Efficacy of an assisted low-intensity programme of perioperative pelvic floor muscle training in improving the recovery of continence after radical prostatectomy: a randomized controlled trial


Daniele Tienforti, Urologia, Policlinico ‘Agostini Gemelli’, 00168, Rome, Italy. e-mail: danieletienforti@yahoo.it


Study Type – Therapy (RCT)

Level of Evidence 1b

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

Peri-operative pelvic floor muscle training reduces urinary incontinence for men undergoing radical prostatectomy (RP).

A preoperative biofeedback session, combined with postoperative pelvic floor muscle training, and assisted sessions on a monthly basis only, is an effective low-intensity programme to improve recovery of continence in patients undergoing RP.


  • • To evaluate the efficacy of preoperative biofeedback (BFB) combined with an assisted low-intensity programme of postoperative perineal physiokinesitherapy in reducing the incidence, duration and severity of urinary incontinence (UI) in patients undergoing radical prostatectomy (RP).


  • • A prospective, single-centre, randomized controlled clinical study was designed.
  • • The intervention group received a training session with BFB, supervised oral and written instructions on Kegel exercises and a structured programme of postoperative exercises on the day before open RP. After RP, patients received control visits, including a session of BFB, at monthly intervals only.
  • • The control group received, after catheter removal, only oral and written instructions on Kegel exercises to be performed at home. Patients received control visits at 1, 3 and 6 months after catheter removal.
  • • At each visit the number of incontinence episodes, the number of pads used and patient-reported outcome measures (International Consultation on Incontinence Questionnaire on Urinary Incontinence [ICIQ-UI], [ICIQ]-Overactive Bladder [OAB], University of California, Los Angeles-Prostate Cancer Index [UCLA-PCI], International Prostate Symptom Score-Quality of Life [IPSS-QoL]) were assessed in both groups. All patients were followed-up for a period of at least 6 months after catheter removal.
  • • The primary outcome was the recovery of continence, strictly defined as a ICIQ-UI score of zero.


  • • Overall, 34 consecutive patients were eligible and 32 were available for the final analysis: 16 patients for each study group. The two groups were homogeneous for all pre- and intraoperative features examined.
  • • In the intervention group, continence had been achieved by six, eight and 10 patients at 1-, 3- and 6-month follow-ups, respectively, vs no patients (P= 0.02), one patient (P= 0.01) and one patient (P= 0.002) in the control group at each follow-up, respectively.
  • • The analysis of the UCLA-PCI and ICIQ-OAB scores, the number of incontinence episodes per week and the number of pads per week showed significant differences in favour of patients in the intervention group at 3 and 6 months.
  • • Patients in the intervention group reported better IPSS-QoL scores at all follow-up times but the difference did not reach statistical significance.


  • • Preoperative BFB combined with a postoperative programme of perineal physiokinesitherapy and assisted sessions on a monthly basis only, is a treatment strategy significantly more effective than the standard care in improving recovery of continence in patients undergoing RP.
  • • The impact on QoL appeared less evident, although a trend for a better QoL was observed in the intervention group.



urinary incontinence


radical prostatectomy


post-prostatectomy urinary incontinence


pelvic floor muscle training


quality of life.


