Context
Although open radical cystectomy (ORC) is still the standard approach, laparoscopic radical cystectomy (LRC) and robot-assisted radical cystectomy (RARC) are increasingly performed.
Objective
To report on a systematic literature review and cumulative analysis of pathologic, oncologic, and functional outcomes of RARC in comparison with ORC and LRC.
Evidence acquisition
Medline, Scopus, and Web of Science databases were searched using a free-text protocol including the termsrobot-assisted radical cystectomyorda Vinci radical cystectomyorrobot* radical cystectomy. RARC case series and studies comparing RARC with either ORC or LRC were collected. A cumulative analysis was conducted.
Evidence synthesis
The searches retrieved 105 papers, 87 of which reported on pathologic, oncologic, or functional outcomes. Most series were retrospective and had small case numbers, short follow-up, and potential patient selection bias. The lymph node yield during lymph node dissection was 19 (range: 3–55), with half of the series following an extended template (yield range: 11–55). The lymph node–positive rate was 22%. The performance of lymphadenectomy was correlated with surgeon and institutional volume. Cumulative analyses showed no significant difference in lymph node yield between RARC and ORC. Positive surgical margin (PSM) rates were 5.6% (1–1.5% in pT2 disease and 0–25% in pT3 and higher disease). PSM rates did not appear to decrease with sequential case numbers. Cumulative analyses showed no significant difference in rates of surgical margins between RARC and ORC or RARC and LRC. Neoadjuvant chemotherapy use ranged from 0% to 31%, with adjuvant chemotherapy used in 4–29% of patients. Only six series reported a mean follow-up of >36 mo. Three-year disease-free survival (DFS), cancer-specific survival (CSS), and overall survival (OS) rates were 67–76%, 68–83%, and 61–80%, respectively. The 5-yr DFS, CSS, and OS rates were 53–74%, 66–80%, and 39–66%, respectively. Similar to ORC, disease of higher pathologic stage or evidence of lymph node involvement was associated with worse survival. Very limited data were available with respect to functional outcomes. The 12-mo continence rates with continent diversion were 83–100% in men for daytime continence and 66–76% for nighttime continence. In one series, potency was recovered in 63% of patients who were evaluable at 12 mo.
Conclusions
Oncologic and functional data from RARC remain immature, and longer-term prospective studies are needed. Cumulative analyses demonstrated that lymph node yields and PSM rates were similar between RARC and ORC. Conclusive long-term survival outcomes for RARC were limited, although oncologic outcomes up to 5 yr were similar to those reported for ORC.
Patient summary
Although open radical cystectomy (RC) is still regarded as the standard treatment for muscle-invasive bladder cancer, laparoscopic and robot-assisted RCs are becoming more popular. Templates of lymph node dissection, lymph node yields, and positive surgical margin rates are acceptable with robot-assisted RC. Although definitive comparisons with open RC with respect to oncologic or functional outcomes are lacking, early results appear comparable.
Radical cystectomy and pelvic lymph node dissection (PLND) is the gold standard treatment for muscle-invasive bladder cancer (MIBC) and high-risk non–muscle-invasive disease [1] . Patients undergoing this operation can experience 66% recurrence-free survival at 10 yr after surgery [2] . The addition of neoadjuvant platinum-based chemotherapy has been shown to improve overall survival (OS) rates by approximately 5% [3] . Robot-assisted radical cystectomy (RARC) was initially described by Menon et al in 2003 [4] . Over time, many international centers have adopted RARC.
Oncologic outcomes from large population-based cohorts of RARC with lengthy follow-up are lacking. Early on in RARC history, surrogates for oncologic control were reported using positive surgical margin (PSM) rates and lymph node yields. More recently, 5-yr survival figures have become available. The majority of these outcomes, however, capture institutions early in their learning curves and incorporate patients potentially selected for the robotic technique, thus avoiding more advanced-stage or technically difficult cases. Data on functional consequences of RARC are even more limited; therefore, the quality of nerve sparing and its effect on potency recovery and continence are inadequately understood.
Because of the expanding evidence available in the field of RARC, and in preparation for the Pasadena international consensus meeting on best practice in RARC and urinary reconstruction, we performed a systematic literature review of perioperative, functional, and oncologic outcomes of RARC in comparison with open radical cystectomy (ORC) and laparoscopic radical cystectomy (LRC).
