To develop a technique of nerve-sparing robot-assisted radical cystoprostatectomy (RRCP) for patients with bladder cancer.
To develop a technique of nerve-sparing robot-assisted radical cystoprostatectomy (RRCP) for patients with bladder cancer.
Robotic assistance should enhance the ability to preserve the neurovascular bundles during laparoscopic radical cystectomy. Thus we undertook RRCP and urinary diversion using a three-step technique. First, using a six-port approach and the da Vinci Surgical System (Intuitive Surgical, Sunnyvale, CA, USA), one surgeon carried out a complete pelvic lymphadenectomy and cystoprostatectomy using a technique developed specifically for robotic surgery. The neurovascular bundles were easily identified and dissected away, the specimen entrapped in a bag and removed through a 5–6 cm suprapubic incision. Second, a different surgical team exteriorized the bowel through this incision and created a neobladder extracorporeally. Third, the neobladder was internalized, the incision closed and the primary surgeon completed the urethro-neovesical anastomosis with robotic assistance.
RRCP was carried out in 14 men and three women by the primary surgeon (M.M.). The form of urinary reconstruction was ileal conduit in three, a W-pouch with a serosal-lined tunnel in 10, a double-chimney or a T-pouch with a serosal-lined tunnel in two each. The mean operative duration for robotic radical cystectomy, ileal conduit and orthotopic neobladder were 140, 120 and 168 min, respectively. The mean blood loss was < 150 mL. The number of lymph nodes removed was 4–27, with one patient having N1 disease. The margins of resection were free of tumour in all patients.
We developed a technique for nerve-sparing RRCP using the da Vinci system which allows precise and rapid removal of the bladder with minimal blood loss. The bowel segment can be exteriorized and the most complex form of orthotopic bladder can be created through the incision used to deliver the cystectomy specimen. Performing this part of the operation extracorporeally reduced the operative duration.
Radical cystoprostatectomy is the standard surgical procedure for muscle-invasive bladder cancer, but it is a formidable operation, associated with significant complications even in expert hands . There have been major advances in laparoscopy for urological oncology and a few surgeons have performed laparoscopic radical cystectomy (LRC) alone with extracorporeal diversion [2–6] or with a completely intracorporeal ileal conduit [7,8] or different form of continent urinary diversion [9–12]. However, these procedures are technically challenging, perhaps even more so than conventional open cystectomy. Robotic assistance has helped to refine one technically challenging oncological procedure, laparoscopic radical prostatectomy [13,14]. This prompted us to examine if robotic assistance would facilitate LRC and herein describe the technique that we currently use for robotic radical cystoprostatectomy (RRCP).
Between April 2002 and February 2003, 14 men and three women with TCC of the bladder underwent RRCP using the daVinciTM surgical system (Intuitive Surgical, Sunnyvale, CA, USA). The technique was developed in two patients in Detroit and modified in 15 in Mansoura, Egypt. All RRCPs were performed by one surgeon (M.M.), experienced in robotic techniques. In most cases, the urinary system was reconstructed by a second surgical team with extensive experience in open cystectomy. In the first five patients a surgical approach that mimicked the steps of ‘open’ radical cystectomy was used. In patients 6 to 17, the initial approach was through an extended transverse incision in the ‘cul de sac’, dissecting out the ureter, the adnexal structures and developing the rectovesical plane. The present report focuses on the technique of nerve-sparing RRCP in men and does not provide an in-depth description of the various techniques of urinary reconstruction used.
Bowel preparation with antibiotic coverage was initiated 1 day before the procedure. On the morning of surgery a broad-spectrum antibiotic and 5000 units of subcutaneous heparin were administered. Bilateral leg compression stockings were applied. After inducing anaesthesia the patient was placed in an extended lithotomy position to extend the space between the pubis and the umbilicus, with a 45° Trendelenberg tilt (to allow the bowels to fall back and open the pelvic cavity). Sufficient padding was applied around the shoulder and pressure points, and the arms were tucked in. A nasogastric tube and a urethral catheter were inserted.
A six-port transperitoneal approach was used; the pneumoperitoneum was created using a Veress needle (Ethicon Endo-Surgery, Inc., Albuquerque, NM) at the proposed site of the primary port in the sub or supra-umbilical region, depending on the body stature of patient. A 12-mm primary port was inserted and the entire peritoneal cavity inspected with a 30° laparoscope. The two 8-mm ports for the robotic instruments were placed 2.5 cm lateral to the rectus muscle and 2 cm below the umbilicus on either side. A second 12 mm port was placed in the right iliac fossa 5 cm above the anterior superior iliac spine in the anterior axillary line. The fifth port (5 mm) was placed between the robotic and 12 mm port on right side. The sixth 5-mm port was placed on the left side between primary port and left robotic port ≈ 5 cm below the umbilicus. All secondary ports were inserted under laparoscopic vision using a 30° laparoscope.
