Robot-assisted laparoscopic vs open radical cystectomy: comparison of complications and perioperative oncological outcomes in 200 patients


Correspondence: Ashok K. Hemal, Wake Forest University, Medical Center Blvd., Winston-Salem, NC 27157, USA.




  • To compare perioperative morbidity and oncological outcomes of robot-assisted laparoscopic radical cystectomy (RARC) to open RC (ORC) at a single institution.

Patients and Methods

  • A retrospective analysis was performed on a consecutive series of patients undergoing RC (100 RARC and 100 ORC) at Wake Forest University with curative intent from 2006 until 2010.
  • Complication data using the Clavien system were collected for 90 days postoperatively.
  • Complications and other perioperative outcomes were compared between patient groups.


  • Patients in both groups had comparable preoperative characteristics.
  • The overall and major complication (Clavien ≥3) rates were lower for RARC patients at 35 vs 57% (P = 0.001) and 10 vs 22% (P = 0.019), respectively.
  • There were no significant differences between groups for pathological outcomes, including stage, number of nodes harvested or positive margin rates.


  • Our data suggest that patients undergoing RARC have perioperative oncological outcomes comparable with ORC, with fewer overall or major complications.
  • Definitive claims about comparative outcomes with RARC require results from larger, randomised controlled trials.

American Society of Anesthesiology


(open) (robot-assisted) radical cystectomy


Wake Forest University Baptist Medical Center


Bladder cancer will be diagnosed in 70 000 Americans, with nearly 15 000 deaths expected from this disease in 2012 [1]. While most patients present with an indolent, non-muscle-invasive disease, some have an aggressive form that poses risks of metastases and death. For these patients, radical cystectomy (RC) should be considered.

RC is a procedure that has been plagued with high rates of perioperative morbidity and mortality. The perioperative mortality rate for RC was 20% before the 1970s [2]. Improvement in anaesthetic care and a better understanding of the anatomy and surgical technique, together with stage migration, have helped drop the rate to ≈2.5% [3-5]. Despite significant progress with respect to mortality, complication rates are still as high as 70% when strict criteria are employed [6]. Furthermore, it appears a plateau has been reached for improvement in perioperative outcomes using this procedure, with no significant improvement in complication rates over the past several decades [7].

In an on-going attempt to decrease surgical morbidity associated with open RC (ORC), we sought to examine the impact of robot-assisted laparoscopic RC (RARC) on perioperative outcomes by examining a consecutive series of patients undergoing RARC and comparing them to an historical control group of patients undergoing ORC at our institution.

Patients and Methods

An Institutional Review Board-approved retrospective chart review was initiated on all patients undergoing RC at Wake Forest University Baptist Medical Center (WFUBMC) between 2006 and 2010. A cohort of the last 200 patients undergoing RC with curative intent for bladder carcinoma at WFUBMC, 100 ORC performed by four surgeons and 100 RARC performed by two surgeons (A.K.H. and A.K.K.), were included in this analysis. It is important to note that neither robotic surgeon performed an ORC during this timeframe. Patients seeing all six surgeons after January 2008 were offered the option of RARC or ORC.

We queried our database for patient demographics, preoperative disease characteristics, perioperative variables, and pathological outcomes. Of note, transfusion rates reflect transfusions that occurred at any time during a patient's hospitalisation for RC. The Clavien system was used to characterise complications and 90-day complication rates are reported [8]. The American Society of Anesthesiology (ASA) classification system was used to report preoperative comorbidity [9]. Standard descriptive methods were used to characterise the two cohorts. The Student's t-test was used for parametric and the Mann–Whitney U-test for nonparametric comparisons of continuous data. The Fisher's exact test was used to compare categorical data between the two groups.

ORC was performed via a midline abdominal incision in the standard fashion. The RARC technique was first described by Menon et al. [10] in 2003, and has subsequently been published by several authors [11]. We perform RARC with subtle modifications from the original description [12]. Of note, the lymphadenectomy boundaries used during dissection include the genitofemoral nerve laterally, the bladder medially, Cloquet's node distally, the obturator nerve and its vessels posteriorly, and the mid-common iliac vessels proximally for both ORC and RARC. All urinary diversions performed at the time of RARC in this cohort of patients were performed extracorporeally.