We report on the systematic review and cumulative analysis of oncologic and functional outcomes of RARC. We systematically examined lymph node yields, PSMs, cancer-specific survival (CSS), disease-free survival (DFS), and OS. In addition, functional outcomes after RARC, including urinary continence and erectile function, were systematically examined.
A systematic literature search was initially performed in September 2013 using the Medline, Scopus, and Web of Science databases. The searches included only a free-text protocol using the termsrobot-assisted radical cystectomyorda Vinci radical cystectomyorrobot* radical cystectomyin all the fields of the records for Medline and Scopus searches and in theTitleandTopicfields for the Web of Science search. No limits were applied. A full update of the searches was performed on April 28, 2014.
Two authors (G.N. and B.Y.) separately reviewed the records to select RARC case series and studies that compared RARC with ORC and RARC with LRC. Discrepancies were resolved by open discussion. Other significant studies cited in the reference lists of the selected papers were evaluated, as were studies published after the systematic search.
All noncomparative studies reporting the following data on RARC were collected: intraoperative and perioperative data (operative time, blood loss, transfusion rate, in-hospital stay, readmission, complication rates), functional data (urinary continence, erectile function), and oncologic data (PSMs, lymph node yield, DFS, CSS, OS). The present review included only studies reporting on functional and oncologic data.
Studies reporting on partial cystectomy, prostate-sparing cystectomy, salvage cystectomy, cystectomy for urachal cancer or benign disease, single-case reports, pure laparoscopic (or mixed) series, or laparoendoscopic single-site or natural orifice transluminal endoscopic surgery for radical cystectomy; experimental studies on animal models; congress abstracts; review papers; editorials; population-based studies; and book chapters were not included in the review. All data retrieved from the selected studies were recorded in an electronic database.
All papers were categorized according to the 2011 levels of evidence (LOEs) for therapy studies: LOE 1, systematic review of randomized trials or n-of-1 trials; LOE 2, randomized trial or observational study with dramatic effect; LOE 3, nonrandomized controlled cohort/follow-up study; LOE 4, case series, case–control study, or historically controlled study; or LOE 5, mechanism-based reasoning [5] . Papers were categorized according to the IDEAL recommendations [6] .
Cumulative analysis was conducted using Review Manager v5.2 software designed for composing Cochrane Reviews (Cochrane Collaboration, Oxford, UK). Statistical heterogeneity was tested using the chi-square test. Apvalue <0.10 was used to indicate heterogeneity. Where there was a lack of heterogeneity, fixed-effects models were used for the cumulative analysis. Random-effects models were used in case of heterogeneity. For continuous outcomes, the results were expressed as weighted mean differences and standard deviations (SDs); for dichotomous variables, results were given as odds ratios (ORs) and 95% confidence intervals (CIs). Because of limitations in the Review Manager v5.2 software, meta-analysis of continuous variables was possible only when rough data were presented as mean and SD. Authors of the papers were contacted to provide missing data, whenever necessary. For all statistical analyses, two-sidedp < 0.05 was considered statistically significant.
Figure 1 shows a flowchart of this systematic review of the literature.
In total, 65 surgical series[4], [7], [8], [9], [10], [11], [12], [13], [14], [15], [16], [17], [18], [19], [20], [21], [22], [23], [24], [25], [26], [27], [28], [29], [30], [31], [32], [33], [34], [35], [36], [37], [38], [39], [40], [41], [42], [43], [44], [45], [46], [47], [48], [49], [50], [51], [52], [53], [54], [55], [56], [57], [58], [59], [60], [61], [62], [63], [64], [65], [66], [67], [68], [69], and [70]and 22 comparative studies[71], [72], [73], [74], [75], [76], [77], [78], [79], [80], [81], [82], [83], [84], [85], [86], [87], [88], [89], [90], [91], and [92]reported on pathologic, oncologic (n = 18), or functional (n = 9) outcomes of RARC.
Most surgical series were retrospective, single-center studies (LOE 4). Exceptions included prospective studies[8], [22], [28], [35], [36], [43], [52], [59], [64], [66], [81], and [87]and some multi-institutional retrospective collaboration studies[16], [23], [25], [30], [45], [48], [54], [61], and [68]. Only two of the comparative studies were randomized[74] and [83](LOE 2b); all other comparative studies were nonrandomized, whether prospective or retrospective (LOE 4).
