Open retropubic prostatectomy (RP) is one of the standard surgical approaches for treating clinically localized prostate cancer.1 Recently, aside from the established open retropubic or perineal approach, conventional laparoscopic prostatectomy and robot-assisted radical prostatectomy have experienced increasing popularity.2,3 Today, most clinics have a preferred surgical approach, based on their surgical school, and favourable outcomes are reported with all techniques.2–8 In a recent review article, no superiority, in terms of functional and oncological outcome, was reported for any particular technique except for lower blood loss and transfusion rates associated with robotic prostatectomy.9 However, all techniques are under constant refinement and it is important that updated results of existing techniques are published to allow an ongoing and timely comparison of the available approaches. Such studies aid the fair judgement of developments in each technique and help to better assess functional outcome, efficacy and morbidity. We report our current technique of open nerve-sparing RP, including cancer control rates, functional outcome and perioperative morbidity.
Equipment recommendations for open nerve-sparing retropubic prostatectomy
We consider a self-retraining retractor system (Bookwalter autoretractor,™ Codman, Le Locle, Switzerland), which has 4- to 5-fold magnification surgical loupes and a xenon headlamp for optimal exposure and illumination of the operative field. No specific positioning of the patient is necessary.
We recommend spinal anaesthesia and additional total intravenous anaesthesia. Intravenous fluid replacement should be limited to 500 ml until the prostate is removed. Based on an individual fast-track concept, discharge from hospital in a good physical condition is possible 3–4 days after surgery and a minimum hospital stay of 4 days postoperatively is required in Germany for legal reasons.
Open radical retropubic prostatectomy – step-by-step
Incision of the endopelvic fascia and preparation of the dorsal vein complex
After an 8–12 cm subumbilical vertical medial skin incision, Retzius' space is established. The endopelvic fascia is incised and fibres of the levator ani muscle are carefully pushed away. However, fibres from the levator urethrae are preserved to maintain the anterior position of the urethra. The attachments of the detrusor apron, also called the puboprostatic ligaments, are isolated and then removed. Throughout the procedure, no coagulation enhancers should be used in the vicinity of the neurovascular bundle or the surface of the prostate to avoid damage to the nerves.
A superficial stay suture (3-0 monofile suture using a UR-6 needle) is created distally to the prostate and will later be used for the selective ligation of the dorsal vein complex. The lateral parts of the fascia below the 10 and 2 o'clock positions of the sphincter muscle, also called Mueller's ligaments, are not touched as they separate the sphincter from the neurovascular bundle (Figures 1 and 2). To prevent backflow bleeding, an additional 2-0 Vicryl suture is placed in the midline of the prostate.
The trans-section of the exposed dorsal vein complex, situated between the continuation of the endopelvic fascia superiorly and the fascia of circular striated sphincter muscle inferiorly, is initiated near the apex of the prostate. The dorsal vein complex is dissected without any ligation until the fascia of the external sphincter is visible. While the fascia of the external sphincter is incised, care is taken that the muscle fibres of the external sphincter that are directly underneath are kept intact. The blood loss during the trans-section is typically moderate, averaging 150–250 ml of blood loss per patient.
The selective suturing of the distal parts of the dorsal vein complex between the 10 and 2 o'clock positions involves two layers: the ventral part of the dorsal vein complex, which consists of the continuation of the endopelvic fascia; and the dorsal part, which is covered by the fascia of the external sphincter. This approach ensures that functional tissue of either the sphincter fibres or the urethra is not incorporated into the ligation and that traction to the adjacent tissue is avoided (Figure 3).
Intrafascial nerve-sparing procedure
The dissection of the neurovascular bundles starts on the anterosuperior part of the prostate with an incision to the parapelvic fascia, because the majority of nerves run on the inferolateral part of the prostate. The neurovascular bundles are mobilized and lateralized before the urethra is dissected (Figure 3). For careful dissection, the attached levator fascia and periprostatic fascia are carefully lifted and incised at the anterior aspect of the prostate above the 10 and 2 o'clock positions. The underlying space, including the nerves, fatty tissue and small vessels supplying the prostate, is identified. When the correct plane of dissection is entered, the shiny, smooth and reflecting surface of the prostatic capsule becomes visible.
