Totally intracorporeal robot-assisted radical cystectomy: optimizing total outcomes
We performed a systematic literature review to assess the current status of a totally intracorporeal robot-assisted radical cystectomy (RARC) approach. The current ‘gold standard’ for radical cystectomy remains open radical cystectomy. RARC has lagged behind robot-assisted prostatectomy in terms of adoption and perceived patient benefit, but there are indications that this is now changing. There have been several recently published large series of RARC, both with extracorporeal and with intracorporeal urinary diversions. The present review focuses on the totally intracorporeal approach. Radical cystectomy is complex surgery with several important outcome measures, including oncological and functional outcomes, complication rates, patient recovery and cost implications. We aim to answer the question of whether there are advantages to a totally intracorporeal robotic approach or whether we are simply making an already complex procedure more challenging with an associated increase in complication rates. We review the current status of both oncological and functional outcomes of totally intracorporeal RARC compared with standard RARC with extraperitoneal urinary diversion and with open radical cystectomy, and assess the associated short- and long-term complication rates. We also review aspects in training and research that have affected the uptake of RARC. Additionally we evaluate how current technology is contributing to the future development of this surgical technique.
Since 2003, robot-assisted radical cystectomy (RARC) has been gradually adopted as a surgical alternative to open cystectomy. As recently as 2010 the number of centres performing this surgery appeared limited, with only ∼500 cases being reported in the worldwide literature , but by 2012 this had increased to ∼1000 cases . Currently the vast majority of RARC in the USA are completed with an extracorporeal approach to the urinary diversion. In a recent multi-institutional report from the USA only 3% of patients had a totally intracorporeal approach ; however, there is accumulating evidence that both RARC and totally intracorporeal RARC are gaining acceptance, with increasing numbers of cases using both approaches being completed in the USA and Europe. We are also seeing the first publications of large single-centre series of RARC  and totally intracorporeal RARC .
The adoption of RARC as a viable alternative to open radical cystectomy has been impeded to date by the lack of long-term oncological and functional outcomes . The extent of a robot-assisted pelvic lymph node dissection has also been questioned, but data are now emerging from multicentre studies, as well as large single-centre series, showing similar lymph node yields and promising medium- and long-term oncological outcomes for RARC [5, 7]. The benefits of a totally intracorporeal RARC approach include the potential advantages of laparoscopy avoiding a mini-laparotomy with less intra-operative blood loss, quicker return of bowel function, shorter hospital stay with earlier return to normal activities, which would all theoretically allow more timely administration of adjuvant chemotherapy if required. There is also a shorter learning curve when compared with pure laparoscopic surgery [8-10].
Although RARC has additional direct peri-operative costs when compared with other treatment methods , RARC may provide overall cost savings when indirect costs, such as length of stay, time back to normal activities and complication rates, are considered . Radical cystectomy with urinary diversion by any approach remains one of the most complex and morbid operations performed in urology. It is associated with high complication rates even in the hands of experienced surgical teams . Complication rates have been shown to decrease with experience, but remain high overall, even in high-volume specialist centres [3, 13-15]. While open radical cystectomy remains the ‘gold standard’ treatment for carcinoma invading the bladder muscle and high-risk non-muscle-invasive bladder cancer, minimally invasive approaches are continually being refined and re-assessed. A recent meta-analysis concluded that complication rates were actually reduced with a robotic approach compared with open surgery . As technology develops and techniques evolve it is increasingly likely that more surgeons will perform a robotic approach to manage bladder cancer, reflecting the adoption seen in radical prostatectomy surgical management. By completing RARC with an intracorporeal urinary diversion, are we making an already challenging procedure more complex and potentially increasing complication rates for our patients? Alternatively, is it logical to commence RARC with a minimally invasive approach, only to convert to an open procedure to complete the operation? In the present review we assess the literature to see if the current evidence supports a totally intracorporeal RARC approach.