Table 1 summarizes the number of lymph nodes recovered in published RARC series. The majority of studies (86%) reported extent of lymph node dissection (LND), with more centers performing extended LND (ELND) in recent series. Standard LND typically involved the removal of obturator, internal iliac, external iliac, and some portion of the common iliac lymph nodes bilaterally. ELND templates typically brought the proximal extent up to the aortic bifurcation or inferior mesenteric artery. Approximately half of the analyzed studies reported following an extended template of dissection.
Table 1 Lymph node yields in robot-assisted radical cystectomy series
| Reference | Institution | IDEAL stage | Cases, no. | Study design | Extension of LND | Operative time, min | Retrieved nodes, no. | pN+, % | Metastatic nodes, median, no. | Complications due to LND |
|---|---|---|---|---|---|---|---|---|---|---|
| Menon et al, 2003 [4] | Henry Ford Hospital | 1 | 17 | Retrospective | Standard | – | – | 6 | – | – |
| Menon et al, 2004 [7] | Henry Ford Hospital | 1 | 3 female | Retrospective | Standard | – | 12 | 0 | – | – |
| Guru et al, 2007 [8] | Roswell Park Cancer Institute | 1 | 20 | Prospective | Standard | 44 | 13 | 15 | 1 | – |
| Mottrie et al, 2007 [9] | O.L.V.–Clinic | 2a | 27 | Retrospective | Extended | – | 23 | 9 | – | – |
| Pruthi et al, 2008 [71] | UNC | 2a | 20 | Retrospective | Standard | – | 19 | 10 | – | – |
| Hemal et al, 2008 [10] | All India Institute of Medical Sciences | 1 | 6 | Retrospective | Standard | – | 12 | 17 | – | – |
| Lowentritt et al, 2008 [11] | Tulane University | 2a | 4 | Retrospective | Standard | – | 12 | 25 | – | – |
| Murphy et al, 2008 [12] | Guy's Hospital | 2a | 23 | Retrospective | Standard | – | 16 | 9 | – | – |
| Park et al, 2008 [13] | Yonsei | 2a | 4 | Retrospective | Standard | – | 17 | 0 | – | – |
| Pruthi et al, 2008 [14] | UNC | 2a | 12 female | Retrospective | Standard, then extended | – | 19 | 17 | – | – |
| Pruthi et al, 2008 [15] | UNC | 2b | 50 | Retrospective | Standard | – | 19 | 20 | – | – |
| Wang et al, 2008 [72] | Cornell | 2b | 33 | Retrospective | Standard | – | 17 | 19 | – | – |
| Woods et al, 2008 [16] | Mayo Arizona Tulane University |
2b | 27 | Multi-institutional | Extended | – | 12.3 | 33 | 3.1 | 0 |
| Yuh et al, 2008 [17] | Roswell Park Cancer Institute | 2a | 54 | Retrospective | Extended | – | 17 | – | – | – |
| Gamboa et al, 2009 [18] | University of California, Irvine |
2a | 41 | Retrospective | Standard | – | 23 | 14 | 4 | – |
| Pruthi et al, 2009 [19] | UNC | 2b | 50 | Retrospective | Standard, then extended | – | 19 | 16 | – | – |
| 10 female | 19 | |||||||||
| 40 male | 18 | |||||||||
| Palou Redorta et al, 2009 [20] | Barcelona Autonomous University | 2a | 9 | Retrospective | Extended | 60 | 10 | 0 | – | – |
| Yuh et al, 2009 [21] | Roswell Park Cancer Institute | 2b | 73 | Retrospective | Extended | – | 19 | – | – | – |
| Guru et al, 2010 [22] | Roswell Park Cancer Institute | 2a | 26 | Prospective | Extended | – | 21 | 29 | 1 | Internal iliac artery injury: 1 |
| Hellenthal et al, 2011 [23] | IRCC | 2b | 437 | Multi-institutional | – | – | 17 | 20 | – | – |
| Josephson et al, 2010 [24] | City of Hope Cancer Center | 2b | 58 | Retrospective | Extended | – | 27 | 24 | – | – |
| Kang et al, 2010 [25] | Multicenter | 2b | 71 standard LND 33 extended LND |