Most of the dissection is carried out with two instruments, i.e. the da Vinci long-tip forceps and the cautery hook. Alternatively, the bipolar coagulating forceps and the articulate scissors can be used, and we used these two instruments particularly for the nerve-sparing part of surgery. Two needle holders are used for suturing. The laparoscopic team uses grasping forceps and the suction for retraction and exposure.
In the first five patients a bilateral pelvic lymphadenectomy was undertaken initially, ligating the superior and inferior vesical pedicles as the operation proceeded. However, developing and transecting the rectovesical pedicles was difficult, particularly in obese patients, in patients with a narrow pelvis, in those with bulky tumours, or in those with pelvic inflammation secondary to bilharziasis. As essentially all the patients had one or more of these characteristics, we modified the technique, performing the posterior dissection initially. We reasoned that the angled lenses, combined with the wristed instrumentation, would allow the development of the rectovesical plane and initial preservation of the neurovascular bundles even before the anterior and latter bladder dissection.
The posterior dissection is carried out using the 30° lens, looking downward. An inverted U-shaped incision is made in the peritoneum of the cul de sac. In many patients, the course of the lower ureters can be seen as a peritoneal fold that extends from the iliac bifurcation to the posterior bladder wall. When this is apparent the vertical limbs of the U follow this course, extending to a point ≈ 2.5 cm proximal to the bifurcation of the common iliac artery. The laparoscopic assistants provide equal countertraction on the transected peritoneal folds and the surgeon dissects all fatty and fibrovascular tissue off the posterior peritoneal fold. The posterior layer of Denonvillier's fascia is incised in the midline and the plane between the rectum and the bladder developed as far inferiorly as is easily possible. The planes are extended laterally, such that a broad dissection front is maintained. This leads to the ureter, which lies on the under-surface of the posterior peritoneum. The ureters are dissected to the bifurcation of the iliac vessels proximally, and the vesico-ureteric junction (VUJ) distally. In males, the VUJ can be identified immediately inferior to the crossing vas deferens on the posterior bladder surface. It is important not to dissect the vas deferens off the posterior surface of the bladder, to maintain this anatomical landmark. The inferior vesical pedicle is usually encountered during this phase of the dissection and must be secured and divided (Fig. 1). The ureter is then clipped, transected and the margins sent for frozen-section analysis. The seminal vesicles are identified immediately medial to the lower end of the ureters and dissected down to their base. To preserve the nerves, which are located close to the tips of the seminal vesicles, this dissection is immediately next to the walls of the seminal vesicle, between the vesicle and the posterior layer of Denonvillier's fascia (which can be seen with exquisite clarity if the dissection has been in the proper planes). The rectoprostatic plane is now developed as feasible by dividing Denonvillier's fascia (Fig. 2).
For the next step, the 30° lens is used, looking upwards. The peritoneum is incised in the line of the external iliac artery from the apex of the U, proximally, to the inguinal ligament distally. This incision is lateral to the medial umbilical ligament and transects the vas deferens. The incision then curves medially under the rectus abdominis, transecting the medial umbilical ligaments and the urachus. A standard pelvic lymphadenectomy is done using the 30° downward lens, which is essential for the proximal dissection, over the common iliac artery. All nodal tissue is cleared from the genitofemoral nerve laterally to the bladder wall medially, and from the distal common iliac artery superiorly to the lateral circumflex iliac vein and the node of Cloquet inferiorly (Fig. 3A). The obturator fossa is cleared of nodal tissue, preserving the obturator nerve, but sacrificing the obturator vessels if necessary. The nodal tissue is cleared around the iliac vessels (Fig. 3B). The nodal tissue seems to form two natural packages, one attached to the bladder wall and one lateral to this. Lymphadenectomy is the most difficult part of the operation, because the tissue contains many small blood vessels that have to be meticulously coagulated. Otherwise, they retract into the tissues and give rise to haemodynamically insignificant but visually annoying oozing. This impairs visibility and may obscure the detection of precise tissue planes.
For control of the bladder pedicles and dorsal vein complex, the superior vesical pedicle is clipped and transected at its origin (Fig. 4). The anterior trunk of the internal iliac artery continues as the inferior vesical artery which gives off vesical branches (secured during the earlier posterior dissection) and terminates as the prostatic artery. This vessel is dissected until it bifurcates into the urethral artery and capsular artery. The urethral artery is clipped and transected, but the capsular artery, which forms the vascular part of the neurovascular bundle of Walsh, is preserved. Identifying the capsular artery enables subsequent preservation of the neurovascular bundles. The endopelvic fascia is now opened lateral to the prostate and the prostato-urethral junction is identified. The dorsal vein complex is secured using a suture of 0 polyglactin on a CT1 needle (Fig. 5), as previously described [13–14]. The apex of the urethra is freed from the rectourethralis using blunt dissection.