Materials and Methods
We performed a systematic literature search using the terms ‘RARC’, ‘robotic cystectomy’, ‘robot-assisted’, ‘oncological outcomes’, ‘functional outcomes’, ‘complication rates’, ‘learning curve’, ‘totally intracorporeal RARC’, ‘intracorporeal neobladder’ and ‘intracorporeal urinary diversion’, in the Medline, Embase and PubMed databases. The ‘related articles’ function was also used to broaden the search, and the computer search was supplemented with manual searches for the reference lists of all retrieved studies. Randomized controlled trials, systematic reviews, controlled cohort studies and observational studies were reviewed. The literature search revealed >200 publications, of which mainly studies published in the English language were reviewed. Excluding the vast majority of case reports and irrelevant studies, ∼120 studies were available for citation. Of these <20 publications focused on a totally intracorporeal RARC approach. Tables were created incorporating the results of totally intracorporeal RARC for future literature comparison.
The peri-operative outcomes for totally intracorporeal RARC with neobladder formation (Table 1) and with ileal conduit formation (Table 2) are summarized below. Series that do not differentiate outcomes between conduit and neobladder urinary diversion are recorded as ‘overall results’.
Table 1. Summary of robot-assisted intracorporeal neobladder series
|Collins et al. ||2013||70||420||500||21 (0–52)||1.5||9||Clavien I–II complications for <30 days and long-term (>30 days): 13 and 11%, respectively. Clavien III–V complications <30 days: 31%; long-term: 19%.|
|Goh et al. ,||2012||8||450||225||55 (44–74)||0||8||Clavien I–II complications for <30 and 31–90 days: 63 and 0%, respectively. Clavien III–V complications <30 days: 25%; 31–90 days: 13%.|
|Kang et al. ||2012||1||585||500||29.7 (overall series)||0||14||Postoperative paralytic ileus, no Clavien III–V complications.|
|Canda et al. ||2011||25||594 (overall series)||429.5 (overall series)||24.8 (8–46) (overall series)||3.7 (overall series)||10.5 (overall series)||Overall results, Clavien I–II complications for <30 days and 31–90 days: 33 and 15%, respectively. Clavien III–V complications <30 days: 15%; 31–90 days: 11%.|
|Pruthi et al. ||2010||3||330||221 (overall series)||NR||NR||5||NR|
|Sala et al. ||2006||1||720||100||NR||NR||5||0|
|Beechen et al. ||2003||1||510||200||NR||NR||10||0|
|Overall, mean|| ||107||473||458||25||1.9||9.4|| |
Table 2. Summary of robot-assisted intracorporeal ileal conduit series
|Collins et al. ||2013||43||292||200||21 (0–57)||11.6||9||Early complication rate (<30 days): 24% low grade and 39% high grade (3–5). Late complication rates (>30 days): 20% high grade.|
|Azzouni et al. ||2012||100||352||300||24 (NR)||4||9|| |
Early complication rate (<30 days): 50% low grade and 13% high grade.
31–90-day complication rates: 66% low grade and 15% high grade.
|Goh et al. ||2012||7||450||225||57.5 (22–95)||0||8|| |
Early complication rate (<30 days): 71% low grade and 0% high grade.
31–90-day complication rates: 14% low grade and 0% high grade.
|Kang et al. ||2012||3||510||400||29.7 (overall series)||0||NR||No complications.|
|Canda et al. ||2011||2||594 (overall series)||429.5 (overall series)||24.8 (8–46) (overall series)||3.7 (overall series)||10.5 (overall series)||Overall results, Clavien I–II complications for <30 days and 31–90 days: 33% and 15%, respectively. Clavien III–V complications <30 days: 15%; 31–90 days: 11%.|
|Pruthi et al. ||2010||9||318 (overall series)||221 (overall series)||NR||NR||4.5||NR|
|Balaji et al. ||2004||3||691||250||NR||NR||7.3||Early complication rate (<30 days): 33% low grade and 0% high grade. Late complications not reported.|
|Overall, mean|| ||176||333||255||23.5||5.5||8.2|| |
Although there is ongoing debate as to whether it is better to commence with the lymph node dissection or the radical cystectomy during RARC, both of these parts of the operation are completed before the urinary diversion. It is therefore unlikely that long-term data will show any difference in oncological outcomes between standard RARC with extracorporeal diversion and totally intracorporeal RARC. With RARC having been performed since 2003 we are now starting to see the results of medium-term oncological outcome data  as well as early randomized controlled trial outcomes, which have until now been lacking [17, 18].
