Platinum Priority – Editorial
Referring to the article published on pp. 1042–1050 of this issue

“The Devil Is in the Details”: Randomized Trial of Robotic Versus Open Radical Cystectomy

By: Mihir M. Desai and Inderbir S. Gill lowast

European Urology, Volume 67 Issue 6, June 2015, Pages 1053-1055

Published online: 01 June 2015

Abstract Full Text Full Text PDF (187 KB)

Refers to article:

Comparing Open Radical Cystectomy and Robot-assisted Laparoscopic Radical Cystectomy: A Randomized Clinical Trial eulogo1

Bernard H. Bochner, Guido Dalbagni, Daniel D. Sjoberg, Jonathan Silberstein, Gal E. Keren Paz, S. Machele Donat, Jonathan A. Coleman, Sheila Mathew, Andrew Vickers, Geoffrey C. Schnorr, Michael A. Feuerstein, Bruce Rapkin, Raul O. Parra, Harry W. Herr and Vincent P. Laudone

Accepted 21 November 2014

June 2015 (Vol. 67, Issue 6, pages 1042 - 1050)

“The Devil Is in the Details”—in 2011, this was the title of an insightful editorial about a paper on robotic radical prostatectomy [1] . In 2015, this title seems equally relevant for examining the prospective randomized trial of open versus robotic radical cystectomy (RC) by Bochner et al in this month's issue ofEuropean Urology [2] .

To start, we must recognize a basic fact about this randomized trial. It compared open and robotic cystectomy only in terms of the extirpative component (ie, RC and lymphadenectomy [LND]). The reconstructive component (urinary diversion [UD]) was done identically in each arm, with the same open surgery, by the same open surgeons, and with the same open technique.

It is common knowledge that dominant driver of the overall perioperative complications and morbidity from this operation is not its extirpative component (RC) but rather its reconstructive component (UD). Because UD was performed with open surgery in both arms of this study, it follows that this study cannot truly answer the primary question that the authors set out to evaluate: Does robotic RC have reduced complications compared with open RC? To be fair, at the time this trial was initiated, robotic intracorporeal UD was somewhat in its early stages and thus may not have been a practical consideration for this team.

We applaud the authors for conceiving and completing a surgical prospective randomized controlled trial [2] . This admirable, rare accomplishment moves the debate forward, and some important issues are addressed by this level 1 trial. In the setting of a potentially lethal cancer, robotic RC provides oncologic outcomes equivalent to those of open RC; positive surgical margin rates, lymph node yield, and lack of early recurrence—all surrogates for oncologic quality of surgery—were similar between the groups. In addition, overall perioperative complications were equivalent between the groups, although wound complications were significantly greater in the open cohort (14% vs 3.3%;p = 0.04); however, as mentioned, the design of this study does not allow it to truly address this issue. Finally, robotic RC was associated with less blood loss.

Yet, more questions were raised than answered.

What was the basis for powering the study with an expectation of a 20% reduction of morbidity in the robotic cohort? Why not 15% or 25%? Given that the primary driver of morbidity, the UD, was performed by open surgery by the same surgeons using the same technique in both arms, it could be argued that the principal findings of the study (similar complications, length of stay, quality of life [QOL] outcomes) were only to be expected.

Why were only 25% of eligible patients ultimately enrolled in this single-institution study? Such a low level of patient accrual introduces the possibility of selection bias. Who selected and counseled the patients during the initial selection interview? Did robotic surgeons participate equally in this critical process or was it controlled primarily by the open surgeons?

Before embarking on this study, what was the comparative prior RC experience of the participating open and robotic surgeons? Important prior work from the authors’ institution has already shown that surgical volumes and experience are of huge importance and predict outcomes [3] . The open RC surgeons at Memorial Sloan Kettering Cancer Center (MSKCC) have performed literally thousands of open RCs to date; however, the prior RC experience of MSKCC robotic surgeons is unclear. Prior experience with “thousands of robotic pelvic procedures” does not necessarily translate immediately into facility with robotic RC; we can attest to this from personal experience. Precisely how many robotic RCs had this team performed before starting this trial? Given the absence of any peer-reviewed publication of robotic RC from MSKCC (to our knowledge), it is likely that this number was rather low.

Blood transfusion rates were not reported. This is surprising. Transfusion rates are important and may affect long-term oncologic outcomes [4] . In daily practice, surgeons typically and intuitively consider blood loss as a surrogate for the technical quality of surgery. Blood loss was superior in the robotic arm.

Assessment of early perioperative morbidity is lacking. We are at a loss to understand the rationale of writing a protocol evaluating minimally invasive surgery in which the first documentation of QOL and physical and emotional functioning is delayed until 3–6 mo after surgery and not obtained sooner. The primary benefit of minimally invasive surgery is likely quicker andearlypostoperative recovery. Because no immediate postoperative recovery data were evaluated (eg, analgesic requirements, bowel function, quantity and timing of oral intake, postoperative nutrition), questions about earlier perioperative recovery remain unanswered.