Urinary incontinence (UI) remains a clinically important complication after radical prostatectomy (RP) [1]. The reported prevalence of patients with post-prostatectomy UI (PPUI) varies between 1% [2] and 87% [3]. This enormous variability is determined by many factors, e.g. surgical technique, time and methods of assessment and definition of continence [1–4]. There has been much discussion in the literature about preoperative conditions that may predispose some patients to a greater or lesser risk of developing PPUI. The factors that are most often cited are age [5–9], body mass index [7,9,10], size of the prostate [7,10], presence of LUTS [5,8,11,12] and Gleason score [5,8,9]. The aetiology of PPUI has not yet been fully understood, but dysfunction of the bladder neck as well as intraoperative damage of the nerves and sphincter may play a causal role [13–15]. Non-invasive treatments are the first-line therapy for PPUI in the 6 to 12 months after RP. In particular, pelvic floor muscle training (PFMT) is recommended [16–18]; however, the efficacy of this conservative management is not certain. Many studies have investigated the effect of peri/postoperative PFMT, with or without biofeedback (BFB), on the recovery of continence after RP, but results are conflicting. Existing clinical trials, in most cases, are not randomized and controlled, there is no standardization of performed treatment, and the definition of incontinence varies widely from study to study. In several studies PFMT was performed without BFB, or the correct execution of the exercises was not checked by medical staff so it is often difficult to determine whether the training was ineffective owing to inherent inefficiency or because it was incorrectly performed [4,19]. Furthermore, several published studies involved PFMT programmes that were very time-consuming for caregivers, limiting their feasibility and widespread use in everyday clinical practice.

The aim of the present study was to evaluate the efficacy of one session of preoperative BFB, combined with an assisted, low-intensity postoperative programme of PFMT that was not excessively time-consuming for caregivers, in reducing the incidence, duration and severity of PPUI and in improving the quality of life (QoL) of patients undergoing RP.



Between December 2009 and March 2010, 38 consecutive patients who underwent standard open retropubic RP at our institution for clinically localized prostate cancer (cT1a-cT2b) were screened. Patients were included if they could regularly attend an ambulatory schedule. Exclusion criteria were prior diseases with a possible impact on urinary continence, preoperative radiotherapy and any medical condition that could limit participation in the training programme.


A prospective, single-centre, randomized trial was designed. All eligible patients gave informed signed consent before randomization. The randomization schedule was generated by computer and was stratified with a 1:1 allocation. All RP procedures were performed according to Walsh's anatomical technique by three experienced surgeons. Participants were unblinded to treatment assignment, but surgeons and person scoring the evaluation questionnaires were blinded throughout the duration of the study. A nerve-sparing RP was performed in patients with Gleason score ≤ 6, low-stage (T1c) disease and PSA < 10 ng/mL, and the bladder neck was preserved when possible.


The intervention group received a supervised training session with BFB from a dedicated caregiver (T.D.) the day before RP and immediately after catheter removal, and oral and written instructions on pelvic floor muscle contractions and a structured programme of exercises (three sets daily of 10 min each of 5-s contractions then 5-s relaxations) to be performed at home while lying, sitting and standing. Exercise frequency was recorded in a training diary. During the preoperative session of BFB, details were provided about the anatomy and physiology of the lower urinary tract and pelvic floor muscles and wrong execution was corrected. Patients in the intervention group received control visits at monthly intervals after removing the catheter, with a new session of BFB and motivation to maintain adherence to treatment at each visit. The treatment was recommended as long as the patient used pads (safety pad included). For the execution of BFB sessions a professional system for the rehabilitation of UI (PelveenCare®, Coloplast, Bologna, Italy) was used. Subjects were placed in the supine position, with hips flexed to ≈60°, then the anal probe was inserted and a reference electrode was placed on the anterior superior iliac spine. Each assisted BFB session lasted 20 min and was performed by the same operator. Patients in the control group received the standard care at our institution. They were not given formal education on PFMT. After catheter removal, these patients received only oral and written instructions from the urologist on pelvic floor muscle exercises to be performed at home (three sets daily of 10 min each), until recovery of complete continence, without using a diary. Treatment side effects were also recorded. The work was carried out in accordance with the ethical standards of the appropriate institutional committee on human experimentation and with the last revision of the Helsinki Declaration.


All patients were followed-up for a period of at least 6 months after catheter removal. Outcome assessment was performed at each monthly visit for the intervention group and at 1, 3 and 6 months after catheter removal for the control group. All patients were given the following self-administered, validated, condition-specific questionnaires: ICIQ-UI [20], ICIQ-Overactive Bladder (OAB) [20], UCLA-Prostate Cancer Index (PCI) [21], IPSS-QoL [22], by a nurse, blinded to the randomization. The primary outcome measure was the self-reported recovery of continence 6 months after catheter removal. Continence was strictly defined as an ICIQ-UI score of zero. The secondary outcome measures were number of incontinence episodes per week, number of pads used per week, OAB symptoms measured by the ICIQ-OAB score, urinary function measured by the UCLA-PCI score, and the impact of incontinence on QoL measured by the IPSS-QoL score.


