Effectiveness of preoperative pelvic floor muscle training for urinary incontinence after radical prostatectomy: a meta-analysis

Radical prostatectomy (RP) is the most common treatment for patients with localized prostate cancer[1]. In addition to traditional open RP, currently more advanced surgical methods, such as robot-assisted laparoscopic RP and laparoscopic RP, are widely used. However, despite improvements in surgical technique, urinary incontinence (UI) remains a significant and bothersome sequela after RP that may dramatically worsen a patient’s quality of life[2–4]. UI after RP has been attributed to urethral sphincter deficiency or injury, and to bladder dysfunction such as detrusor overactivity, impaired bladder-filling sensation, and low bladder compliance[5]. Therapeutic strategies for UI include conservative treatment, pharmacotherapy, penile clamping, and artificial urinary sphincter[6]. A conservative treatment approach to UI has the advantage of noninvasiveness. The theoretical basis of pelvic floor muscle training (PFMT) is that repeated volitional contractions of selected pelvic floor muscles may improve their strength and efficiency during periods of increased intra-abdominal pressure, and PFMT has been shown to be effective in relieving stress UI in women[7]. However, despite the popularity of PFMT as conservation treatment for UI, a Cochrane review concluded that there was insufficient evidence to show whether it was effective or not[8]. The reason may be the duration of PFMT was not long enough. Actually, many studies reported that their study patients started PFMT after catheter removal[9–12]. However, others investigated the effects of preoperative PFMT on the duration and severity of UI for patients with RP and had positive results[13–16]. But a recent randomized controlled trial (RCT) in which patients started PFMT 3 weeks before surgery[17] had results that were inconsistent with other studies, and another RCT, in which patients began PFMT with biofeedback 4 weeks before surgery even appeared to show greater benefit for the control group[18]. These contradictory results may be attributable to sample size and/or RCT quality. It is therefore essential to determine whether patients with additional preoperative PFMT regain urinary continence earlier than patients with only postoperative PFMT after RP. As such, we intended to provide the best available evidence through a strict meta-analysis. We hypothesized that additional preoperative PFMT would have a positive effect on urinary incontinence after RP, at least at the early recovery stage.

Our meta-analysis attempted to clarify the controversial issue of whether additional preoperative PFMT would hasten the resolution of urinary incontinence after RP. After a comprehensive search of electronic databases and strict assessment of quality, five RCTs were deemed qualified for inclusion. However, the pooled analysis suggested no benefit from additional preoperative PFMT for UI after RP at 1- and 3-, 6-, and 12-month time points, which represented early, interim, and late recovery phases in the present analysis.

Urinary incontinence is most severe in the early postoperative phase; therefore, two of the five included studies[16, 21], which were conducted to determine the effectiveness of preoperative PFMT in the early recovery stage (within 3 months) indicated significantly decreased duration and severity of early urinary incontinence after RP. These positive results formed the foundation of our hypothesis. However, results of the pooled analysis were inconsistent with our hypothesis and indicated there was no benefit from additional preoperative PFMT, even at the very early stage of recovery (within 1 month). It must be pointed out, however, that the findings should be interpreted with caution because of the unavailability of evidence regarding additional preoperative PFMT and considerable between-study clinical heterogeneity, one aspect of which is the different durations of preoperative exercise. Two trials[16, 21], which started PFMT 4 weeks before surgery, had findings in favor of the experimental group, whereas two other trials commenced exercise 3 weeks[17, 20] and 2–4 weeks[17, 20] before surgery but did not find a significant benefit from preoperative PFMT. The inconsistent results may be because of the different durations of preoperative exercise; it can be assumed that while a short duration of exercise provides time for awareness of the pelvic floor muscles, a longer preoperative exercise period to enhance strength and endurance would provide greater benefit[17]. Another critical issue is the frequency of PFMT. Novara[25] points out that the greater the frequency of PFMT, the better its efficacy, regardless of the method chosen.