Retrospective | Standard Extended |
– | 15.7 24.7 |
10 | – | – |
| Kasraeian et al, 2010 [26] | Montsouris Institute | 2a | 9 | Retrospective | Extended | – | 11 | 22 | – | 0 |
| Kauffman et al, 2011 [27] | Cornell | 2b | 85 | Retrospective | Extended | – | 19 | 15 | – | – |
| Kwon et al, 2010 [28] | Kyungpook National University | 2a | 17 | Prospective | Standard | – | 6 | 6 | 1 | 0 |
| Lavery et al, 2011 [29] | Ohio State University | 2a | 15 | Retrospective | Extended | 107 | 41.8 | 20 | – | 0 |
| Martin et al, 2010 [30] | Mayo Arizona Tulane University |
2b | 59 | Multi-institutional | Extended | – | – | 34 | – | – |
| Ng et al, 2010 [73] | Cornell | 2b | 83 | Retrospective | Standard | – | 16 | 16 | – | – |
| Nix et al, 2010 [74] | UNC | 3 | 21 | RCT | Standard | – | 19 | 19 | – | – |
| Pruthi et al, 2010 [31] | UNC | 2b | 100 | Retrospective | Standard, then extended | – | 19 | 20 | – | – |
| Richards et al, 2010 [75] | Wake Forest University | 2b | 35 | Retrospective | Extended | – | 16 | 29 | – | – |
| Akbulut et al, 2011 [32] | Ankara Ataturk Training and Research Hospital | 2a | 12 | Not reported | Extended | – | 21.3 | 42 | – | 8 |
| Canda et al, 2012 [33] | Ankara Ataturk Training and Research Hospital | 2a | 27 | Not reported | Extended | – | 24.8 | 22 | – | – |
| Davis et al, 2011 [34] | University of Texas M.D. Anderson Cancer Center | 2a | 11 | Retrospective | Extended | 117 | 43 | 9 | 1 | – |
| Jonsson et al, 2011 [35] | Karolinska Institute | 2b | 45 36 neobladder 9 ileal conduit |
Prospective | Standard Extended |
– | 19 19 27 |
20 17 33 |
– | – |
| Khan et al, 2011 [36] | Guy's Hospital | 2a | 50 | Prospective | – | – | 17 | – | – | – |
| Manoharan et al, 2011 [37] | University of Miami | 2a | 14 | Retrospective | Standard | – | 12 | – | – | – |
| Martin et al, 2011 [76] | Mayo Arizona | 2b | 19 | Retrospective | – | – | 16 | – | – | – |
| Richards et al, 2011 [38] | Wake Forest University | 2b | 60 | Retrospective | Extended | – | 17 | 30 | – | Lymphocele: 1 |
| Schumacher et al, 2011 [39] | Karolinska Institute | 2b | 45 | Retrospective | Standard 49%, extended 31% | – | 22.5 | – | 1.5 | Lymphocele: 2 |
| Shah et al, 2011 [40] | Ohio State University | 2b | 30 | Retrospective | Extended | – | – | 30 | – | – |
| Torrey et al, 2011 [41] | City of Hope Cancer Center | 2b | 34 | Retrospective | Extended | – | 28.9 | – | – | – |
| Cho et al, 2012 [42] | Hallym University College of Medicine | 2b | 35 | Retrospective | Standard | – | – | 6 | – | – |
| Goh et al, 2012 [43] | Keck School of Medicine, University of Southern California, Los Angeles | 2a | 15 | Prospective | Superextended | – | 55 | 26 | – | – |
| Lau et al, 2012 [44] | City of Hope Cancer Center | 2b | 23 (aged >80 yr) | Retrospective | Extended | – | 20.4 | 22 | – | – |
| Mmeje et al, 2013 [45] | Mayo Arizona UNC |
2b | 50 | Multi-institutional | Extended | – | 18 | 100 | 3 | – |
| Poch et al, 2012 [46] | Roswell Park Cancer Institute | 2b | 56 | Retrospective | – | – | 25 | 16 | – | – |
| Richards et al, 2012 [77] | Wake Forest University | 2b | 20 (aged >75 yr) | Retrospective | Extended | – | 17 | 35 | – | – |
| Saar et al, 2013 [47] | Saarland University | 2b | 62 | Retrospective | – | – | 14.2 | 21 | – | – |
| Smith et al, 2012 [48] | Mayo Arizona UNC, Tulane University |
2b | 227 | Multi-institutional | – | – | 18 | 20 | – | – |