The variables considered in the present study were analysed and described as mean (sd) or range for continuous variables, and proportions for dichotomous variables. CIs used were 95%. The differences between groups were analysed using the Mann–Whitney U-test and Fisher's exact test, as appropiate. A two-sided P value of <0.05 was considered to indicate statistical significance. The software used was GraphPad Prism 5 for Windows. Power was based on the primary outcome variable. Using the two-sided test to differentiate between proportions and assuming a type 1 error of 0.05, a total sample size of 30 patients would provide a power of 80% to detect a difference between groups of ≥25% in the proportion of patients achieving continence. Expecting a 10% dropout rate, we randomized 34 eligible patients.


Of 38 patients screened for this study, 34 fulfilled the eligibility criteria, and were randomized into the control group (17 patients) and the intervention group (17 patients). Later, one patient refused to participate in the study because of intolerance to the insertion of the rectal probe and one patient was excluded because of intra-operative surgical complications during RP, leaving a total of 32 patients (Fig. 1). These patients were available for evaluation at all scheduled visits and the postoperative results presented refer to these patients.

Figure 1.

Study flow chart.

The two groups of patients were homogeneous for all preoperative, intraoperative and postoperative characteristics used for the present study (Table 1). Table 2 compares the continence status based on ICIQ-UI questionnaire scores in both groups during follow-up. In the intervention group, continence (ICIQ-UI score = 0) was achieved by six, eight and 10 patients at 1, 3 and 6 months after catheter removal, respectively. Six patients had not achieved continence at the 6-month follow-up. In the control group, only one patient achieved continence. The difference between groups was statistically significant at each reported follow-up time.

Table 1.  Patient characteristics
CharacteristicsControl group, N= 16Intervention group, N= 16 P
 Age (range)67 (60–74)64 (52–74)n.s.
 Mean (sd) body mass index24.4 (2.9)27.0 (3.8)n.s.
 Mean (sd) prostate weight, g38.5 (15.0)33.0 (9.6)n.s.
 Mean (sd) preoperative PSA, ng/mL9.2 (8.5)8.66 (9.0)n.s.
 Preoperative hormonal therapy, n66n.s.
 Preoperative LUTS, n68n.s.
 Previous TURP, n12n.s.
 Surgeons, n   
 Nerve-sparing procedure, n   
  Pelvic lymphadenectomy118n.s.
  Mean (sd) blood loss, mL700 (244)603 (305)n.s.
 Mean (range) no. of days with catheter14.1 (5–25)14.6 (5–26)n.s.
 Postoperative hormone therapy, n44n.s.
 Postoperative radiotherapy, n47n.s.
 Pathological stage, n   
  not evaluable20 
 Gleason score, n  n.s.
  not evaluable32n.s.
 Surgical margins, n   
Table 2.  Comparison of continent patients (ICIQ-UI = 0) in the two groups at 1, 3 and 6 months after catheter removal
Follow-upIntervention groupControl group P

As shown in Figs 2 and 3, analysis of UCLA-PCI and ICIQ-OAB scores showed significant differences in favour of the intervention group at 3- (403.81 vs 272.44, P= 0.006 and 10.12 vs 13.19, P= 0.04, respectively) and 6 months' follow-up (422.50 vs 274.25, P= 0.003 and 9.06 vs 12.62, P= 0.01, respectively).

Figure 2.

Comparison of mean UCLA-PCI scores in the intervention (patients) and control (controls) groups at 1, 3 and 6 months after catheter removal.

Figure 3.

Comparison of mean ICIQ-OAB scores in the intervention (patients) and control (controls) groups at 1, 3 and 6 months after catheter removal.

