Complete puborectalis, puboperinealis muscle and urethral rhabdomyosphincter preservation in laparoscopic radical prostatectomy: Anatomical landmarks to achieve early urinary continence

Objectives To describe our surgical technique of “muscle‐sparing” laparoscopic radical prostatectomy and to review relevant anatomical landmarks during the procedure. Methods This was a prospective non‐controlled case series of 120 consecutive patients who underwent laparoscopic radical prostatectomy, always carried out by the same surgeon (OL). The median follow‐up period was 33 months. Dissection of the puboperinealis and puborectalis muscle consists of the precise dissection of the puborectalis and puboperinealis muscles from the periprostatic fascia. Rhabdomyo‐dissection consists of an approach that spares the external urethral sphincter from the ventral surface of the prostate and membranous urethra. Clinical data were collected in a dedicated database. Intraoperative variables, postoperative complications and outcomes of urinary continence were assessed. A descriptive statistical analysis was carried out. Results Continence rates were 70.8%, 83.3% and 92.5%, at 0–2, 3–4 and 5–8 weeks after removal of the urethral catheter, respectively; 96.6% and 98.3% at 6 and 12 months after surgery. The positive surgical margin rate associated with rhabdomyo‐dissection was 8.3%. Conclusions Laparoscopic radical prostatectomy with dissection of the puboperinealis and puborectalis muscle, and rhabdomyo‐dissection is an oncologically safe procedure, associated with very early recovery urinary continence in most patients. It is a technique that can be applied in most cases, as long as there is no invasion of the ventral side of the prostate.


Introduction
Post-radical prostatectomy urinary incontinence is a frequently found complication, and is found in 1-47% of patients, depending on the series. [1][2][3] Furthermore, in 5-25% of such cases, further surgical procedures are required to correct the condition. 4 Multiple surgical techniques have been described, for instance, techniques with preservation of: Retzius space, bladder neck, VVSS, nerve bundles, puboprostatic ligaments, deep dorsal penis venous complex, maximal urethral length, endopelvic fascia, detrusor apron and anterior suspension of the bladder neck, and also techniques with posterior urethral reconstruction: Rocco stitch, Denonvilliers' fascia reconstruction, pubourethralis ligament reconstruction, endopelvic fascia preservation and arcus tendinous fascial reconstruction, all of them with the aim of avoid post-surgical urinary incontinence, but none have yielded more than either uncertain or unreproducible results. 2,[5][6][7][8][9][10][11] In most cases, urinary incontinence is caused by damage, either to the intrinsic or extrinsic muscle systems that play an active role in urinary continence, as well as damage to the vascular/nerve pedicles involved in the process. 3,[12][13][14][15][16][17][18] In this study, we present a simple technical variation, aimed at sparing the muscle systems and vascular/nerve pedicles in order to achieve high rates of urinary continence in the early postoperative period (the first 2 months after surgery) for such patients.

Methods
The present study was a prospective, non-controlled case series of 120 consecutive patients who underwent LRP, always carried out by the same surgeon (OL).
The ethics committee of Bellvitge University Hospital, Institut Catala de la Salut, Generalitat de Catalunya, Barcelona, Spain approved the prospective collection of data (approval number: PR145/19 (CSI 19/24)), and all patients provided written informed consent.
Patients with organ-confined prostate cancer were considered candidates for LRP and included in this study.

Surgical technique
1. Laparoscopic approach to the pelvic subperitoneal space.
We use a 30°laparoscopic lens. We carry out a blunt dissection of the space of Retzius, sparing the parietal peritoneum and its contents.
2. DPPM: After cutting the endopelvic fascia, we proceed to a precise dissection of the puborectalis muscle from the periprostatic fascia until reaching medial fibroadipose tissue. Using the 30°laparoscopic lens, the surgeon is able to change the view to 180°and thus carry out a careful dissection of the puboperinealis muscle from the underlying perineal fascia and the lateral wall of the puboprostatic ligaments (Fig. 1). In this phase, in most cases, subtotal sectioning of the puboprostatic ligaments is required (Fig. 2). Subsequently, the surgeon proceeds to the ligation of the deep DVP, using a hemostatic suture. 3. Section the neck of the bladder with scissors, and dissect and section the cervical urethra. Then section the vesicoprostatic muscle and dissect the VVSS.
4. Working with scissors, section the ventral and dorsal layers of the Denonvilliers' fascia, followed by the dissection of the prerectal space through to its most caudal portion, where we find the perineal body (Fig. 3).
5. Anterolateral approach to the neurovascular bundle with dissection and complete preservation thereof in those cases in which it is indicated.
6. Dissection and clamping with 17-mm and 13-mm Hemo-lok of the successive dorsolateral neurovascular pedicles through to the apex of the gland.
7. RMD: This consists of the dissection of the EUS apron from the ventral surface of the gland, advancing along the cleavage plane between the periprostatic fascia, the ventral slope and the McNeal AFMS, below the vascular plane, until reaching the beginning of the membranous urethra, which we carefully dissect from the interior of the PA in order to "recover" the fibers of the rhabdomyosphincter from the inside of the PA. With this procedure, we virtually preserve ad integrum the total length of the membranous urethra and its corresponding sphincter complex -EUS (Figs 4-6).
8. Vesicourethral anastomosis with V-Lok 3/0 15-cm barbed sutures; for the urethra, the stitches should be placed caudally to the EUS to avoid tearing the sphincter while, dorsally, they should include the perineal body (basal plate) to "incorporate" the membranous urethra together with the EUS, as part of the system of abdominopelvic pressure, thus facilitating adequate urinary continence.