| Styn et al, 2012 [78] | University of Michigan | 2b | 50 | Retrospective | – | – | 14.3 | 12 | – | – |
| Sung et al, 2012 [79] | Samsung Medical Center | 2b | 35 | Retrospective | Standard | – | 19.1 | 26 | – | Lymphocele: 1 |
| Treiyer et al, 2012 [49] | Saarland University | 2b | 91 | Retrospective | Standard | – | 14.5 | 14 | – | – |
| Tsui et al, 2012 [50] | Chang Gung Memorial Taiwan | 2a | 8 | Retrospective | Standard | – | 3 | 12.50 | – | – |
| Yuh et al, 2012 [51] | City of Hope Cancer Center | 2b | 196 | Retrospective | Extended | – | 28 | 22 | – | Lymphocele: 3 |
| Collins et al, 2013 [52] | Karolinska Institute | 2b | 113 | Prospective | Extended 56%, standard 34%, limited 5%, none 5% | – | 21 | 20 | – | Lymphocele: 5 |
| Johar et al, 2013 [53] | IRCC | 2b | 939 | Multi-institutional | – | – | 18.1 | 26 | – | – |
| Maes et al, 2013 [80] | Metro Health Hospital | 2b | 14 | Retrospective | Extended | – | 11.9 | 7 | – | – |
| Marshall et al, 2013 [54] | IRCC | 2b | 765 | Multi-institutional | Extended 58%, standard 40%, no LND 2% | – | 18 | 27 | – | – |
| Musch et al, 2014 [81] | Klinikin Essen–Mitte | 2b | 100 | Prospective | – | – | 26.5 | 20 | – | Lymphocele: 4 |
| Nazmy et al, 2014 [55] | City of Hope Cancer Center | 2b | 209 | Retrospective | Extended | – | – | 22 | – | Lymphocele: 3 |
| Nepple et al, 2013 [82] | Washington University | 2b | 36 | Retrospective | Standard | – | 17 | 22 | – | – |
| Parekh et al, 2013 [83] | University of Texas Health Sciences Center at San Antonio | 3 | 20 | RCT | Standard | – | 11 | 20 | – | – |
| Tyritzis et al, 2013 [56] | Karolinska Institute | 2b | 70 | Retrospective | Standard 43% Extended 48% |
– | 21 | 14 | – | Lymphocele: 6 Lymphedema: 1 |
| Xylinas et al, 2013 [57] | Cornell | 2b | 175 | Retrospective | Standard | – | 19 | 17 | – | Lymphocele: 2 |
| Phillips et al, 2014 [69] | Seward St. Elizabeth Medical Center | 2b | 23 (>80 yr) | Retrospective | Extended | – | 19 | – | – | |
| Raza et al, 2014 [70] | Roswell Park Cancer Institute | 2b | 99 | Retrospective | – | – | 20.7 | 36 | – | – |
| Yuh et al, 2014 [58] | City of Hope Cancer Center | 2b | 162 | Retrospective | Extended | – | 28 | 23 | – | – |
| Total | 19.3 | 23 |
IRCC = International Robotic Cystectomy Consortium; LND = lymph node dissection; RCT = randomized controlled trial; UNC = University of North Carolina.
The lymph node yield from all series was 19 (range: 3–55). Initial descriptions using a standard template of dissection achieved yields of 18 lymph nodes [59] . Number of lymph nodes recovered with an ELND ranged from 11 to 55. Abaza et al adopted a robotic template similar to the open technique, including external iliac, obturator, hypogastric, common iliac, and presacral up to the aortic bifurcation; the mean lymph node yield was 37.5 (SD: 13.2), demonstrating that lymph node counts could mirror those of open dissection if the same template was followed [88] . In a study of open completion LND after robot-assisted ELND in 11 men, Davis et al removed only an additional 4 lymph nodes with an open approach after 43 were removed with robot assistance [34] . Time of LND was rarely reported, although it ranged from 44 min in standard LND to 117 min in ELND[8] and [34]. The lymph node–positive rate was 22%. In series with >20 RARCs, lymph node–positive rates ranged from 6% to 42%. Reports of vascular injuries were rare, and lymphocele rates were 0–9%.