Also the number of incontinence episodes per week (Fig. 4) and the number of pads per week (Fig. 5) were significantly lower for patients in the intervention group at both 3- (3.84 vs 14 and 1.50 vs 6.25, respectively) and 6-month follow-ups (2.72 vs 13.06 and 1.31 vs 4.625, respectively).

Figure 4.

Comparison of incontinence episodes per week in the intervention (patients) and control (controls) groups at 1, 3 and 6 months after catheter removal.

Figure 5.

Comparison of pads used per week in the intervention (patients) and control (controls) groups at 1, 3 and 6 months after catheter removal.

Patients in the intervention group reported lower IPSS-QoL scores (better QoL) than those in the control group at all follow-up times but the difference was not significant (Fig. 6). No patient complained of any treatment side effect or discomfort during follow-up.

Figure 6.

Comparison of IPSS-QoL mean values in the intervention (patients) and control (controls) groups at 1, 3 and 6 months after catheter removal.


The results of the present study show that a single preoperative supervised training session with BFB, combined with a postoperative programme of PFMT, including assisted training sessions on a monthly basis only, is effective in improving the recovery of continence after open RP. At 3- and 6-month follow-up, the rate of incontinent patients, pad use and number of incontinence episodes per patient and other patient-reported outcomes were significantly worse in the control group than in the intervention group. We observed poor results in our control group, probably owing to two main factors: (i) the very strict definition of continence as having a ICIQ-UI score of zero (zero frequency and no urine leakage and no interference with everyday life); and (ii) the use of only verbal instruction in the control group (in many other studies patients received more detailed instructions and a more intensive protocol for PFMT [23–28].

Few studies have investigated the benefits of treatment that includes preoperative BFB. Bales et al. [23] randomized 100 men scheduled to undergo RP to a group receiving graded PFMT with BFB 2–4 weeks before surgery or to a control group performing PFMT without BFB. The BFB group was instructed to continue exercises four times per day until surgery and to resume exercises after surgery. The control group received written and brief verbal instructions in pelvic muscle exercises before surgery and again after catheter removal. In that study the rate of continence (defined by the use of one pad or less per day) at time points 1, 2, 3, 4 and 6 months after surgery was not significantly different between the two groups. Those authors concluded, therefore, that preoperative BFB training did not improve the outcome of PFMT on urinary continence. Lilli et al. [24] randomized 90 candidates for retropubic RP. Forty-five patients underwent preoperative PFMT and BFB, while the remaining 45 (control group) had preoperative PFMT only. Patients started PFMT 2 weeks before surgery and continued during the immediate postoperative period and thereafter at home. Patients who were completely dry and used no pads were defined as continent. In that study too, the authors observed no significant differences between the two arms during follow-up, but there was a significant correlation between follow-up time and continence improvement in each group. We suggest that not only the different definitions and methods of assessment of continence, but also the type of the intervention and control could explain the discrepancy between the results of the aforementioned studies and those of our study. We believe that the caregiver's postoperative guidance, the regular visits and self-monitoring with exercise diaries established in our intervention protocol were important in that they increased long-term compliance with PFMT and thereby improved continence outcomes. Goode et al. [29], in a randomized controlled trial, recently used a similar postoperative strategy (although with a more intensive intervention including a visit every 2 weeks and the association of PFMT with behavioural therapy), as well as strategies to prevent stress and urge leakage, fluid management, and self-monitoring with bladder diaries. In that study, they showed the effectiveness of their treatment for PPUI persisting even more than 1 year (1–17 years) after surgery. As already noted by Ribeiro et al. [30], in most negative studies [23,24,26,28] control patients received a more intensive regimen of PFMT and significant instruction, quite different from the standard care, and that may have affected the results.