The neurovascular bundles are gently pushed laterally and inferiorly using the blunt tip of the scissors. The dissected fascia and vessels are clipped with 5 mm rectangular titanium clips and 3 or 5 mm 45° angle titanium clips, or selective stitches [5–0 resorbable polydioxanone (PDS) sutures], to control arterial bleeding. The neurovascular bundle can be released when the periurethral area is 3–5 cm proximal to the base of the prostate and the bundle can be released into the perivesical fat to avoid tension on the bundle and to expose the prostatic pedicles.
Full functional length preparation of the urinary sphincter
To preserve a maximum amount of functional tissue, the preparation of the sphincter and urethra begins at the apex. Fibres of the circular striated sphincter muscle, covering the apex outside of the prostate, are carefully pushed distally with blunt scissors until the longitudinal smooth muscle fibres become visible. Subsequently, the longitudinal smooth muscle fibres, running inside the prostate, are incised. Transection of the longitudinal fibres is performed approximately 3 mm within the prostate, superior to the apex of the prostate. The remaining muscle fibres are gently pushed distally for preservation. Subsequently, the urethra is incised anteriorly and the preparation is performed for two-thirds of its circumference. By beginning the subtle dissection with the fibres covering the apex of the prostate, the entire length of the functional urethra is preserved (Figure 4).10
Anastomotic sutures and prostate removal
Six sutures are required for a complete anastomosis. At the 3 o'clock, 1 o'clock, 11 o'clock and 9 o'clock positions, a UR-6 needle with a 3-0 PDS suture is passed through the ligated dorsal vein complex, which is used as an anchor. The needle is then passed through the mucosa of the urethra and only a small part of the sphincter muscle is incorporated. At the 5 o'clock and the 7 o'clock positions, the last two 3-0 PDS sutures are generated with a UR-6 needle. These also fix the dorsal part of the sphincter to Denonvilliers' fascia and the raphe of the sphincter muscle for traction of the whole membranous urethra (Figure 5), which is similar to the technique described by Rocco et al.11 As mentioned above, perivesical fat is mobilized during the nerve-sparing procedure to release the bundles and avoid traction during the surgery. In addition, the release of the fatty tissue up to 3–5 cm proximal to the base of the prostate allows an exact visualization of the prostatic pedicles, which then are selectively ligated or clipped.
At the base of the prostate, the Denonvilliers' fascia is incised for preparation of the seminal vesicles. By incising the Denonvilliers' fascia, a ventral part of this fascia is left on the specimen to prevent positive margins. The part of the Denonvilliers' fascia covering the seminal vesicles is incised and is left in situ in order to protect the underlying neurovascular bundle. The tips of the seminal vesicles are identified and the adjacent vessels are clipped and dissected with 5 mm titanium clips. No thermal energy is used to protect the integrity of the nerves running close to the tips of the seminal vesicles. If required, the bladder outlet is narrowed with a tennis-racket suture (3-0 Vicryl) and eversion of the mucosa. The operating field is checked and if bleeding close to the neurovascular bundle occurs, 5-0 PDS sutures, or clips, are used to control the bleeding.
End of the procedure
The needles of the six anastomotic sutures are placed through the everted bladder mucosa and tied in a single-knot technique to avoid compromising the blood supply, which results in a stricture rate of < 1%. The wound is closed using a self-resorbable intracutaneous suture, which has good aesthetic results. Whether a drain is carried out is left to the discretion of the surgeon.
Indication for nerve-sparing and lymph node dissection
The indication that nerve-sparing surgery (bilateral vs. unilateral vs. no nerve sparing) and lymph node dissection is required is based on nomograms used to predict the probability of side-specific extraprostatic extension and lymph node invasion.12,13 To further increase the oncological safety of a nerve-sparing procedure, frozen sections (NeuroSAFE)14 of the lateral surface of the prostate are taken from every patient.15 In the Martini-Clinic, preoperative nomography and intraoperative sections are used to avoid positive surgical margins in inadequately indicated nerve-sparing procedures (i.e. in patients with capsular penetration) and to identify as many candidates for nerve-sparing RP as possible. As the advantages of nerve-sparing RP are obvious compared to non-nerve-sparing RP, it seems necessary to identify as many suitable cancers for nerve-sparing RP as possible. Using nomography to estimate the side-specific likelihood of organ confinement in combination with NeuroSAFE intraoperative frozen section analysis in the Martini-Clinic, the rate of patients undergoing nerve-sparing RP for pathologically organ-confined disease is as high as 98.3%. These results corroborate that we strive to carry out nerve-sparing RP as much as possible using frozen section analysis to minimize side-effects on functional outcome.