The accumulating oncological outcome data for RARC are supported by encouraging short- and intermediate-term outcome reports [3, 14, 15, 19]. Until recently, most oncological data had been on surrogate markers of oncological outcome, namely lymph node yields and positive surgical margin (PSM) rates. A current meta-analysis concluded that RARC is associated with higher lymph node yields compared with open radical cystectomy. The two surgical techniques appeared to be equivalent in terms of PSM rates .
Several large series have recently reported their data including a multi-institutional analysis of RARC in 227 patients . In that series, 53% of patients had ≤pT2 disease, 15% had pT3/T4 disease and 20% had node-positive disease with the remainder having pT0. The mean (range) number of nodes removed was 18 (3–52). There was a PSM in five cases (2.2%), all with T3/T4 disease. Whilst concerns about the lack of tactile feedback in T3/T4 disease remain , some series have had excellent PSM rates. Pruthi et al.  analysed their outcomes in 100 consecutive RARC cases, 40% were ≤pT1, 27% were pT2, 13% were pT3/T4 and 20% were node-positive. The mean (range) number of lymph nodes removed was 19 (8–40). They had a PSM rate of 0%. At a mean follow-up of 21 months, 15 patients had disease recurrence and disease-specific survival was 94%.
Medium-term oncological data are now emerging, Collins et al.  reported on their first 113 consecutive totally intracorporeal RARCs performed since 2003. On surgical pathology, 48% of patients had ≤ pT1, 27% had pT2, 13% had pT3 and 12% had pT4 disease. The mean (range) number of lymph nodes removed was 21 (0–57) and 20% of patients had lymph node-positive disease. PSMs were found in six patients (5.3%). Of the six PSMs, five were in patients who had undergone ileal conduit diversion, and one patient had T3 and four had T4 disease. Only one patient undergoing an orthotopic neobladder had a PSM, attributable to a falsely reported ureteric margin on frozen section at the time of the operation. The median (range) follow-up was 25 (3–107) months. Using Kaplan–Meier analysis, cancer-specific survival was 81% at 3 years and 67% at 5 years.
Currently, there is a limited amount of data on functional outcomes . Whereas in robot-assisted radical prostatectomy (RARP) surgery there is a potential trade-off between oncological and functional outcomes , in RARC functional outcomes are dependent on various factors and surgical choices, e.g. continent vs non-continent diversion, with additional variables such as natural voiding vs required intermittent self-catheterization. Although continence rates after RARC are directly related to the surgical approach, they are influenced by multiple factors including patient age and mental status, an intact and innervated urethral sphincter, urethral length, low-pressure/large-capacity reservoir (>300 mL), absence of bacteriuria, and completeness of voiding. Continence after orthotopic bladder substitution continues to improve up to 12 months after surgery. It is therefore preferable to assess continence stratified by daytime vs night-time continence and by gender [22, 23] (Table 3). Similar conclusions were reached in the 2012 EAU International Consultation on Bladder Cancer  which reviewed the data published on urinary diversion between 1970 and 2012 and found that in patients with open radical cystectomy and orthotopic bladder substitution, day- and night-time continence is achieved in 85–90% and 60–80%, respectively.
