Literature review of factors affecting continence after radical prostatectomy.

Radical prostatectomy (RP) is the most common cause of stress urinary incontinence (UI) in men. Several anatomic structures affect or may affect urinary continence - urethral sphincter, levator ani muscle, puboprostatic ligaments, bladder neck, endopelvic fascia, neurovascular bundle - and understanding of the anatomy of pelvic floor and urethra is crucial for satisfactory functional outcome of the procedure. Surgical techniques implemented to improve continence rates include nerve-sparing procedure, bladder neck preservation/plication, urethral length preservation, musculofascial reconstruction, puboprostatic ligaments preservation or seminal vesicle preservation. Perioperative (preoperative and postoperative) pelvic floor muscle training (PFMT) aims to shorten the duration of postoperative UI and thus, improve early continence rates postoperatively. In the review, complex information regarding anatomical, intra- and perioperative factors affecting urinary continence after RP is provided, including description of important anatomical structures, possible implications for surgical technique and evaluation of different PFMT strategies in perioperative period.

Radical prostatectomy (RP) is the most common cause of sphincteric urinary incontinence (UI) in men. Advances in operative technique have reduced the rate of post prostatectomy incontinence (PPI). However, the burden of PPI remains high and is even expected to rise due to the increasing number of procedures performed.1 Besides stress UI, overactive bladder symptoms caused by detrusor overactivity and impaired detrusor compliance de novo may occur during the first 12 months after RP.2 Early PPI of different grades is higher than thought and may affect up to 96% of patients.3 Mean continence rates 12 months after surgery are 89-100% for robot-assisted laparoscopic prostatectomy (RALP) and 80-97% for open radical retropubic prostatectomy (RRP).4 Post prostatectomy incontinence affects both physical activity and social well-being and therefore has a significant impact on the quality of life.5 The precise aetiology of PPI has not been completely understood. There are several risk factors, including preoperative (age, continence status prior to surgery), intraoperative (surgical technique, surgical experience), and postoperative factors.6 Several factors and limitations must be considered in interpreting the assumed outcomes of RP including PPI - individual surgeon factors (level of surgical ability, completion of the learning curve, the same surgical step performed differently), variable definition of continence/incontinence, use of validated tool to evaluate PPI, methods used for data collection, clinical stage or biopsy parameters and patient’s age.7 In general, significantly better continence rates after RP are observed in men under 70 years of age than in men over 70 years.8 There has not been sufficient evidence suggesting a significant relationship between body weight and postoperative RP, and prostate volume and PPI.9 Neither has a relationship between preoperative physical activity and PPI been observed.10 Preoperative detrusor overactivity (DOA) is associated with higher risk of post prostatectomy incontinence. Preoperative detrusor overactivity was found in 34% of patients still incontinent 6 months after RP on baseline urodynamic evaluation before surgery.11 Several treatment options have been proposed for the management of PPI; pharmacotherapy, pelvic floor muscle training (PFMT), bulking agents, fixed or adjustable male slings, and compression devices including the artificial urinary sphincter (AUS). Whereas different strategies can be applied to patients with mild-to-moderate PPI, AUS is considered the standard treatment for moderate-to-severe PPI.12 Due to different surgical approaches/modifications and PFMT strategies, the aim of the review was to summarise the anatomical, surgical, and perioperative factors that affect, or may affect, postoperative urinary continence after RP.

Anatomical aspects. Puboprostatic ligaments

Puboprostatic ligaments are paired fibrous streaks that originate from the endopelvic fascia. They attach to the lower fifth of the pubic bone, lateral to the symphysis and the junction of the external urethral sphincter and prostate.13 They support the external sphincter and sustain the urethra in its position in the pelvic floor. It is still unclear whether they are a part of a muscle or not and whether these ligaments also contain muscle fibres.14 Because of the direct connection between the anterior bladder wall and the pubic bone (), the term “pubovesical” ligaments has been proposed.15 The intactness of the urethral suspensory mechanism appears to have a relevant role in the preservation of urinary continence - a puboprostatic ligament-sparing approach allows the preservation of the maximal urethral length and the anterior urethral support remains intact, leading to an earlier return of continence.16 Stolzenburg observed a significant decrease in the period to early continence in the group of patients with puboprostatic ligament-sparing nerve-sparing endoscopic extraperitoneal RP (nsEERP) compared with standard nsEERP. No difference was found between the 2 groups after 3 months (no PPI recorded).17

