
Between September 2009 and July 2011, 275 patients with localized or locally advanced prostate cancer were scheduled for ORP or RARP in the University Hospitals Leuven. A total of 180 patients met the inclusion criteria (Fig. 1) and signed written informed consent. Patients who did not want to participate for several reasons (n = 81) were asked to report their date of continence. Compared with nonparticipating (excluded) patients, included patients had less ORP and more RARP surgery ( p = 0.001), more patients with an intermediate D’Amico risk group and fewer patients with a high D’Amico risk group ( p < 0.001), and fewer patients with non– nerve-sparing surgery and more patients with bilateral nerve-sparing surgery ( p = 0.001). Other characteristics, such as age and surgical margin status, were comparable between both groups (data not shown). One hundred eighty patients were included; 91 patients were randomized to the intervention group and 89 patients to the control group. Ten patients were lost to follow-up. Figure 1 shows the flow of patients and reasons for dropout. Patients were very strict toward their appointments with the physiotherapist and the performance of exercises.
SD = standard deviation; ORP = open radical prostatectomy; RARP = robotassisted laparoscopic radical prostatectomy.
All characteristics of both groups were comparable (Table 1). Median time to continence was 30 and 31 d for the control group and the experimental group, respectively ( p = 0.878). The median amount of first-day UI was 108 and 124 g for the experimental and control groups, respectively ( p = 0.880). 3.1.
Primary outcomes
Figure 2 shows the Kaplan-Meier analyses for time to continence (24-h pad test). Time to continence was comparable between experimental and control groups during the first year after surgery ( p = 0.878). Additionally, Cox regression with correction for the two strata (age and type of surgery) gave comparable results ( p = 0.773; hazard ratio: 1.047 [0.768–1.425]) (Table 2). Four patients in the control group and two patients in the experimental group remained incontinent 12 mo after surgery. Compared with controls, patients in the intervention
Table 2 – Cox regression for time to urinary continence according to type of treatment B
Treatment group Type of surgery Age
0.046 0.430 0.519
SE
0.158 0.164 0.173
Wald
0.083 6.850 9.035
df
1 1 1
p value
0.773 0.009 0.003
B = B coefficient; SE = standard error; df = degrees of freedom; HR = hazard ratio; CI = confidence interval.
HR
1.047 1.537 0.595
95% CI Lower
Upper
0.768 1.114 0.424
1.425 2.122 0.835
EUROPEAN UROLOGY 64 (2013) 766–772
[(Fig._1)TD$IG]
769
275 paents scheduled for a radical prostatectomy
Excluded, because of: - Cognive problems (2) - Non–Dutch speaking (6) - Simultaneous other surgery (6) - Transport problems (26) - Lack of me (4) - Psychosocial/other medical problems (2) - Refused parcipaon (5) - Insisted on preoperave PFMT (12) - Not approachable (6) - Not enough me between diagnosis and date of planned surgery (26)
180 paents randomized
Experimental group (n = 91)
Drop-outs (n = 5): Died (n = 1) Cerebrovascular accident (n = 1) Transport problems (n = 3)
Assessed at 1 mo, 3 mo, and 6 mo (n = 86)
Drop-outs (n = 1): Refused further parcipaon (n = 1)
Assessed at 12 mo (n = 85)
Control group (n = 89)
Drop-outs (n = 2): Transport problems (n = 2)
Assessed at 1 mo and 3 mo (n = 87)
Drop-outs (n = 2): Refused further parcipaon (n = 2)
Assessed at 12 mo (n = 85)
Fig. 1 – Trial profile. PFMT = pelvic floor muscle training.
group had comparable cumulative incidence rates for continence and average amount of urine loss (24-h pad test) at all time points (Table 3). 3.2.
Secondary outcomes
The point prevalence of urinary continence, defined as 0 or 1 g on the 1-h pad test and the VAS, was comparable for both groups at 1 mo, 3 mo, 6 mo, and 12 mo after RP (Table 4). Voiding symptoms (IPSS) did not differ between the experimental and control groups at any time point. Only one aspect of the KHQ, incontinence impact, was in favor of the experimental group at 3 mo and 6 mo after surgery ( p = 0.008 and p = 0.024, respectively).