NeuroSAFE frozen sections are cut from the apex to the base of the complete dorsolateral part of the prostate and the sides of the slice are inked in different colours (Figures 6 and 7).15 If tumour cells extend to the outer surface on microscopic evaluation, we remove the corresponding neurovascular bundle and its adjacent tissue. Although NeuroSAFE frozen section analysis does not completely eliminate the risk of positive surgical margins, it does effectively minimize the risk.
Perioperative parameters and morbidity
Preoperative parameters and morbidity were based on the data from 1150 patients, whereas the functional results reported below are based on 637 preoperative potent men. The median operating time was 175 minutes (range 95–215 minutes), the median blood loss was 590 ml (range 130–1800 ml) and the transfusion rate was 4.5%. There was no rectal injury, ureteral injury or perioperative death. All patients were routinely investigated by pelvic ultrasound before discharge to detect a pelvic haematoma (present in 5.3% of patients) or lymphocele (present in 7.5% of patients). Surgical revision for pelvic haematoma was necessary in 0.4% of patients and lymphoceles were treated only when compression of the external iliac vein was seen by Doppler sonography; these cases (0.8% of patients) were managed by puncture therapy. No patient had to undergo laparoscopic fenestration for lymphoceles; however, deep vein thrombosis was observed in 1.3% of patients 14 days postoperatively.
Urinary continence 1 year after surgery
The reported number of required pads for urine leakage after nerve-sparing RP is commonly used to assess postoperative urinary continence. In the Martini-Clinic, validated questionnaires (International Continence Society and European Organization for Research and Treatment of Cancer Quality of Life C30 questionnaires) are routinely conducted 1 year after surgery. Continence results were stratified by age and extent of nerve sparing (the unilateral vs. the bilateral approach).
In men aged 70 years, urinary continence 1 year after nerve-sparing RP was reported to be between 93.2% and 97.4% and did not seem to be influenced by the extent of nerve sparing (Table 1). Among men > 70 years of age, urinary continence was observed in 94.5% after bilateral nerve-sparing RP; however, after unilateral nerve-sparing RP, only 84.1% achieved continence, suggesting a meaningful effect of the extent of nerve sparing on urinary continence in this patient group.16 Some 2.6–6.8% of patients < 70 years of age reported needing two pads per day, and this did not seem to be influenced by the extent of nerve sparing. However, among men aged > 70 years, 3.7% needed two pads after bilateral nerve-sparing RP and 10.7% needed two pads after unilateral nerve-sparing RP. More than two pads were used by 0.7% of men ≤ 70 years who underwent bilateral nerve-sparing RP and up to 1.8% of men who underwent unilateral nerve-sparing RP. Among men > 70 years, more than two pads were used by 1.8% and 5.2% of patients after bilateral and unilateral nerve-sparing RP, respectively. There was no complete urinary incontinence.
Erectile function 1 year after surgery
Evaluation of postoperative potency was restricted to men with documented good preoperative erectile function who underwent a unilateral or bilateral nerve-sparing RP and in whom information on postoperative erectile function 1 year after surgery was available. The five-item version of the International Index of Erectile Function (IIEF-5) was given before surgery and 12 months after RP for all patients.17 There are several approaches to describe postoperative erectile function and the IIEF-5 score is one of the most commonly used questionnaires. Many centres do not report postoperative IIEF scores but report patients' ability to perform intercourse after RP instead. Such results strongly correlate; nevertheless, the number of potent patients substantially varies by the type of evaluation and the definition of being potent. Question 2 of the IIEF-5 questionnaire asks for erections sufficient for penetration after sexual stimulation. By this definition, potency rates varied from 84% to 92% in men who underwent a bilateral nerve-sparing RP and from 58.3% to 70% following unilateral nerve-sparing RP (Table 2). The use of proerectile medication, such as phosphodiesterase type 5 (PDE-5) inhibitors, was left to the patient's discretion and 20% of men used such medications. The potency rates of only 25–59% of men returned to a normal score of >19 (based on the IIEF questionnaire).