Aggregate operative time was reportedly 2 h longer in the robotic cohort; however, Bochner et al [2] did not provide the time breakdown between the extirpative and reconstructive components for each cohort. UD was performed identically by open surgery in each arm and presumably took the same amount of time. Consequently, it is rather surprising that robotic RC/LND took longer than open RC/LND by almost 2 h. The explanation is likely twofold. First, LND was performed up to the aortic bifurcation or higher in 78% of robotic cases and only 48% of open cases, and this would add to the robotic operative time. As an important side note, the percentage of patients with disease higher than clinical T2 was greater in the open cohort (49.5% vs 57.8%); it is unclear why, surprisingly, the anatomic template of LND was actually less rigorous in the open cohort (78% vs only 48% up to or beyond aortic bifurcation). Second, in the robotic cohort, some operative time would naturally be lost when repositioning the patient from the robotic to the open approach. In evaluating these data, a recent report on time efficiency during robotic RC is pertinent [5] . Desai et al documented median operative time of 2 h for the extirpative part of the operation, which was divided approximately equally between the cystectomy and LND. Another recent report documented mean total operative time of 7.5 h for the entire robotic operation performed intracorporeally (robotic RC with completely intracorporeal orthotopic neobladder) in 132 patients [6] . In contrast, this MSKCC trial reports an operative time of 7.6 h, but all urinary diversions were performed extracorporeally (and included the less time-consuming ileal conduit in 45% of patients) [2] . Absence of data on operative time breakdown in this level 1 study prevents an educated analysis of the true reasons for this time differential. Also unclear is whether the entire diversion was performed extracorporeally or whether some steps, such as urethroileal anastomosis, were performed robotically.

Cost is an important consideration in comparing open and robotic RC. This complex operation is associated with significant perioperative morbidity and a considerable 30-d readmission rate. A complete cost analysis would ideally include not only intraoperative and immediate postoperative costs but also delayed costs due to hospital visits, readmissions, and complications. The somewhat limited cost analysis presented by Bochner et al [2] found an overall benefit in favor of open RC with a cost differential of $3920 and $1739 for neobladder and ileal conduit, respectively. We would logically expect that costs for the open/extracorporeal UD component of the operation were relatively similar between the respective cohorts, given that they were performed by the same open surgeons using the same technique. So, then, why was thecost differentialmore than double for neobladder versus ileal conduit? These data imply one of two things: Either robotic RC/LND preceding an extracorporeally performed neobladder cost $2181 more than robotic RC/LND preceding an extracorporeally performed ileal conduit or the open neobladder after robotic RC was more expensive than the open neobladder after open RC, even though the same open surgeon(s) performed both diversions. Neither makes inherent sense.

Finally, the longer robotic operative time in this study is the main driver of its two primary conclusions—longer operative time and higher costs in the robotic cohort—ergo, the need to scrutinize and understand these data in more detail. Theoretically speaking, in this trial, had operative times been similar between the two cohorts, the two main arguments of the study against robotic RC would go away. This discrepancy further underscores the unfortunate problem of the missing data on operative time breakdown already discussed.

Other concerns have recently been raised regarding robotic RC and diversion surgery [7] . These include concerns regarding compromised technique for extracorporeal and intracorporeal robotic orthotopic neobladder, compromised functional outcomes of robotic orthotopic neobladder, decreased usage of continent diversion, and inappropriate outcomes reporting. It is heartening that all of these concerns have either been addressed already or are in the process of being addressed. Indeed, “quality of surgery is critical to success” [7] . Convincing data already exist attesting to high levels of robotic technical quality (less blood loss, acceptable complication profile) and robotic oncologic quality (low negative margin rates, high nodal yields). With regard to robotic functional quality, each essential technique or principle of open orthotopic neobladder has already been completely duplicated intracorporeally with the robotic approach [8] —detubularization, cross (double) folding, achieving a spherical shape, exclusive use of intracorporeal suturing—all to achieve larger radius, greater volume, and ultimately lower pressures.

The University of Southern California (USC) technique routinely incorporates the above essentials [8] . Preliminary data on functional outcomes of robotic orthotopic neobladder seem encouraging. We performed urodynamic evaluation in 12 patients at a median of 8.6 mo (range: 2–19.5) after robotic RC with intracorporeal orthotopic neobladder. Urodynamic data were as follows: postvoid residual of 82 ml (range: 0–325), compliance of 33 ml/cm H2O (in the normal range), first sensation at 227 ml, and bladder capacity on cystometrogram of 514 ml (range: 339–1001). No patient had detrusor overactivity, and two patients had incontinence on provocative maneuvers. Furthermore, Bladder Cancer Index urinary domain, function, and bother data appeared similar those for open orthotopic neobladder (unpubl. data, V. Satkunasivam, Los Angeles, CA, USA). At USC, continent orthotopic neobladder is used in approximately 35–40% of patients undergoing robotic RC; these data compare favorably with national data for open RC with orthotopic neobladder. The sweeping critique of inappropriate outcome reporting may have had some truth to it in the initial robotic RC publications; however, most contemporary robotic series report complications and outcomes using standard classifications.