The preoperative characteristics of the ORC and RARC groups are listed in Table 1. There were no significant differences in sex, age, body mass index, ASA classification, number of previous abdominal surgeries, and exposure to pelvic radiation or neoadjuvant chemotherapy. Of interest, 78% of RARC and 73% of ORC patients had an ASA classification ≥3. Furthermore, 52% of RARC and 45% of ORC patients had undergone previous abdominal or pelvic surgery.

Table 1. Clinical characteristics
Sex, n (%)   
Male73 (72)74 (72)0.132
Female27 (28)29 (28) 
Age, year   
Mean (range)66 (34–86)67 (47–90)0.822
Median (IQR)67 (61, 74)67 (62, 73) 
BMI, kg/m2   
Mean (range)27.1 (16–45)26.5 (17–42)0.183
ASA Class, n (%)   
228 (27)23 (22) 
367 (66)76 (74) 
46 (7)4 (4) 
Number of previous: n (%)   
Abdominal surgery46 (45)54 (52)0.414
Prostate cancer7 (7)9 (9)0.262
Abdominal radiation6 (6)4 (4)1.0
Systemic chemotherapy10 (10)10 (10)0.597

Operative and postoperative characteristics are shown in Table 2. The mean operating times were significantly shorter for ORC (393 min) than for RARC (451 min, P < 0.001). Most patients in both groups underwent ileal conduit urinary diversion, with 97% of RARC and 83% of ORC patients undergoing this form of diversion (P = 0.032). Blood loss and transfusion rates were lower in the RARC group than the ORC group, at 423 vs 986 mL and 15 vs 47%, respectively (P < 0.001). No patients required conversion from RARC to ORC.

Table 2. Operative and postoperative characteristics
Mean (range) operative duration, min393 (239–639)451 (259–692)<0.001
Diversion, n (%)   
Ileal conduit84 (83)100 (97)0.032
Continent cutaneous5 (5)0 
Orthotopic neobladder12 (13)3 (3) 
Mean (range) EBL, mL986 (200, 4000)423 (50, 2600)<0.001
Required transfusion, n (%)47 (47)15 (15)<0.001
Hospital stay, days   
Mean (range)12.2 (5–45)7.8 (3–21)<0.001
Median (IQR)8 (7, 12)6 (5, 8) 
90-day re-admission rate, n (%)18 (18)18 (17) 
90-day complications, Clavien grade, n (%)   
043 (43)67 (65) 
I7 (7)1 (1) 
II30 (30)25 (24) 
III11 (11)6 (6) 
IV11 (11)3 (3) 
V01 (1) 
Total58 (57)36 (35)0.001
Major complications22 (22)10 (10)0.019

The overall 90-day complication rate was 35% in the RARC group and 57% in the ORC group (P = 0.001), with a major complication rate (Clavien ≥3) of 10% and 22%, respectively (P = 0.019). There was one death in the RARC group. This was a 66-year-old man with cirrhosis who had completed neoadjuvant chemotherapy. His case was unremarkable, but on postoperative day 18 he developed a subacute aorto-bilateral iliac thrombosis from which he was unable to recover. In addition, one patient in the RARC group was noted on postoperative day 3 to have left gastrocnemius compartment syndrome, requiring fasciotomy. There were no enduring sequelae in this patient.

Perioperative pathological outcomes are given in Table 3. In all, 30% of RARC and 24% of ORC patients had lymph node positive disease (P = 0.308). Lymph node yields were comparable between the groups, with mean node counts of 17.7 for RARC and 15.7 for ORC (P = 0.551). There were positive margins in 11% of ORC and 12% of RARC patients (P = 0.864).

Table 3. Pathological characteristics
CharacteristicORC, n (%)RARC, n (%)P
Pathological stage:   
pT0, Ta, Tis, T1, T253 (53)60 (58)0.491
pT3, T447 (47)43 (42) 
Pathological positive nodes (pN+)24 (24)31 (30)0.308
Mean (range) total lymph nodes15.7 (0–39)17.7 (2–52)0.551
Positive margin11 (11)12 (12)0.864


We describe herein our experience with RARC and compare perioperative characteristics, oncological outcomes and complication rates with our mature ORC results. The findings suggest that RARC is associated with longer operating times, less blood loss, lower transfusion rates, and decreased hospital stays. Importantly, a significant decrease in overall and high-grade complication rates with comparable short-term oncological outcomes was seen when comparing RARC with ORC. To our knowledge, the present study represents the largest of only three series directly comparing the complication and short-term oncological outcomes of these two approaches [13-15].