Table 3. Extracted functional results from published data at 12 months after totally intracorporeal RARC with neobladder formation
|Tyritzis et al. ||70 (62 men)||88||72||70 (66 for men)||58|
|Goh et al. ||8||75||NR||NR||NR|
|Canda et al. ||17||64.7||17.6||92||NR|
|Sala et al. ||1||100||0||NR||NR|
If we consider totally intracorporeal continent urinary diversion, in most published series a Studer neobladder has been created [23, 25-27] and, although current cohorts are small, functional outcomes reported are encouraging. Tyritzis et al. , in a series of 70 patients, reported daytime continence of 88.2% in male patients who had undergone nerve-sparing surgery, whilst night-time continence reached 73.5% at 12 months. Similar rates were achieved for males who had undergone non-nerve-sparing surgery at 12 months (83.3 and 88.9%, respectively). Of female patients, 66.7% were found to be continent during the day and 66.7% at night at 12 months. All continence rates showed significant improvement at 12 months compared with the 6-month follow-up. A total of 81.2% of male patients were potent with or without phosphodiesterase type 5 inhibitor medication at 12 months. In that series, all eight female patients received a nerve-sparing procedure by preserving the autonomic nerves on the anterior vaginal wall. Of the evaluated male nerve-sparing, male non-nerving-sparing and female patient groups 84.4, 23.8 and 66.7% of patients, respectively, were sexually active postoperatively. Goh et al.  reported daytime continence in six out of eight patients at a mean (range) follow-up of 3.1 (3–21) months. Canda et al.  reported daytime continence in 11 out of 17 patients at a mean (range) follow-up of 6.4 (2–12) months, four had mild and two had severe daytime incontinence.
In sexual functionality in females, important outcome measures after the reconstruction of the vagina include both the ability to have sexual intercourse and the absence of dyspareunia.
Complication rates from published single-centre series are summarized in Tables 1 and 2. The International Robotic Cystectomy Consortium (IRCC) recently published their accumulated data complication rates . Although the majority of these cases were performed with an extraperitoneal approach to urinary diversion, it is a large database of 939 patients with at least 90-day follow-up and is therefore a good comparative measure. In all, 41% (n = 387) and 48% (n = 448) of patients experienced a complication within 30 and 90 days of surgery, respectively, and 29% had grade 1–2 and 19% had grade 3–5 complications. Gastrointestinal, infectious and genitourinary complications were most common (27, 23 and 17%, respectively). On multivariable analysis, increasing age group, neoadjuvant chemotherapy and receipt of blood transfusion were independent predictors of any and high grade complications, respectively. The 30- and 90-day mortality rates were 1.3 and 4.2%, respectively . In a further paper, the IRCC looked at the difference in postoperative complications between patients undergoing extracorporeal urinary diversion compared with intracorporeal urinary diversion. No difference in the re- operation rates at 30 days was noted between the groups. The overall 90-day complication rate was not significantly different between the groups, but a trend favouring intracorporeal urinary diversion was noted (41 vs 49%, P = 0.05). Gastrointestinal complications were significantly lower in the intracorporeal urinary diversion group (P ≤ 0.001). Overall patients with intracorporeal urinary diversion were at a lower risk (32%) of experiencing a postoperative complication at 90 days (odds ratio 0.68, 95% CI 0.50–0.94, P = 0.02) .
Yuh et al.  reported a large single-centre series of RARC with extraperitoneal urinary diversion, where 80% (156/196 patients) experienced a complication of any grade ≤90 days after surgery. A total of 475 adverse events (113 major) were recorded, with 365 adverse events (77%) occurring ≤30 days after surgery. A total of 35% of patients experienced a major complication within the first 90 days. The 90-day mortality rate was 4.1%.
Surgical morbidity after RARC with any approach is therefore substantial when reported using a standardized reporting methodology; however, the majority of complications are low grade and a recent meta-analysis concluded that RARC is a minimally invasive alternative to open radical cystectomy with fewer overall peri-operative complications and is associated with a shorter length of hospital stay .
Complication rates during RARC have been shown to decrease with experience . The fact that the majority of published totally intracorporeal RARC series are currently small may indicate that complication rates with this approach have further to decrease. Factors that are key to reducing complication rates include patient selection, the peri-operative care pathway, the intra-operative technique and equipment choice .