Considering the variation among studies and that the data for time to continence and quality of life were insufficient for a pooled analysis, we provided only a narrative description of the secondary outcomes. Outcomes between different trials were conflicting, and this meta-analysis was unable to provide a consistent conclusion. The two trials[17, 21] reporting the outcome of time to continence used different measurements and definitions of continence, and produced entirely different results. Geraerts et al.[17] indicated that the median time to continence was almost the same for both groups, while Patel et al.[21] found that a PFMT program begun preoperatively could significantly reduce the duration of early incontinence. More evidence is needed to corroborate these results. Similarly, the outcome of quality of life reported by three trials[16–18] will also require further investigation to resolve these inconsistent results. Geraerts et al.[17] found the impact of incontinence on quality of life to be smaller in the preoperative-exercise group but no explanation was given for the better quality-of-life scores obtained by the experimental group. The authors also evaluated satisfaction with preoperative PFMT, and all patients in the experimental group expressed satisfaction at receiving PFMT before surgery despite the lack of reduction in their postoperative duration of incontinence. The study by Centemero et al.[16] showed that 75% of patients in the intervention group reported a high degree of satisfaction with starting PFMT before surgery. Considering that UI is a complication that causes patients a great deal of stress and has a particularly negative impact on quality of life, any intervention that can shorten its duration is worth trying. More importantly, men receiving additional PFMT before surgery showed a high degree of satisfaction, so we still recommend PFMT, with its advantage of noninvasiveness, as the main strategy for prevention and treatment of UI after RP even though our meta-analysis did not clearly demonstrate a benefit.

As is often the case with a meta-analysis, our study has several limitations that could bias the final results. First, we included only five studies, based on our strict inclusion criteria, which was not enough to draw a strong conclusion. Second, funnel plots were not created to assess publication bias because of the limited available evidence; we merely included the studies published in English, which may widen the publication bias and affect the analysis. The third major limitation was the clinical heterogeneity caused by non-standardized treatment regimens and outcome measures. Treatment regimens varied between studies in surgical technique, frequency of PFMT (four times or 30 min daily), definitions of continence and incontinence, durations of follow-up (3, 6, or 12 months), timing of initiation of preoperative PFMT (2–4 weeks before surgery), and outcome measurements (24-h pad test, 1-h pad test, different questionnaires). For example, in our meta-analysis, the surgical techniques included open prostatectomy, and laparoscopic and robit-assisted laparoscopic radical prostatectomy, however, because of an insufficient number of qualified RCTs, we could not perform a subgroup analysis by category of surgical technique, which was a source of bias affecting the results. It is noteworthy that factors associated with resolution of UI are complicated and include a patient’s age, tumor stage, and history of previous lower urinary tract dysfunction[26]. These probably also decrease the credibility and validity of the results of this meta-analysis.

Our meta-analysis suggests that additional preoperative PFMT did not improve the rate of reestablishment of continence after RP at the early (within 3 months), interim (6-month), or late (1-year) stage. However, we were unable to draw a definitive conclusion regarding the outcomes of time to continence and quality of life. Further high-quality RCTs with larger samples, standardized treatment regimens, and credible outcome measurements are warranted to confirm or refute our results.

Wei Wang, PhD, Director of Nursing Education Center, Head Nurse of Department of Urology, The First Affiliated Hospital, School of Medicine, Zhejiang University.

Qing Mei Huang, BSN, Nursing graduate student of Nursing Education Center, The First Affiliated Hospital, School of Medicine, Zhejiang University.

Feng Ping Liu, MSN, doctoral student of nursing, Nursing Education Center, The First Affiliated Hospital, School of Medicine, Zhejiang University.

Qi Qi Mao, MD, PhD; Attending Physician, Department of Urology, The First Affiliated Hospital, School of Medicine, Zhejiang University.

Wei Wang and Qing Mei Huang are co-first authors.