Some studies have shown a positive impact of preoperative BFB on postoperative urinary continence. Sueppel et al. [31], in their study of 16 patients treated with BFB before surgery, compared continence outcomes at 6 weeks and 12 months. The results appeared positive on several measures, but statistical comparisons were not possible owing to the small sample size. Burgio et al. [32] performed a randomized controlled trial comparing preoperative behavioural training to standard care in 125 men, 53–68 years old, who underwent RP. Patients were randomized to one preoperative session of BFB-assisted behavioural training plus daily home exercise or a usual care control condition, consisting of simple postoperative instructions. The main outcome measures were time to continence as derived from bladder diaries, incontinence severity (proportion with severe/continual leakage), pad use, Incontinence Impact Questionnaire score, psychological distress and QoL. In their study, preoperative behavioural training significantly decreased time to continence and the proportion of patients with severe/continual leakage at the 6-month endpoint. There were also significant differences between the groups for self-reported urine loss with coughing, sneezing, and getting up from lying down, but no differences were found on return to work and usual activities or QoL measures.

The results from the study by Van Kampen et al. [33] suggest that ‘individual’ training should be offered to patients after RP. As noted by Novara [34] and Griebling [35], the key points in the studies on post-prostatectomy rehabilitation are the definition of the right frequency of PFMT and whether the same results could have been achieved after interventions with less time-consuming follow-up by the therapist. For example Overgård et al. [36] investigated the effects of PFMT in a randomized controlled trial in which the intervention group was involved in an intensive protocol of PFMT, including weekly training sessions guided by a physiotherapist or by a DVD for those patients living far from the hospital. After 6 months, there was a clinically relevant difference in continence status between the active group (79% continent patients) and the control group (58% continent patients; P= 0.061). Twelve months after RP, the difference became clinically and statistically significant: 92% of patients in the active group were continent compared with 72% of patients in the control group (P= 0.028). The study concluded that continence rates were similar 3 months after RP in patients performing intensive PFMT with or without follow-up instructions by a physiotherapist, but in the period up to 1 year the group receiving physiotherapist-guided training showed a significantly lower UI rate than patients training on their own. As noted by Lazzeri [37], Overgård's findings suggest that physiotherapist guidance increases long-term adherence to PFMT and thereby improves continence rates over time more than information provided to patients for training on their own. It means that the long-lasting presence of a physiotherapist might improve the capacity of the patient to re-recognize his pelvic floor musculature and to re-locate sensations after surgery by an ‘afferent’ modulation on pelvic innervation. From a resource standpoint, however, a high-intensity regimen is difficult to implement in clinical practice and may lead to a high patient drop-out rate.

In the present study, the use of a low-intensity, supervised programme requiring control visits including a BFB session on a monthly basis only, thus less time-consuming for caregivers, more feasible for peripheral urological clinics and probably more cost-effective, yielded a significant improvement, compared with standard care, in recovery of continence that was already evident at 1 month after surgery.

The QoL value, measured by the IPSS-QoL score, was not significantly better in the intervention group, although there was a trend for better QoL in this group. Other randomized controlled trials on PPUI have not shown an improvement in QoL measures, even when better continence outcomes were achieved in the intervention group [26,32,38]. The authors of those studies suggested that patients may have found ways to circumvent the impact of incontinence on well-being or confounded differences in incontinence with other QoL issues such as recovery from surgery, anxiety about cancer or sexual dysfunction [32].

The limitations of the present study are the relatively small number of patients and controls and the short follow-up period of 6 months. Although our results are very clear at 6 months, a longer follow-up would have provided more information on the long-term effect of our PMFT programme, although this should take into account the fact that in several other studies there is a point at which continence recovery in intervention and control groups begins to converge.

In conclusion, this prospective randomized study showed that a preoperative BFB, combined with our assisted, low-intensity programme of PFMT, is a significantly more effective treatment strategy than standard care in improving recovery of continence in patients undergoing RP. The impact on QoL appeared less evident, although a trend for a better QoL was observed in the intervention group. We recommend this not excessively time-consuming programme of PFMT as a valid alternative to the standard care of providing patients undergoing RP only with oral and written information on pelvic floor exercises.


None declared.