So, what is the contemporary landscape of robotic RC? Tremendous strides have been made in the past few years. In 2015, robotic RC is no longer experimental; worldwide, >2000 robotic RCs have been performed at multiple institutions by multiple surgeons, attesting to its reproducibility. In the United States, its use has increased 21-fold, from 0.6% in 2004 to 12.8% in 2010 [9] . At USC, use of robotic RC increased from 4% of all RCs in 2009 to 42% in 2014 (unpubl. data, I. Gill, Los Angeles, CA, USA). Minimally invasive cystectomy has excellent 5- and 10-yr oncologic outcomes[10] and [11]. The complete menu of robotic intracorporeal diversion options, including continent cutaneous diversion, is now available.

These are still early days for robotic RC. Nevertheless, the current trends in robotic kidney and prostate surgery should provide some insights into the shape of things to come. A majority of kidney and prostate cancer surgery is already done with a minimally invasively approach, mostly robotic. Reasonable people can debate, even differ, on whether these trends are based on clearly demonstrated or merely perceived benefits. Yet, the current reality is what it is. These trends seem unlikely to revert toward open surgery. Given the ongoing growth trajectory of robotic RC, why would one expect the cystectomy story to be any different?

We agree that any morbidity advantage of robotic over open RC remains to be demonstrated. To address this issue, the necessary next step is to perform a prospective randomized study comparing open and completely intracorporeal robotic RC. Once such data are available, the real question would be how much superior does robotic RC have to be to justify its routine use? At this writing, it can be stated confidently that, with regard to technical rigorousness and quality of surgery, robotic RC with intracorporeal UD already equals open RC in terms of extirpative, oncologic, and reconstructive results at select centers of excellence. We are confident that with increasing experience and ongoing technical and technological refinements, operative time and cost efficiency will also equal those of open surgery; however, this remains to be demonstrated. We commend Bochner et al [2] on moving the discussion forward and look forward to additional trials.

Conflicts of interest

The authors have nothing to disclose.


  • [1] J.A. Eastham, P.T. Scardino. The devil is in the details. Eur Urol. 2011;59:708-709 Crossref
  • [2] B.H. Bochner, G. Dalbagni, D.D. Sjoberg, et al. Comparing open radical cystectomy and robot-assisted laparoscopic radical cystectomy: a randomized clinical trial. Eur Urol. 2015;67:1042-1050
  • [3] C.B. Begg, E.R. Riedel, P.B. Bach, et al. Variations in morbidity after radical prostatectomy. N Engl J Med. 2002;346:1138-1144 Crossref
  • [4] Soubra A, Zabell JR, Adejoro O, Konety BR. Effect of perioperative blood transfusion on mortality for major urologic malignancies. Clin Genitourin Cancer. In press.
  • [5] M.M. Desai, A.L. de Abreu, A.C. Goh, et al. Robotic intracorporeal urinary diversion: technical details to improve time efficiency. J Endourol. 2014;28:1320-1327
  • [6] M.M. Desai, I.S. Gill, A.L. de Castro Abreu, et al. Robotic intracorporeal orthotopic neobladder during radical cystectomy in 132 patients. J Urol. 2014;192:1734-1740
  • [7] R.E. Hautmann, H.W. Herr, R.S. Pruthi, M. Aron. Robotic radical cystectomy—is the diversion the Achilles’ heel?. J Urol. 2014;192:1601-1603
  • [8] A.C. Goh, I.S. Gill, D.J. Lee, et al. Robotic intracorporeal orthotopic ileal neobladder: replicating open surgical principles. Eur Urol. 2012;62:891-901 Crossref
  • [9] J.J. Leow, S.W. Reese, W. Jiang, et al. Propensity-matched comparison of morbidity and costs of open and robot-assisted radical cystectomies: a contemporary population-based analysis in the United States. Eur Urol. 2014;66:569-576
  • [10] G.P. Haber, I.S. Gill. Laparoscopic radical cystectomy for cancer: oncological outcomes at up to 5 years. BJU Int. 2007;100:137-142 Crossref
  • [11] D.C. Snow-Lisy, S.C. Campbell, I.S. Gill, et al. Robotic and laparoscopic radical cystectomy for bladder cancer: long-term oncologic outcomes. Eur Urol. 2014;65:193-200 Crossref


Section of Robotic Surgery, Catherine and Joseph Aresty Department of Urology, USC Institute of Urology, Keck School of Medicine of USC, University of Southern California, Los Angeles, CA, USA

lowast Corresponding author. USC Institute of Urology, University of Southern California, 1441 Eastlake Ave, Suite 7416, Los Angeles, CA 90089, USA. Tel. +1 323 865 3794; Fax: +1 323 865 0120.

Place a comment

Your comment *

max length: 5000