Detrusor apron ending with pubic insertion - puboprostatic/pubovesical ligament. B - bladder, PB - pubic bone, P - prostate

External urethral sphincter

The external urethral sphincter (EUS) surrounds the membranous urethra (). Its location corresponds with the site of the peak urethral closing pressure and is considered the principal structure ensuring continence after RP.13 Although the term “striated sphincter” has been widely used, the EUS contains both smooth muscle bundles (inner longitudinal layer and outer circular layer) and striated muscle bundles (that form a “true” rhabdosphincter separated from the muscles of the pelvic floor).18 The EUS is horseshoe (or omega) shaped and does not connect dorsally. Despite this configuration, urethral pressure at the external sphincter increases uniformly along the entire circumference during bladder filling. The striated component exerts its function from the prostate apex to the bulb, whereas the inner muscle component extends to the verumontanum19 (). The EUS is supplied by 2 somatic nerves - the pudendal nerve and a branch of the sacral plexus running on the surface of the levator ani muscle (that is why transection of the pudendal nerve does not lead to ablation of sphincter activity).13 Although pelvic floor muscles and the internal urethral sphincter participate in the continence mechanism, protection of the EUS should always be the main goal of the surgeon.14 Urethral pressure profilometry (UPP) shows a significant decrease in both maximum urethral closure pressure (MUCP) and functional profile length (FPL) after RP, and lower preoperative MUCP and FPL are associated with increased risk of PPI.20 Another risk factor for PPI and longer time to achieve continence is the shorter length of the urethral sphincter on preoperative endorectal MRI.21 Conversely, a longer preoperative or postoperative membranous urethral length measured by endorectal MRI is associated with superior continence.22

Male urethral sphincter image showing: A) the internal urethral sphincter. B) the external urethral sphincter.

The relationship between the prostatic apex and the external urethral sphincter. SMS - smooth muscle sphincter, SS - striated sphincter, PA - prostatic apex, CS - colliculus seminalis.

Internal urethral sphincter (vesical sphincter, musculus sphincter vesicae)

The smooth muscle at the level of the bladder neck is distinct from the rest of the bladder. The middle muscular layer with circular fibres forms a preprostatic sphincter, which is generously supported by adrenergic innervation. Stimulation of adrenergic fibres produces an effective closure of the bladder neck, which avoids retrograde ejaculation and assures continuous urinary continence. It maintains urinary continence even in the case of a destroyed external urethral sphincter.13

Pelvic floor muscles

The levator ani muscle is the innermost muscle of the anterior pelvis. Its anteromedial component, located next to the urethral sphincter, is called the pubourethral muscle, although the term “puboperinealis” muscle has also been used.9 This muscle is relaxed during voiding. During the volitional contraction of the pelvic floor (referring to the attempt to stop the urinary flow during voiding), the urethrovesical junction and anorectal junction move upwards and forwards, the bulb of the penis moves ventrally and there is a small displacement of the ventral urethral margin dorsally at the level of EUS, as demonstrated by perineal ultrasound.23 Apical prostatic dissection during RP may damage the pubourethral muscle fibres and pudendal nerve branches innervating EUS, which run close to the pubourethral muscle.24 There is a relatively solid fascia or interface between the levator ani and the EUS, which also contains veins and nerves (originating from the periprostatic neurovascular bundle).25 Preservation of the levator ani fascia (a part of endopelvic fascia) protects the innervation of the levator ani muscle and EUS.26 Other components of the levator ani muscle include the puborectalis muscle (PRM) located laterally and the rectourethralis muscle (RUM) located between the perineal body and anorectal junction (). The anterior part of PRM inserts to the inferior pubic rami and the posterior parts connect to the anorectal junction. This muscle is important for the closure of vaginal and anal orifices, but its contribution to urethral pressure increase has also been demonstrated in the rabbit.27 The relationship between the perineal body and RUM remains controversial and little is known of the histological relationship between EUS and RUM - the membranous urethra is attached to the rectourethralis muscle through EUS, which is closely fixed to the ventral portion of RUM.28 However, other studies failed to demonstrate any contact of RUM with the urethra itself and suggest that the term is a misnomer.29 When pelvic anatomy was compared by MRI study before and after recovering urinary continence after RP, both an increase in PRM thickness and movement of the bladder neck upwards and forwards was observed.30 There was no significant change in the membranous urethral length. These results support the use of periurethral suspension and pelvic floor exercises for the resurgence of continence after RP.