Table 2 summarizes the studies assessing the effects of patient characteristics and particular surgical aspects on lymph node yields in RARC series. Cumulative analysis from the International Robotic Cystectomy Consortium (IRCC) with respect to lymphadenectomy in 437 patients found a median of 17 lymph nodes removed, with a 20% node-positivity rate [23] . Patient age and sex did not affect the performance of lymphadenectomy. In a different series, increasing body mass index (BMI) did not appear to negatively affect lymph node yield, with >20 lymph nodes removed in normal, overweight, and obese patients [46] .
Table 2 Impact of patient characteristics and surgical aspects on lymph node yield in robot-assisted radical cystectomy series
| Reference | Institution | IDEAL stage |
Cases | Study design | Extension of LND | Operative time, min | Retrieved nodes, no. | pN+, % |
|---|---|---|---|---|---|---|---|---|
| Patient BMI | ||||||||
| Poch et al, 2012 [46] | Roswell Park Cancer Institute | 2b | 56 | Retrospective | – | – | 25 | 16 |
| BMI <25: 14 | 22 | 7 | ||||||
| BMI 25 to <30: 21 | 23 | 14 | ||||||
| BMI ≥30: 21 | 20 | 24 | ||||||
| Case volume | ||||||||
| Guru et al, 2008 [59] | Roswell Park Cancer Institute | 2a | 1–12 | Prospective | Extended | 46 | 33% >13 | 8 |
| 13–24 | 44 | 66% >13 | 33 | |||||
| 25–36 | 41 | 83% >13 | 25 | |||||
| 37–47 | 43 | 72% >13 | 64 | |||||
| 48–58 | 56 | 91% >13 | 18 | |||||
| Pruthi et al, 2008 [60] | University of North Carolina | 2b | 50 | Retrospective | Standard | – | 19 | – |
| 1–10 | 21 | |||||||
| 11–20 | 19 | |||||||
| 21–30 | 20 | |||||||
| 31–40 | 17 | |||||||
| 41–50 | 20 | |||||||
| Richards et al, 2011 [38] | Wake Forest University | 2b | 60 | Retrospective | Extended | – | 17 | 30 |
| 1–20 | 17 | |||||||
| 21–40 | 19.1 | |||||||
| 41–60 | 14.4 | |||||||
| Schumacher et al, 2011 [39] | Karolinska Institute | 2b | 45 | Retrospective | Standard 49% | – | 22.5 | – |
| Extended 31% | ||||||||
| 1–15 | Standard 40% | |||||||
| Extended 7% | ||||||||
| 16–30 | Standard 47% | |||||||
| Extended 53% | ||||||||
| 31–45 | Standard 60% | |||||||
| Extended 33% | ||||||||
| Prior RARP experience | ||||||||
| Hayn et al, 2010 [61] | IRCC | 2b | 496 | Retrospective | – | – | 17.8 | – |
| ≤50 previous RARP: 83 | 13.7 | |||||||
| 51–100 previous RARP: 187 | 19.8 | |||||||
| 101–150 previous RARP: 176 | 19.6 | |||||||
| >150 previous RARP: 50 | 11.8 * | |||||||
* Statistically significant.
BMI = body mass index; RCC = International Robotic Cystectomy Consortium; LND = lymph node dissection; RARP = robot-assisted radical prostatectomy.
It is interesting to note that in single-institution series, Richards et al [38] , Schumacher et al [39] , Guru et al [59] , and Pruthi et al [60] did not find higher lymph node yields with increasing sequential case numbers. However, in the IRCC, performance of lymphadenectomy was positively correlated with surgeon and institution volume but was reduced in patients with more advanced disease (pT4 stage), which may reflect operative avoidance of bulky nodal tissue.
Table 3 summarizes the occurrence of PSMs reported in the RARC series. The reported PSM rates were 5.6% (range: 0–26%). However, in series of >100 patients, margin rates ranged between 4% and 9%[48] and [53]. PSMs were reported in 1–1.5% of patients with pT2 disease and 0–25% of patients with pT3 and higher disease. PSM rates from the IRCC in 939 cases were 9% [53] .