The learning curve is an important aspect to consider when developing a totally intracorporeal RARC service, as it affects delivery costs as well as patient outcomes. Studies seem to show a shorter learning curve for RARC compared with that for RARP [9, 32, 33] and yet the uptake of RARC has been much slower. It is not clear whether the reduced learning curve reflects self-selection by surgeons skilled to commence RARC. The learning curve is a difficult variable to assess, with no accepted standard definition to measure. It can be defined as a ‘self-declared’ time point when the surgeon feels comfortable performing the operation , but it may be more useful to assess single defined outcomes or a combination of defined outcome measures. On an individual's learning curve in a specific procedure, the surgeon may have peaked on certain measures such as PSM rates or complication rates, whilst they continue to improve in other measures such as length of time to complete the operation . Defining specific outcomes also makes it easier to reliably assess progress on the learning curve and identify areas of strength and those requiring improvement. There is evidence that a structured approach to totally intracorporeal RARC and mentorship can positively affect the individuals' learning curve [5, 10]. There is also evidence that surgeons experienced at open radical cystectomy have a shorter learning curve in RARC in terms of oncological outcomes such as PSM rates and lymph node yields. The IRCC concluded that an acceptable level of proficiency can be achieved after 30 cases for standard RARC .
The IRCC also looked at the effect of previous experience at RARP on the learning curve for RARC and concluded that previous RARP case volume might affect early operating times, blood loss and lymph node yield. They also concluded that surgeons with greater RARP experience are more likely to perform RARC on patients with advanced tumours. Previous RARP experience, however, did not appear to affect the surgical margin status for RARC .
Although it is difficult to predict new innovations there are strengths and trends currently evident with a robotic approach that are likely to be built on. Whilst a perceived weakness of RARC is the lack of proprioception, some experimental studies have shown that, although this is a disadvantage in simple tasks such as knot-tying, the deleterious effects from lack of tactile feedback such as breaking of threads and dropping of needles are overcome early in the learning curve [36, 37]. Three-dimensional vision has been shown to improve task performance especially with complex procedures , such as intracorporeal urinary diversion. The three-dimensional vision, magnification and removal of ‘tremor’ will also be advantageous in in vivo microscopy where images are magnified several hundred times and surgeons can potentially operate at cellular level . As RARC techniques evolve to incorporate greater magnification, the lack of proprioception may become increasingly irrelevant. The in-tile technology enables the development of real-time imaging as well as real-time discussion with histopathologists, which may both further affect PSM rates and increase lymph node yields .
With the advent of single-port surgery more complex instrumentation is currently being developed. The arrival of the four-arm robotic system greatly improved retraction in RARC, although a skilled laparoscopic assistant is still required. Newer instruments may incorporate multiple ‘arms’ in a single device that will enable retraction at several points. In intracorporeal neobladder formation this should have significant advantages. New robotic instruments/staplers to aid suturing and closure of the neobladder would further aid accessibility to this approach.
Robotic simulators are constantly being refined and complex multiple-step procedures such as RARC and urinary diversion can only benefit from their widespread introduction [41, 42]. To aid training and regulation there is likely to be increasing standardization of RARC surgery techniques . In the future, collaborative telementoring and collaborative telesurgery may contribute to improved outcomes via network links between centres .
Bladder cancer is the fourth and ninth most common malignancy in males and females, respectively in the USA and one of the most expensive cancers to manage . Totally intracorporeal RARC is the latest minimally invasive surgical option for muscle-invasive bladder cancer. Current reports have shown less blood loss, a shorter hospital stay, and equivalent complication rates when compared with a traditional open radical cystectomy approach. A totally intracorporeal approach may be advantageous in reducing postoperative complication rates compared with an extracorporeal urinary diversion . Long-term oncological results, although currently limited, are encouraging.