The seminal vesicles are used as an operative landmark to avoid injury to the neurovascular bundles and we follow the principles described earlier [13,14]. The dissection is in the plane between the posterior surface of seminal vesicle and the posterior layer of Denonvillier's fascia. Monopolar coagulation is avoided, and the da Vinci articulated scissors and bipolar forceps used for this step. The dissection should be meticulous and stay close to the prostatic surface, reflecting the lateral pelvic fascia off the prostate. Such precision is possible because the vesical and prostatic pedicles have been controlled at this point. The neurovascular bundle is then reflected off laterally, leaving a layer of Denonvillier's fascia on the surface of the rectum. (Fig. 6)
The urethra is divided at the apex of prostate with the help of articulated robotic scissors (Fig. 7). The posterior striated sphincter should be divided carefully. An attempt is made to gain the maximum length of urethra, which would help subsequently in the anastomosis with the neobladder.
The specimen is entrapped in a laparoscopic Endocatch II bag (US Surgical, Norwalk, USA) and retrieved through a 5–6 cm incision placed midway between the umbilicus and pubic symphisis (Fig. 8A,B). Through the same incision a segment of ileum is extracted, isolated, detubularized and reconfigured extracorporeally (Fig. 9). It was possible to conduct the following through the bladder-retrieval incision: the ileal conduit (in three), a W-pouch with a serosal-lined tunnel (in 10) or a double chimney or T-pouch with a serosal-lined tunnel (in two each). The technique of orthotopic substitution is described in previous reports [15,16].
The pouch is placed in the pelvis and a Foley catheter passed urethrally into the pouch, through the neobladder neck. The pouch is pulled down to the urethra. The abdominal incision is closed and the robot re-docked for anastomosis of the neobladder with the urethra (Fig. 10A,B). The urethro-neovesicostomy is performed robotically with a continuous double-armed 3–0 polydioxanone suture, as previously described for radical prostatectomy [13,14].
The mean operative duration for the RRCP was 140 min, and the mean blood loss < 150 mL. The operative duration for the ileal conduit and orthotopic neobladder were 120 min and 168 min, respectively. Surgical margins were negative in all cases; one patient had N1 disease. Thirteen patients had associated bilharziasis, with significant periureteric, perivesicular and perivesical scarring.
The operation was not completed in one patient because of a malfunction of the lens. One patient was re-explored for bleeding. While no active bleeding point was seen, there was a port-site haematoma, and oozing at the site of the urethro-neovesical anastomosis (constructed using open surgery in this instance).
LRC has been reported by several authors [2–12], but robot- assisted cystectomy has not. Parra et al. performed the first laparoscopic simple cystectomy for infection in a retained bladder but the first case of radical cystectomy with reconstruction of the ileal conduit extracorporeally was reported by Sanchez de Badajoz et al.. Since then there have been various reports of LRC but the urinary diversion was performed extracorporeally from the site of removal of the specimen or by a mini-laparotomy incision [4–6]. Puppo et al. also reported five cases of LRC with transvaginal extraction of the specimen. Gill et al. first reported two cases of LRC and intracorporeal ileal conduit formation, and subsequently performed five more cases . Turk et al. were the first to report five cases of LRC and intracorporeal continent (recto-sigmoid pouch) urinary diversion and transrectal specimen retrieval. Gill et al. also performed LRC and an orthotopic neobladder (Studer) in two patients. With the exception of the Turk et al. and Puppo et al., all surgeons used an abdominal incision to remove the specimen, but Gill et al. used either an abdominal or a vaginal incision [3–12].
The concerns about intraoperative blood loss and subsequent need for transfusion have always been associated with radical cystectomy. In a recent report, despite various technical modifications to reduce the blood loss during open radical cystectomy, the estimated median blood loss was 600 mL, with a third of patients required transfusion, and it was also noted that two previous reports had a mean and median blood loss of 1800 and 1700 mL, respectively . One of the most notable advantages of LRC seems to be the strikingly small blood loss, with almost negligible transfusion rates, although it is not consistent in all reports [2–12]. However, the procedure is difficult to learn and restricted to those well versed in complex laparoscopic surgery.
We sought to develop a technique for LRC that would use the advantages of robotic technology and maximize the benefits of the incision required to remove the bladder. This was the foundation behind the three-step concept, developed from our collective experience with > 450 robotic radical prostatectomies and 3000 open cystectomies. To our knowledge, there are no previous reports of nerve-sparing or robotic radical cystectomy. Our approach allows the neurovascular bundles to be identified and preserved with great precision. Our technique combines the unique advantages of robotic and open surgery to create a procedure that is safe, quick and precise.
In conclusion, this feasibility study of a three- step robot-assisted, nerve-sparing radical cystoprostatectomy shows that the procedure combines the oncological concepts of open surgery with the technical nuances of robotic surgery. The procedure was precise, gentle and safe. The collaboration of surgeons experienced in pelvic robotic surgery  and those with extensive experience in open radical cystectomy  enhanced the development of these concepts. However, the development is not finished but continues, with the possible final goal of robotic cystectomy with complete intracorporeal urinary reconstruction. While this is technically feasible it seems to waste the opportunity for efficiency that the specimen retrieval incision offers. It can be done, but does it need to be? Our opinion might change once more efficient automatic suturing devices appear.
laparoscopic radical cystectomy
robotic radical cystoprostatectomy.