Table 3 Positive surgical margins in robot-assisted radical cystectomy series
| Reference | Institution | IDEAL stage | Cases, no. | Study design | Pathologic stage, % | Overall PSM rate, % | PSM location | PSM rate, % | ||
|---|---|---|---|---|---|---|---|---|---|---|
| ≤pT2 | ≥pT3 | ≤pT2 | ≥pT3 | |||||||
| Menon et al, 2003 [4] | Henry Ford Hospital | 1 | 17 | Retrospective | – | – | 0 | – | 0 | 0 |
| Yohannes et al, 2003 [62] | Creighton University | 1 | 2 | Retrospective | 0 | 100 | 0 | – | 0 | 0 |
| Menon et al, 2004 [7] | Henry Ford Hospital | 1 | 3 female | Retrospective | 66 | 33 | 0 | – | 0 | 0 |
| Rhee et al, 2006 [84] | University of Virginia | 1 | 7 | Retrospective | 43 | 57 | 0 | – | 0 | 0 |
| Guru et al, 2007 [8] | Roswell Park Cancer Institute | 1 | 20 | Prospective | 40 | 60 | 15 | Prostate: 1 Ureter: 1 Vagina: 1 |
0 | 25 |
| Mottrie et al, 2007 [9] | O.L.V.–Clinic | 2a | 27 | Retrospective | 78 | 22 | 4 | Ureter: 1 | – | – |
| Pruthi et al, 2008 [71] | UNC | 2a | 20 | Retrospective | 70 | 20 | 0 | – | 0 | 0 |
| Hemal et al, 2008 [10] | All India Institute of Medical Sciences | 1 | 6 | Retrospective | 67 | 33 | 0 | – | 0 | 0 |
| Lowentritt et al, 2008 [11] | Tulane University | 2a | 4 | Retrospective | 25 | 75 | 0 | – | 0 | 0 |
| Murphy et al, 2008 [12] | Guy's Hospital | 2a | 23 | Retrospective | 74 | 17 | 0 | – | 0 | 0 |
| Park et al, 2008 [13] | Yonsei | 2a | 4 | Retrospective | 50 | 50 | 0 | – | 0 | 0 |
| Pruthi et al, 2008 [14] | UNC | 2a | 12 female | Retrospective | 58 | 25 | 0 | – | 0 | 0 |
| Pruthi et al, 2008 [15] | UNC | 2b | 50 | Retrospective | 66 | 14 | 0 | – | 0 | 0 |
| Pruthi et al, 2009 [19] | UNC | 2b | 50 10 female 40 male |
Retrospective | 66 50 70 |
18 30 15 |
0 | – | 0 | 0 |
| Wang et al, 2008 [72] | Cornell | 2b | 33 | Retrospective | 72 | 28 | 6 | Perivesical fat: 2 | 0 | 22 |
| Woods et al, 2008 [16] | Mayo Arizona Tulane University |
2b | 27 | Multi-institutional | – | – | 7 | – | 0 | – |
| Yuh et al, 2008 [17] | Roswell Park Cancer Institute | 2a | 54 | Retrospective | 44 | 56 | 13 | – | 0 | 23 |
| Gamboa et al, 2009 [18] | University of California, Irvine | 2a | 41 | Retrospective | – | – | 5 | – | 0 | – |
| Palou Redorta et al, 2009 [20] | Barcelona Autonomous University | 2a | 9 | Retrospective | 66 | 33 | 11 | – | – | – |
| Yuh et al, 2009 [21] | Roswell Park Cancer Institute | 2b | 73 | Retrospective | 45 | 55 | 10 | – | 0 | 18 |
| Guru et al, 2010 [22] | Roswell Park Cancer Institute | 2a | 20 | Prospective | 62 | 38 | 4 | – | 0 | 9 |
| Hayn et al, 2010 [61] | IRCC | 2b | 482 | Multi-institutional | 64 | 36 | 7 | – | – | – |
| Hellenthal et al, 2010 [68] | IRCC | 2b | 513 | Multi-institutional | 64 | 36 | 7 | – | 1.50 | 17 |
| Kang et al, 2010 [25] | Multicenter | 2b | 104 | Multi-institutional | 70 | 30 | 5 | – | – | – |
| Kasraeian et al, 2010 [26] | Montsouris Institute | 2a | 9 | Retrospective | 44 | 66 | 0 | – | – | – |