Components of the levator ani muscle. RU - rectourethralis muscle, PP - puboperinealis muscle, PR - puborectalis muscle, SS - striated sphincter, C SMS - circular smooth muscle sphincter, L SMS - longitudinal smooth muscle sphincter, PV/PL - pubovesical/puboprostatic ligament, PB - pubic bone, DVC - dorsal venous complex

Fascias around the prostate

The periprostatic fascia is located between the levator ani fascia and the prostate. In the past, it was thought that this fascia was divided into 2 layers, but histological studies have discovered that the periprostatic fascia is a multilayered connective tissue comprising collagenous fibres, fat tissue, nerves, and blood vessels.31 Traditionally, the standard nerve-sparing (NS) RP is based on interfascial dissection between the lateral layer and the medial layer, whereas the intrafascial technique attempts to get as close to the prostatic capsule as possible (). Despite extensive research of prostate anatomy, the exact anatomy of the fascias around the prostate, as well as the exact relationship between the neurovascular bundle and the fascia remain controversial.29

Interfascial (A) and intrafascial (B) nerve-sparing radical prostatectomy. EF - endopelvic fascia, PF - prostatic fascia, PC - prostatic capsule, NVB - neurovascular bundle, PP - prostatic pedicle

Technical aspects

Various operative techniques have been proposed to improve postoperative continence rates. The basic concept of these techniques is to maintain the normal anatomy and function of pelvic structures as much as possible by their preservation, reconstruction or reinforcement.32 The surgeon’s experience is an important factor, as outcomes of RP are sensitive to small differences in performance.33

Bladder neck preservation/bladder neck plication

ladder neck excision may harm the internal urinary sphincter. To preserve the bladder neck (bladder neck preservation [BNP]), one should sharply dissect the bladder neck off the base of the prostate to maintain most of the circular muscle fibres. This technique seems to hasten the recovery of urinary function; however, in the long term, the continence rates with or without BNP are almost the same.34

Plication of the bladder neck is an effective and relatively simple technical modification to shorten the period of restoration of urinary continence after RP.35 The anterior bladder plication stitch decreases the extension of the bladder neck and EUS at rest and increases the functional urethral length.

Urethral length preservation

To achieve the maximum urethral length, it is important to preserve both EUS and the intraprostatic portion of the membranous urethra. Full-functional urethral length preservation achieved by modified apical preparation increases early continence rates, with a possible maximal effect in patients with a long intraprostatic part of the membranous urethra.36 Between 10-40% of the functional urethra is actually covered by the prostatic apex. Thus, the shape of the prostate can make urethral length preservation difficult, especially in the case of circumferential overlap of the apex over the EUS.37 An advantage in long-term continence rate has not been demonstrated. Precise recognition of the junction between the membranous urethra and the prostatic apex is the most crucial point to maintain maximal urethral length without increasing the risk of positive apical surgical margins.32 Some authors believe that it is better to cut the urethra just as it emerges from the prostatic apex or, if possible, with some dissection into the prostatic apex without any distal dissection, which will not make the actual urethra longer or the patient more continent, but will instead only cut the urethral blood supply and its innervation.38

Musculofascial reconstruction

Anterior reconstruction techniques are characterised by reinforcing the anterior support of the urethra. This is typically achieved by anchoring the vesicourethral anastomosis to the pubic bone or the puboprostatic ligaments.39,40 A single puboperiurethral stitch passed between the urethra and dorsal venous complex and then through the periostium of the pubic bone results in faster recovery of continence and better continence rates at 3 months compared with no periurethral suspension.41 After the urethra is supported ventrally, the angle of the vesicourethral junction is not too obtuse and the anastomosis is stable - an unstable anastomosis may lead to enhanced scar formation with possible anastomotic stricture as a consequence.42 Periurethral fibrosis might also impede the recovery of continence by altering the elasticity of EUS.22