Clinical, pathological and surgical variables assessed
For each patient, we prospectively collected the following clinical, pathological and surgical data.

Clinical data
Age, BMI, PSA (ng/mL), rectal examination, Gleason biopsy score, lymphatic and/or neurovegetative tumor permeation, multiparametric prostate MRI, erectile function and D'Amico Risk Classification data were collected.

Pathological data
Gleason specimen, pTNM, perineural permeation and lymphovascular permeation data were collected. With respect to PSM, we proceeded to: (i) globally assess PSMs; (ii) RMD dependent PSMs, when specifically affecting the anterior face of the prostate apex, between 8 and 4 o'clock.

Surgical data
Time of surgery, NVBP, bladder neck preservation and urethral length (preoperative multiparametric prostate MRI assessment) data were collected.

Follow up
Overall follow up was over a median of 33 months (IQR 20-48.5).
With respect to urinary continence, follow up was at 2, 4 and 8 weeks after removal of the urethral catheter (1 week after surgery), and thereafter, once a month during the first year after surgery.
Data were prospectively collected and stored in a customized database and retrospectively analyzed.
The patients were given the ICIQ-SF questionnaire. The assessment and follow up on regaining urinary continence was carried out by both the Physiotherapy and Urological Oncology Units of Moises Broggi Hospital, Barcelona, Spain.
Patients were considered continent when either no pads, or just pads for the psychological comfort of the dry patient, were used.
Patients were considered potent once they achieved regular intercourse, with or without the use of PDE5. PDE5 was not used systematically for patients with bilateral bundle preservation.
BCR was considered present when PSA is equal or >0.2 ng/mL. We carried out both adjuvant and salvage radiotherapy on selected high-risk patients.

Surgical complications
Complications occurring during surgery or within 60 days after surgery were classified according to the Clavien-Dindo classification (

Statistical analysis
For continuous variables, we show descriptive statistics (average, standard deviation, median, IQR), whereas for categorical variables, we show frequencies and percentages.
We applied Pearson's v 2 -test to detect the differences in levels of urinary continence between different patient groups, and applied McNemar's test to assess the changes in levels of continence over time.
We applied the Kaplan-Meier model to analyze BCR-free survival, and the log-rank test to compare our curves.

Results
We present a series of 120 patients who underwent LRP using the surgical technique described above. Their clinical, pathological and surgical variables are shown below.
The pathological and surgical stages of the patients are shown in Table 3: pTNM shown ≥pT2c in 105 patients (87.4%).

Oncological results
Overall, the PSM rate was 32 (26.6%) patients; PSM was associated with RMD in 10 (8.3%) patients. In no case in the present series did we impose any surgical limitation to patients with prostate tumors located on the ventral part of the prostate gland (TZ or AFMS).
BCR-free survival for all patients receiving radiotherapy showed a test log-rank P = 0.000, signifying a lower BCR-free survival time: the mean estimation of 39.071 months; 95% CI 32.8-45.2 for the Kaplan-Meier estimator (Fig. 8).
There was no stenosis of the urethral anastomosis during the follow-up period.
The rates of urinary continence with respect to the variables evaluated are shown in Table 4.    The rate of recovery of erectile function with bilateral bundle preservation (n = 25) during the first year of follow up was 18 (72%).