Strength and endurance did not significantly differ between the two groups at any time point. However patients with stronger pelvic floor muscles needed less time to become continent ( p = 0.015). Of 180 patients, 23 (13%) received additional radiotherapy. For all patients except three, radiotherapy was started only after continence was achieved. In these three patients, mean urine loss at the start of radiotherapy was 7 g/d, 8 g/d, and 64 g/d, respectively. Sixty-five percent of patients were completely continent during 3 d in the preoperative period. Thirty percent had a small amount of urine loss (range: 1–10 g/d). However, Kaplan-Meier analysis for time to continence indicated that patients without preoperative urine loss achieved continence
770
EUROPEAN UROLOGY 64 (2013) 766–772
Table 3 – Cumulative incidence of continence and average urine loss of patients at 1 mo, 3 mo, 6 mo, and 12 mo after radical prostatectomy based on the 24-h pad test Time since catheter removal, mo
Continent patients, no. (%) Experimental group, n = 85
1 3 6 12
44 67 80 83
Average urine loss, g
Control group, n = 85
(52) (79) (94) (98)
44 71 80 81
(52) (84) (94) (95)
Experimental group, n = 85
Control group, n = 85
90 17 12 2
85 13 3 3
Table 4 – Comparison of the point prevalence of continence after radical prostatectomy at 1 mo, 3 mo, 6 mo, and 12 mo after surgery, according to treatment group Experimental group, no. (%)
Control group, no. (%)
Point prevalence of continence, VAS scale, defined as 1/10 At 1 mo (89/88) 35 (39.3) At 3 mo (88/87) 64 (70.7) At 6 mo (88/86) 73 (82.9) At 12 mo (84/84) 72 (85.7) Point prevalence of continence, 1-h pad test, defined as 0 g At 1 mo (85/86) 42 (49.4) At 3 mo (86/86) 63 (73.3) At 6 mo (86/85) 76 (88.4) At 12 mo (81/83) 68 (83.9) Point prevalence of continence, 1-h pad test, defined as 1 g At 1 mo (85/86) 48 (56.5) At 3 mo (86/86) 71 (82.6) At 6 mo (86/85) 78 (90.7) At 12 mo (81/83) 74 (91.4)
Odds ratio (95% CI)
p value*
38 52 65 62
(43.2) (59.8) (75.6) (73.8)
0.853 1.795 1.572 2.129
(0.469–1.552) (0.951–3.388) (0.749–3.302) (0.975–4.649)
0.648 0.080 0.264 0.083
41 61 73 73
(47.7) (70.9) (85.9) (87.9)
1.072 1.123 1.249 0.717
(0.588–1.953) (0.576–2.187) (0.509–3.068) (0.295–1.742)
0.879 0.865 0.655 0.506
51 71 80 76
(59.3) (82.6) (94.1) (91.6)
0.890 1.000 0.609 0.974
(0.485–1.634) (0.455–2.198) (0.191–1.944) (0.326–2.912)
0.758 1.000 0.566 1.000
CI = confidence interval; VAS = visual analog scale. Fisher exact test; statistical significance was defined as p < 0.05.
*
significantly faster than preoperatively incontinent patients ( p = 0.01). Duration of incontinence, based on the patient-reported date of continence, did not differ between included and excluded patients ( p = 0.121).
[(Fig._2)TD$IG]
Fig. 2 – Time to urinary continence according to type of treatment.
4.
Discussion
Patients who performed PFMT before and after surgery had no shorter duration of postoperative UI than patients who performed PFMT only after catheter removal. Only six patients remained incontinent after 1 yr (range: 6–167 g/d). Four patients opted for a male sling procedure and were continent afterward. Two patients refused surgical treatment because of minimal incontinence rate (6 g/d; n = 1) or other comorbidities (130 g/d; n = 1). Secondary analyses revealed that the 1-h pad test and the VAS did not significantly differ between both groups. The median time to urinary continence and the median amount of first-day UI were approximately the same for both groups. Voiding symptoms were comparable between both groups. Quality of life concerning the impact of incontinence was in favor of the experimental group at 3 mo and 6 mo after surgery. Although both groups recovered continence to the same extent, we cannot explain why the experimental group had better quality-of-life scores. Preoperative urine loss had a significant effect on duration of postoperative UI. Our study had several strengths. First, it was a prospective study offering the same postoperative treatment to both groups and only three sessions of preoperative PFMT to the experimental group. This characteristic makes comparison possible. All patients were followed for 12 mo and evaluated at regular time intervals. All patients measured urine loss daily during 24 h by weighing pads
771
EUROPEAN UROLOGY 64 (2013) 766–772
accurately to 1 g. Additionally, patients performed a 1-h pad test, VAS, IPSS, and KHQ at 1 mo, 3 mo, 6 mo, and 12 mo after surgery. Patients were assessed by a blinded independent assessor and treated by another blinded independent therapist. Patients were operated on by three experienced surgeons, using highly standardized surgical procedures. Finally, our sample size was sufficiently large and based on a priori power analysis. Limitations included the fact that the number of patients after ORP and RARP were unequal (116 vs 54 patients). Another limitation was the short waiting list made only three preoperative sessions possible. Every therapy that shortens duration of incontinence is valuable, but patients perhaps would have benefited from a longer preoperative exercise period to enhance strength and endurance. It can be assumed that three PFMT sessions train only awareness of the pelvic floor muscles. Additionally, it would have been interesting to administer the 1-h