In this article, we describe our current technique of nerve-sparing RP and, therefore, cannot present meaningful oncological long-term follow-up data for this specific cohort. All prostatectomy specimens were inked and underwent a 3 mm step-section procedure for pathological work-up. Using margin status as a surrogate parameter for cancer control, the positive margin rate was 5.2% in pT2 cancers and 27.1% in pT3 cancers. However, because our surgical approach is based on previous reports from the Martini-Clinic, we report on long-term data addressing prostate-specific antigen recurrence-free survival and, more importantly, cancer-specific survival rates.4,18 At 10 years after RP, the biochemical recurrence-free survival rates ranged from 5.9% to 87%, depending on the pathological stage, and cancer-specific survival rates ranged from 72.2% to 98.3%.
Retropubic prostatectomy has been performed for several decades and the groundbreaking work on anatomy and surgical technique was published by Walsh and Donker more than 30 years ago.1 Several studies have published results of this technique and the ongoing modifications.4–8 The refinements observed in this report are certainly no reinvention of the technique, but nuances of the prescribed approach have changed. One modification in the current technique is a more precise dissection of the posterior urethra. The continence rate 1 year after surgery, defined as no pad used or one safety pad used only, increased from 92% in 2006 to 95.5% in our latest questionnaire series in 2011 on more recent patients. The above-mentioned modification may be responsible for that increase and the intrafascial approach for the nerve-sparing procedure and the increased number of NeuroSAFE intraoperative frozen sections, including the complete capsule corresponding to the neurovascular bundle, is another alteration from previous studies. The current positive surgical margin rate in pT2 cancers has fallen in recent years from 9.4% to 5.2%.18 These data have to be viewed in the context that 98.3% of patients with organ-confined disease at final pathology are undergoing a nerve-sparing procedure.
The intrafascial approach has also slightly increased potency rates reported by patients in the present study, with 58.3–92% of men reporting having postoperative erections sufficient for intercourse. Potency assessment based on whether an IIEF score of >19 was reached resulted in potency rates falling to 25–59%, a phenomenon that has already been described by Menon et al.3 In that study, 70% of preoperative potent men had erections sufficient for intercourse 1 year after surgery; however, only half of these patients returned to a normal Sexual Health Inventors for Men score, indicating the necessity of using identical and validated assessment tools when differences in potency rates of published studies are considered.
Cancer control rates cannot be provided on the present cohort as these are recent patients. Nevertheless, the presented data based on the patients who underwent surgery earlier showed favourable recurrence-free and cancer-specific survival rates. Although we cannot prove that such data can be extrapolated to the present patient cohort, we believe that long-term cancer control rates will further improve in the present study. This is based on the fact that positive margin rates are lower in the present study than in previously studied patients.18 The importance of negative margins on outcome has been discussed extensively and we believe that intraoperative frozen sections are one of the keys to achieving favourable functional outcomes without compromising the procedure.19
Interestingly, the recurrence-free survival rate in a previous report from the Martini-Clinic in pT2 cancers at 10 years was 80% and increased to 87% in our recent analysis.18 This change may be due to favourable cancers within defined cohorts, which is often associated with the phenomenon that the year of surgery may have prognostic impact.20
One of the aims of this report is to demonstrate that the decrease in morbidity of RP is based not on the methods used but on the experience of the surgical team. Complication rates, and especially transfusion rates, in the current study are in the range of a high-volume endoscopic series, which highlights our hypothesis.9 Prospective trials are needed to show whether or not any method is advantageous over another.
Open radical RP is still the most frequently performed surgical approach for the treatment of localized prostate cancer. Surgical refinements include a strict intrafascial approach and subtle preservation of the posterior urethra and its insertion. In the hands of experienced surgeons, open radical RP provides excellent long-term cancer control rates, favourable functional outcome and low perioperative morbidity.