Robot-assisted laparoscopic surgery has revolutionised the field of radical prostatectomy and is now being applied to RC. Due to the aggressiveness of bladder cancer and the comorbidity of patients presenting for RC, a careful, systematic evaluation is critical to avoid early adoption of a technique that may not benefit, and in fact may harm, the patient.

There have been several published reports of RARC and ORC, but due to differences in local practice patterns, patient selection and reporting methods, it has been difficult to compare the two techniques [3, 16, 17]. Therefore, a comparison of two cohorts using standardised reporting such as the Clavien system, together with 90-day follow-up, is important for the evaluation of the two approaches. The highest level of evidence is a randomised control trial, of which there is only one published by Nix et al. [13] examining outcomes in 41 select patients, 21 of whom underwent RARC. Given the few patients, the study was only powered to detect a difference in lymph node yield. Thus it is difficult to draw any significant conclusions about complication rates, but there were comparable perioperative benefits and similar oncological outcomes as in our present experience. The only other such published comparison is by Ng et al. [14], who studied 187 patients undergoing RC, 84 of whom underwent RARC. Again, as in the present series, there was a benefit seen in overall and high-grade complications with equivalent perioperative oncological outcomes. In fact, if validated, this decrease in complication rate may be the finding that drives wide-spread adoption of the RARC technique. Comparable complication rates for RARC have been reported by other investigators, suggesting that this technique is generalizable and that this finding of decreased complication rates is robust (Table 4) [14, 18-20].

Table 4. Comparison of complication rates in RARC series
ReferenceNumber of patientsMajor complication, %Overall complication, %
Present study1001035
Khan et al. [18]501034
Smith et al. [20]227730
Ng et al. [14]831048
Hayn et al. [19]1562452

Although RARC had a favourable morbidity profile when compared with ORC, one patient in the present series developed left gastrocnemius compartment syndrome, which was probably a result of his position during RARC. This individual was not obese, was well positioned and was well padded. He was left in stirrups for the urinary diversion part of the procedure. We strongly recommend that the patient should be out of stirrups immediately upon undocking the robot.

It is important to note that the present study involved a comparable group of patients presenting for RC performed by four surgeons. In all, 65% of patients in the RARC group had been exposed to previous surgery or pelvic radiation, and 78% had an ASA class ≥3 (Table 1). Given the importance of this procedure for cancer-specific mortality and the substantial complication rate associated with it, the programme was not entered into lightly. It was only after recognising that an experienced robotic team was in place that we began to develop our RARC programme. Furthermore, one surgeon with extensive prior experience with laparoscopic and robotic surgery (A.K.H.) provided careful and dedicated mentorship of the other robotic surgeon in this series (A.K.K.) [21].

Critical to any cancer surgery are pathological outcomes, which were no different in the two groups. In all, 30% of RARC patients and 24% of ORC patients had node-positive disease, nodal counts were comparable in both groups, and there were positive margins in 12% of RARC and 11% of ORC patients. Of note, all positive margins occurred in patients with locally advanced disease, pT4 in the vast majority, with some patients with pT3 disease. These outcomes are comparable with large contemporary series of RC from high-volume centres of excellence [3, 22]. How these results translate to long-term oncological outcomes remains to be seen, but RARC appears to have comparable intermediate-term oncological outcomes compared with ORC, as was noted by Martin et al. [23].

The present analysis has several shortcomings, particularly the biases that may be introduced by a non-randomised analysis. Despite examining an unselected series of patients, an opportunity may still exist for unmeasured selection biases. One possible source of such a bias could be the higher number of ileal conduits performed in RARC as compared with ORC patients. It is uncertain if this was due to the selection of robotics, a reflection of a different patient demographic, or a trend towards ileal conduit over time. Regardless, the potential impact of this on the measured outcomes is unclear. However, in the study by Ng et al. [14], the rate of continent diversions was comparable in both groups, and a benefit in complication rate was still seen. Finally, a formal cost analysis was beyond the scope of the present study but has been addressed in other studies. While upfront costs of RARC exceed that of ORC, when length of stay and the cost of complications are considered, the cost of RARC has been found to be comparable or possibly lower than ORC [24, 25].

In conclusion, RARC resulted in less blood loss and shorter hospital stays with decreased overall and high-grade complication rates when compared with ORC. Furthermore, perioperative oncological markers were no different between the two approaches. Although encouraging, long-term functional and oncological control rates and randomised, multi-institutional comparisons of these techniques will be required before wide-spread adoption.

Conflict of Interest

None declared.