pad test the day of catheter removal for a second objective incontinence parameter. Patients were encouraged not to stay in bed during the whole day. However, we noticed that several patients tended to be rather inactive to minimalize urine loss. This fact might have caused an underestimate of the first-day UI. Finally, it would have been interesting to add a group without PFMT, but since evidence of the benefits of PFMT exists, adding that group was not allowed [1,2,19]. Twelve months after surgery, 80–97% of patients have regained continence [13,14,20–23]. Although we did not hypothesize about the number of patients with persistent incontinence at 1 yr, results were in line with previous research [1,13,14,23]. A recent review [24] supported the European Association of Urology (EAU) guidelines (page 41) [25] about UI that ‘‘men undergoing some form of PFMT, before or after RP achieved continence more quickly than nontreated men.’’ However, Campbell et al. indicated that findings should be treated with caution, as most trials were of poor to moderate quality [24]. In general, conservative management is advised as the main approach to UI after RP. Several studies consistently demonstrated that PFMT speeded the recovery of UI [2,19,26]. According to the EAU guidelines, approximately 10 trials compared the preventive effect of PFMT prior to RP. However, a very strict analysis of these studies shows that only five studies attempted to investigate the evidence of preoperative PFMT [3,7–10]. Similar to our results, Bales et al. did not find a significant effect of preoperative PFMT, although these authors gave only one preoperative session [9]. In other studies, positive results of preoperative training were mentioned [3,7,8,10,11], although a multitude of existing bias was cited. In our study, both groups recovered continence to the same extent when continence was defined as three consecutive days of 0 g of urine loss (24-h pad test). Additionally, analysis based on a definition of <2 g of urine loss showed no difference in time to continence between the two groups ( p = 0.815). This finding was also confirmed by the 1-h pad test. However, we cannot explain why the experimental group had better quality-of-life scores than the control group. All patients in the experimental group expressed their satisfaction at having received PFMT before surgery.
Patients in the control group were in general rather disappointed in not having had preoperative treatment. Three preoperative sessions of PFMT did not improve postoperative duration of incontinence. 5.
Conclusions
Standard postoperative continence rehabilitation using a strict pelvic floor reeducation scheme could not be improved by adding three preoperative sessions of PFMT. The impact of incontinence on quality of life, however, was less in the experimental group. Author contributions: Inge Geraerts had full access to all the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis. Study concept and design: Geraerts, Van Kampen, Van Poppel. Acquisition
of
data:
Geraerts,
De
Groef,
Van
Poppel,
Joniau,
Van Cleynenbreugel. Analysis and interpretation of data: Geraerts, Van Kampen, De Groef. Drafting of the manuscript: Geraerts, Van Kampen, De Groef. Critical revision of the manuscript for important intellectual content: Van Kampen, Van Poppel, Devoogdt, Joniau, Van Cleynenbreugel. Statistical analysis: Geraerts, De Groef. Obtaining funding: Geraerts, Van Kampen, Devoogdt. Administrative, technical, or material support: Geraerts, Van Kampen. Supervision: Geraerts, Van Kampen, Van Poppel. Other (specify): None. Financial disclosures: Inge Geraerts certifies that all conflicts of interest, including specific financial interests and relationships and affiliations relevant to the subject matter or materials discussed in the manuscript (eg, employment/affiliation, grants or funding, consultancies, honoraria, stock ownership or options, expert testimony, royalties, or patents filed, received, or pending), are the following: None. Funding/Support and role of the sponsor: Agency for Innovation by Science and Technology (Applied Biomedical Research): governmental grant. Acknowledgment statement: The authors acknowledge H. Lemkens, K. Sterckx, and A. Gielen.
References [1] Van Kampen M, de Weerdt W, Van Poppel H, De Ridder D, Feys H, Baert L. Effect of pelvic-floor re-education on duration and degree of incontinence after radical prostatectomy: a randomised controlled trial. Lancet 2000;355:98–102. [2] Manassero F, Traversi C, Ales V, et al. Contribution of early intensive prolonged pelvic floor exercises on urinary continence recovery after bladder neck-sparing radical prostatectomy: results of a prospective controlled randomized trial. Neurourol Urodynamics 2007; 26:985–9. [3] Burgio KL, Goode PS, Urban DA, et al. Preoperative biofeedback assisted behavioral training to decrease post-prostatectomy incontinence: a randomized, controlled trial. J Urol 2006;175:196–201. [4] Ficarra V, Novara G, Rosen RC, et al. Systematic review and metaanalysis of studies reporting urinary continence recovery after robot-assisted radical prostatectomy. Eur Urol 2012;62:405–17. [5] Sanda MG, Dunn RL, Michalski J, et al. Quality of life and satisfaction with outcome among prostate-cancer survivors. N Engl J Med 2008;358:1250–61.