Discussion
Thus, with regard to urinary continence, three factors are seen to be of importance ( Fig. 9): 3,[12][13][14][15][16][17]19 1. The extrinsic or active factor of urinary continence: the puborectalis and MsPP, the most ventromedial fascicles of the levator ani muscle, are responsible for the abrupt interruption of micturition (Figs 1,10). They consist of type II striated muscle fibers and are innervated by the pudendal nerve.
2. The intrinsic or passive factor: the IUS and EUS, which require a certain "muscular tone" to prevent "constant dripping." The last one (EUS) comprises striated muscle fiber (mostly type I) innervated by the pudendal nerve, through its extrapelvic and intrapelvic fibers. The IUS is innervated by the corpus spongiosum nerves (neurovegetative innervation ; Figs 1,10).
3. The neurovegetative factor: Here, we refer to the afferent neurovegetative innervation of the membranous urethral mucosa, which, through a double mechanism (short medullar circuit and long conscious cortical circuit) causes the muscle fibers of the IUS, EUS and levator muscles of the anus to suddenly contract when the first drops of urine reach the urethral portion, thus preventing urinary incontinence. 3,19 Innervation runs through the breadth of the periprostatic fascia, most particularly along its lateral facets, which make up the so-called neurovascular bundles, responsible not only for erection, but also for playing a significant role in the preservation of urinary continence (Figs 1,10). [12][13][14] During radical prostatectomy, to a greater or lesser extent, we cause damage to these three factors and this in turn gives rise to PUI, representing a significant loss in the quality of life for these patients. [12][13][14][15] We propose a straightforward surgical technique deployed in two stages (DPPM and RMD), 14 via a classical retropubic approach through the pelvic subperitoneal space and applicable to any of the three standard techniques (open, laparoscopic or robot-assisted surgery). The goal of the technique is  to preserve the aforementioned neuroanatomical structures, all of which are involved in maintaining male urinary continence. The functional results achieved are excellent, comparable to the best series published, but without the need to preserve the neurovascular bundles for reasons of oncological safety and/or the patients existing erectile dysfunction. 2,[9][10]20 That is, it is a technique that can be applied perfectly well in most organ-confined prostate cancer cases, without causing raised PSM levels.

Median
Confidence interval of 95%  The present series showed a population of patients with a PSM rate within the standard published for other series, in line with the pathological characteristics of the series assessed (Table 3). [21][22][23][24] With regard to the RMD, the PSM rate was very low (8.3%), making RMD both a very safe technique oncologically speaking, as well as offering excellent functional results (Table 4).
Notwithstanding, RMD would not be considered an oncologically safe procedure for tumors established in the TZ and/or the AFMS, particularly when situated on the anterior slope of the PA, due to the high risk of finding PSM (Fig. 11).
In none of the patients in the present series did we find that postoperative radiotherapy resulted in any loss of urinary continence.
With respect to the preservation of the neurovascular bundles, just 25 patients were subject to bilateral preservation with a result of full urinary continence at between 3 and 4 weeks of the postoperative period for 100% of those patients. However, the remaining 95 patients also showed very high rates of urinary continence (Table 4). These results are in line with the current concept of the neuroanatomy of the prostate, according to which the majority of the nerves of the corpus spongiosum, responsible for continence, run along a dorsolateral plane of the bundle on a level with the prostate-rectum confluence, meaning that they would be spared even without a deliberate attempt to do so (Figs 1,10). 3 Fig. 10 Anatomical drawing of the caudal view of the PA and membranous urethra along with its sphincter systems. The extension of the EUS apron along the anterolateral faces of the gland and the arrangement of the most ventromedial muscle fibers of the MsPP are visible. The cavernous and corpus spongiosum nerves run medially to the puboperineal muscle plane, within the multiple sheaths of the conjunctive tissue, vessels, elastic fibers and smooth muscle fibers that make up the periurethral fascial plane. In this figure, the pubic symphysis and puboprostatic ligaments have been eliminated in order to show the above described anatomical elements.
Preoperative T2 WI MRI midline sagittal and coronal sections of the prostate showed a very short perineal urethra (membranous urethral length: mean of 7.3 mm; minimum 4.6 and maximum 9.5) in 11 patients (Table 3). 26,27 Surprisingly, at 8 weeks postoperative, all these patients were fully continent. There are many articles that make reference to a direct, proportional relationship between urethral length and early continence. 26,27 However, in our experience, albeit with just a few cases, such results were not reproduced, most probably due to the fact that our perineal approach to the apex of the gland, with extensive DPPM dissection, offers excellent visibility for the dissection of the urethral sphincter during RMD, enabling total preservation, including in those patients with a short perineal urethra.
In general, we have likewise found a positive correlation between urinary rate continence with age and the bilateral preservation of the neurovascular bundles. During weeks 2 and 4, urinary continence rates are significantly higher for younger patients (P = 0.044 and 0.029, respectively) and for bilateral NVBP (P = 0.034 and 0.012, respectively). 3 We found no correlation with BMI, 28 erectile function presurgery, prostate weight, 29 bladder neck preservation or the length of the urethra (Table 4).
DPPM and RMD were found to be oncologically safe procedures, and associated with a very early recovery of urinary continence in most patients in the first 2 months after surgery.
We can conclude that DPPM and more specifically RMD can be applied in most organ-confined prostate cancer cases, as long as there is no invasion of the ventral side of the prostate. It also offers high rates of early urinary continence, without the need to carry out a bilateral preservation of the neurovascular bundles.
The limitations to the present study are that there were too few cases of short perineal urethra to permit us to draw any conclusions, future studies of this topic will be necessary to clarify its role in urinary continence.