Functional results are similar to those of open series . In the largest published series that assessed functional outcomes after totally intracorporeal RARC with neobladder formation , there was no significant difference in continence rates between nerve-sparing and non-nerve-sparing approaches in men. Although nerve-sparing cystectomy has been shown to be associated with better continence rates , it has previously been hypothesized that this finding could be attributable to either the nerves being preserved or simply to a more meticulous dissection around the prostatic apex with less damage to the external sphincter mechanism and membranous urethra . A RARC approach may benefit both aspects through the excellent views and the EndoWrist® instrument dexterity, enabling the surgeon to perform an optimum dissection.
There is evidence that there is a shorter length of hospital stay with RARC . Multiple factors effect recovery time, including preoperative planning, appropriate surgical treatment options and postoperative enhanced recovery programmes [16, 44]. To date, the biggest impact on length of stay after RARC has come from enhanced recovery protocols . Whether the combination of a totally intracorporeal approach and enhanced recovery protocols will result in further reductions in length of stay remains to be seen. As well as shortened length of hospital stay, laparoscopic surgery results in a shorter time to return to normal activities and this aspect has yet to be evaluated in RARC. Patient selection is crucial to surgical outcomes and a large proportion of patients undergoing radical cystectomy are elderly patients with comorbidities. There is currently evidence of improved complication rates and improved overall outcomes in elderly patients undergoing radical cystectomy with a robotic approach . Successful totally intracorporeal RARC, which avoids significant complications, should be reflected in both shorter length of stay and time taken to resume normal activities, as well as further benefiting susceptible groups such as elderly patients. Nevertheless, RARC may not be a suitable approach for all patients and comorbidites and oncological staging are important considerations. The selection process should include preoperative investigation to ensure fitness for surgery as well as specific counselling about robotic surgery. Patients with reduced pulmonary compliance or poor cardiac output, who cannot tolerate the Trendelenburg position are not candidates for the robot-assisted technique. Furthermore, if the patient has a history of previous extensive abdominal surgery, RARC may be contraindicated. Patients with bulky disease should also be avoided, especially early in the learning curve [5, 15]. An intracorporeal urinary diversion may add extra time to the length of the operation, especially early in the surgeon's learning curve. A recent paper looking at compartment syndrome after robotic operations highlighted that operating times >4 h, peripheral vascular disease and obesity were the main risk factors for compartment syndrome .
Totally intracorporeal RARC surgery is complex surgery with several important outcome measures. It is crucial to optimize both oncological and functional outcomes whilst minimizing complications. Totally intracorporeal RARC has potential advantages over standard RARC with an extracorporeal urinary diversion, but prospective randomized trials comparing totally intracorporeal RARC with standard RARC and open radical cystectomy are needed to confirm this. Measures of optimum outcome should include, negative surgical margins, cancer-specific survival at 3 and 5 years, absence of major complications in the 30- and 31–90-day periods, daytime and night-time continence at 12 months, sexual activity, plus measures of length of hospital stay and time to return to normal activities. Comparative studies should also include quality-of-life or overall satisfaction scores.
A totally intracorporeal RARC is feasible with good early-, intermediate- and long-term oncological, functional and complication rate results. A complete minimally invasive approach shows consistent advantages compared with open radical cystectomy, such as blood loss and length of stay. Currently, open radical cystectomy offers a direct cost advantage because of the outlay and running costs of RARC, but when the indirect costs of complications and hospital stay are considered, RARC may be less expensive. If totally intracorporeal RARC is to become increasingly popular it must deliver on its potential for further lowering complication rates, shortening length of stay and earlier return to normal activities.
The outcomes of prospective randomized controlled trials are awaited to confirm the current findings. Future randomized controlled trials comparing totally intracorporeal with standard RARC and open surgery would benefit from validated questionnaires measuring both oncological and detailed functional outcomes, as well as complication rates at standardized times and recovery measures, including time to return to normal activities.
Conflict of Interest
robot-assisted radical cystectomy
positive surgical margin
robot-assisted radical prostatectomy
International Robotic Cystectomy Consortium