Preservation of the puboprostatic collar (formed by the puboprostatic ligaments, the arcus tendineus of the pelvis and the puboperinealis muscle) is performed by careful separation of the apex from the puboprostatic collar complex. In combination with puboperineoplasty (suspension of the vesicourethral anastomosis to the collar - not only to the puboprostatic ligaments - by 3 sutures on each side), it results in an immediate continence rate of 41% after catheter removal and 71% after 4 weeks.24

The effect of posterior musculofascial plate reconstruction on earlier continence recovery is promising but still controversial.43 The aim of posterior reconstruction is to prevent the retraction of the urethra and EUS. It is originally based on the technique described by Rocco et al44 “Rocco stitch”, which includes 2 sutures joining the posterior semi-circumference of the sphincter to the residual Denonvillier’s fascia on either side (cranial elongation of the dorsal sphincteric wall). This plane is then fixed to the posterior bladder wall with other 2 sutures placed 1 to 2 cm cranial and dorsal to the new bladder neck (EUS is anchored to the posterior aspect of the urinary bladder). The urethro-vesical anastomosis is subsequently performed.44 Regarding postoperative continence rates, only a small statistical advantage after 1 month has been shown; the technique does not influence 3- and 6-month urinary continence.7 Nevertheless, the reconstruction provides greater support to the vesicourethral anastomosis and may improve hemostasis. However, care must be taken not to injure the nerve fibres that run along and through the RUM with the stitches placed during the reconstruction of the dorsal musculofascial plate.28

Total musculofascial reconstruction combines the posterior reconstruction (as described previously) and anterior reconstruction (re-attachment of the tendinous arch of the pelvis and the puboprostatic plate to the bladder neck).45 The cumulative analysis of comparative studies showed a minor statistically significant benefit of total (anterior and posterior) musculofascial reconstruction on urinary continence 1 month and 3 months after RP.7 No differences were found after a longer follow-up.

Seminal vesicle preservation

Some of the neural fibres from the inferior hypogastric plexus (pelvic plexus) that supply autonomic smooth muscle fibres of the EUS are located posterolateral to the seminal vesicles and pass very close to their tips.46 The risk of injury to these nerves can be reduced using the seminal vesicle-sparing technique. Improved postoperative continence rates were observed as well.47

Local hypothermia

The risk of injury to neuromuscular tissues surrounding the prostate can be decreased using local hypothermia. Cold intracorporeal irrigation along with an endorectal cooling balloon lead to effective regional pelvic cooling.48 Despite some positive references regarding postoperative continence rates and return to continence (particularly in older patients),49 this method is rarely used in clinical practice and therefore should be considered experimental.

Nerve-sparing RP

Although there is robust evidence that a NS RP is important for preservation of erectile function, there is controversy over whether the NS technique improves postoperative urinary incontinence.32 Comparing bilateral NS, unilateral NS, and non-NS RALP, no significant difference was found in continence rates at one year after surgery. This suggests that baseline factors and not the physical preservation of the cavernosal nerves predict overall return to continence.50 In comparison of interfascial and intrafascial NS RP, no statistical significance in continence rates was observed between the 2 groups at 12 months.51 In addition, postoperative erectile function is not predictive of urinary continence, suggesting that anatomical factors, rather than innervation, are primarily responsible for continence after RP.9

In conclusion, the role of surgical modifications of RP remains controversial. Rather than strengthening the continence mechanism, they focus on restoration of anatomical structures to their original state as it was before the surgery. It is still unclear how all of these anatomical structures interact. It is well known to all surgeons that even if a particular anatomical structure (for example, EUS) remains intact, some patients are still incontinent postoperatively.14 On the other hand, we have our own experience with a post-poliomyelitis patient, who is fully continent 5 years after open RP, with excellent both oncological and functional outcome, despite bladder symptoms including incontinence being found twice as often in polio survivors than in the general population and incontinence affecting 41% of men.52 Moreover, muscle strength slowly deteriorates in post-polio patients.53 PPI would therefore have definitely been anticipated in this patient. Nevertheless, data on pelvic floor muscle function in post-polio patients are lacking.