772
EUROPEAN UROLOGY 64 (2013) 766–772
[6] Filocamo MT, Li Marzi V, Popolo GD, et al. Effectiveness of early pelvic floor rehabilitation treatment for post-prostatectomy incontinence. Eur Urol 2005;48:734–8.
counseling. Urol Oncol. In press. http://dx.doi.org/10.1016/j. urolonc.2011.10.003. [17] Barry MJ, Fowler Jr FJ, O’Leary MP, et al. The American Urological
[7] Sueppel C, Kreder K, See W. Improved continence outcomes with
Association symptom index for benign prostatic hyperplasia. The
preoperative pelvic floor muscle strengthening exercises. Urol Nurs
Measurement Committee of the American Urological Association.
2001;21:201–10.
J Urol 1992;148:1549–57, discussion 1564.
[8] Parekh AR, Feng MI, Kirages D, Bremner H, Kaswick J, Aboseif S. The
[18] Okamura K, Nojiri Y, Osuga Y. Reliability and validity of the King’s
role of pelvic floor exercises on post-prostatectomy incontinence.
Health Questionnaire for lower urinary tract symptoms in both
J Urol 2003;170:130–3.
genders. BJU Int 2009;103:1673–8.
[9] Bales GT, Gerber GS, Minor TX, et al. Effect of preoperative
[19] Ribeiro LH, Prota C, Gomes CM, et al. Long-term effect of early
biofeedback/pelvic floor training on continence in men undergoing
postoperative pelvic floor biofeedback on continence in men un-
radical prostatectomy. Urology 2000;56:627–30.
dergoing radical prostatectomy: a prospective, randomized, con-
[10] Centemero A, Rigatti L, Giraudo D, et al. Preoperative pelvic floor muscle exercise for early continence after radical prostatectomy: a randomised controlled study. Eur Urol 2010;57:1039–44. [11] Tienforti D, Sacco E, Marangi F, et al. Efficacy of an assisted
trolled trial. J Urol 2010;184:1034–9. [20] Menon M, Shrivastava A, Kaul S, et al. Vattikuti Institute prostatectomy: contemporary technique and analysis of results. Eur Urol 2007;51:648–58.
low-intensity programme of perioperative pelvic floor muscle
[21] Di Pierro GB, Baumeister P, Stucki P, Beatrice J, Danuser H, Mattei A.
training in improving the recovery of continence after radical
A prospective trial comparing consecutive series of open retropubic
prostatectomy: a randomized controlled trial. BJU Int 2012;
and robot-assisted laparoscopic radical prostatectomy in a centre
110:1004–10.
with a limited caseload. Eur Urol 2011;59:1–6.
[12] Tewari A, Srivasatava A, Menon M, Members of the VIPT. A
[22] Lepor H, Kaci L, Xue X. Continence following radical retropubic
prospective comparison of radical retropubic and robot-assisted
prostatectomy using self-reporting instruments. J Urol 2004;171:
prostatectomy: experience in one institution. BJU Int 2003;92: 205–10. [13] Ficarra V, Novara G, Fracalanza S, et al. A prospective, non-randomized trial comparing robot-assisted laparoscopic and retropubic radical
1212–5. [23] Krambeck AE, DiMarco DS, Rangel LJ, et al. Radical prostatectomy for prostatic adenocarcinoma: a matched comparison of open retropubic and robot-assisted techniques. BJU Int 2009;103:448–53.
prostatectomy in one European institution. BJU Int 2009;104: 534–9.
[24] Campbell SE, Glazener CM, Hunter KF, Cody JD, Moore KN. Conser-
[14] Rocco B, Matei D-V, Melegari S, et al. Robotic vs open prostatectomy
vative management for postprostatectomy urinary incontinence.
in a laparoscopically naive centre: a matched-pair analysis. BJU Int 2009;104:991–5. [15] Kim SC, Song C, Kim W, et al. Factors determining functional outcomes after radical prostatectomy: robot-assisted versus retropubic. Eur Urol 2011;60:413–9.
Cochrane Database Syst Rev 2012:CD001843. [25] Lucas MG, Bosch JLHR, Cruz FR, et al. Guidelines on urinary incontinence. Arnhem, The Netherlands: European Association of Urology; 2012. p. 41. [26] Marchiori D, Bertaccini A, Manferrari F, Ferri C, Martorana G. Pelvic
[16] Khoder WY, Trottmann M, Stuber A, Stief CG, Becker AJ. Early
floor rehabilitation for continence recovery after radical prostatec-
incontinence after radical prostatectomy: a community based ret-
tomy: role of a personal training re-educational program. Anticancer
rospective analysis in 911 men and implications for preoperative
Res 2010;30:553–6.