Open RP (ORP) versus RALP

Various studies have compared these 2 surgical approaches in terms of PPI. Some of them showed better results in terms of postoperative continence rates and time to continence in RALP groups;54,55 others did not confirm the results.56,57 A prospective study by Geraerts et al58 showed that patients after RALP were prone to recover urinary continence earlier than those after ORP. However, statistically significant difference in continence rates was lost in subgroup analyses. Therefore these results must be interpreted with caution. There was no difference in PPI rates at 12 months; severity of voiding symptoms and quality of life were significantly better in the RALP subgroup.58 Similarly, O’Neil et al59 reported improved urinary function at 6 months, but not at 12 months, in a population-based study including patients treated with surgery for prostate cancer within the Comparative Effectiveness Analysis of Surgery and Radiation (CEASAR) and Prostate Cancer Outcomes Study (PCOS) prospective studies.59

Pelvic floor muscle training

The PFMT improves the function of the pelvic floor, improving urethral stability.60 During increased activity, the urethra is stabilised by increased urethral closure pressure and downward movements are minimised. Several studies consistently demonstrated the benefit of early postoperative PFMT on recovery of PPI.61,62 The rehabilitation program usually includes pelvic floor contractions, controlled either manually by the therapist or by electromyography biofeedback. After initial guided PFMT, patients continue with a home program of several series of exercises per day until continence is achieved. Biofeedback compared with unassisted PFMT has demonstrated statistically significant superior outcomes.63 The necessity of physiotherapist-guided follow-up PFMT is questionable; patients who are instructed by a physiotherapist seem to adhere longer to PFMT and thereby improved continence rates are recorded, compared with patients training on their own.64 Reports on less intense therapy (instruction and telephone support versus intensive guided PFMT) without significant difference at any time point, permit a different (and more cost-effective) strategy.65 The effect of preoperative PFMT (in combination with postoperative PFMT) has also been investigated. Most studies found positive results.66-68 However, due to a lot of bias, their results should be interpreted cautiously. A recent randomised controlled trial failed to demonstrate shorter duration of PPI in patients with additional preoperative PFMT (3 sessions according to the waiting time for surgery), compared with patients with only postoperative PFMT.69 However, due to patients’ satisfaction with PMFT before surgery, postoperative incontinence had less impact on quality of life in the preoperative PFMT group. Electrical stimulation of the pelvic floor (stimulation of the pudendal nerve and its branches) combined with PFMT does not improve the return to continence more than PFMT alone.70 Behavioural therapy, including PFMT and bladder control strategies, is suitable for patients with persistent urinary incontinence after surgery (more than one year) and resulted in fewer incontinence episodes. Biofeedback and electrical stimulation did not further improve the effectiveness.71 The most recent systematic review and meta-analysis of the effect of preoperative PFMT on PPI including 11 studies, confirmed improvement of postoperative continence at 3 months (36% reduction of PPI risk) but not at 6 months after surgery.72

Early postoperative pharmacotherapy

Duloxetine, a potent inhibitor of the presynaptic re-uptake of serotonin and norepinephrine, has been used for the treatment of stress UI. One study compared PFMT plus duloxetine versus PMFT plus placebo for 16 weeks early after RP (starting on day 10 after catheter removal) followed by 8 weeks of PFMT alone. The PPI was significantly improved in the former group, but the effect did not last towards the end of the study (week 24), indicating that duloxetine accelerates cure rather than increases the number of patients cured.73 Detrusor overactivity and impaired bladder compliance after RP are the rationales for the use of antimuscarinic agents. A study comparing solifenacin versus placebo in patients after RALP who were incontinent 1 to 3 weeks after catheter removal failed to demonstrate difference in time to return to continence. However, there was a significant difference in the number of continent patients at the end of the study (week 12) favoring the solifenacin group.74

In conclusion, post prostatectomy incontinence may be influenced by many factors, including patient’s aspects, surgeon’s experience, operative technique, and continence definition/methodological aspects. There is still much to be known regarding the male continence mechanism, the role of specific structures in maintaining continence and the precise aetiology of post prostatectomy incontinence. Prediction of PPI is therefore difficult. The article reviews current anatomical knowledge and intra- and perioperative management suggested to improve continence rates after RP. Although these modifications shorten the time to continence and improve early continence rates, the long-term continence rates remain almost the same as for the standard anatomical RP.