| Kauffman et al, 2011 [27] | Cornell | 2b | 85 | Retrospective | 64 | 36 | 6 | – | 0 | 16 |
| Kwon et al, 2010 [28] | Kyungpook National University | 2a | 17 | Prospective | 59 | 41 | 0 | – | 0 | 0 |
| Martin et al, 2010 [30] | Mayo Arizona Tulane University |
2b | 59 | Multi-institutional | 47 | 53 | – | – | – | – |
| Ng et al, 2010 [73] | Cornell | 2b | 83 | Retrospective | 61 | 39 | 7 | – | 0 | 19 |
| Nix et al, 2010 [74] | UNC | 3 | 21 | RCT | 67 | 14 | 0 | – | 0 | 0 |
| Pruthi et al, 2010 [31] | UNC | 2b | 100 | Retrospective | 67 | 13 | 0 | – | 0 | 0 |
| Richards et al, 2010 [75] | Wake Forest University | 2b | 35 | Retrospective | 60 | 40 | 3 | – | – | – |
| Akbulut et al, 2011 [32] | Ankara Ataturk Training and Research Hospital | 2a | 12 | Not reported | 58 | 42 | 0 | – | 0 | 0 |
| Canda et al, 2012 [33] | Ankara Ataturk Rraining and Research Hospital | 2a | 27 | Not reported | 56 | 44 | 4 | – | 0 | 4 |
| Davis et al, 2011 [34] | University of Texas M.D. Anderson Cancer Center | 2a | 11 | Retrospective | 92 | 8 | 0 | – | – | – |
| Jonsson et al, 2011 [35] | Karolinska Institute | 2b | 45 | Prospective | 78 | 22 | 2 | – | 0 | 10 |
| Khan et al, 2011 [36] | Guy's Hospital | 2a | 50 | Prospective | 72 | 28 | 2 | – | 0 | 7 |
| Manoharan et al, 2011 [37] | University of Miami | 2a | 14 | Retrospective | – | – | 0 | – | 0 | 0 |
| Martin et al, 2011 [76] | Mayo Arizona | 2b | 19 | Retrospective | 42 | 58 | – | – | – | – |
| Richards et al, 2011 [38] | Wake Forest University | 2b | 60 | Retrospective | 63 | 37 | 10 | – | – | – |
| Schumacher et al, 2011 [39] | Karolinska Institute | 2b | 45 | Retrospective | 78 | 22 | 2 | Ureter: 1 | 0 | 10 |
| Shah et al, 2011 [40] | Ohio State University | 2b | 30 | Retrospective | 65 | 35 | 7 | – | 0 | 22 |
| Cho et al, 2012 [42] | Hallym University College of Medicine | 2b | 35 | Retrospective | 86 | 14 | 3 | – | – | – |
| Goh et al, 2012 [43] | Keck School of Medicine, University of Southern California, Los Angeles | 2a | 15 | Prospective | 67 | 33 | 0 | – | – | – |
| Lau et al, 2012 [44] | City of Hope Cancer Center | 2b | 23 (aged >80 yr) | Retrospective | 61 | 39 | 13 | Ureter: 1 | – | – |
| Mmeje et al, 2013 [45] | Mayo Arizona | 2b | 50 | Multi-institutional | 34 | 66 | 2 | – | – | – |
| UNC | ||||||||||
| Poch et al, 2012 [46] | Roswell Park Cancer Institute | 2b | 56 | Retrospective | 55 | 45 | – | – | – | – |
| Richards et al, 2012 [77] | Wake Forest University | 2b | 20 (aged >75 yr) | Retrospective | 60 | 40 | 5 | – | – | – |
| Saar et al, 2013 [47] | Saarland University | 2b | 62 | Retrospective | 64 | 36 | 2 | – | – | – |
| Smith et al, 2012 [48] | Mayo Arizona | 2b | 227 | Multi-institutional | – | – | 2 | – | – | – |
| UNC, Tulane University | ||||||||||
| Styn et al, 2012 [78] | University of Michigan | 2b | 50 | Retrospective | 60 | 40 | 2 | – | – | – |
| Sung et al, 2012 [79] | Samsung Medical Center | 2b | 35 | Retrospective | 43 | 57 | – | – | – | – |
| Treiyer et al, 2012 [49] | Saarland University | 2b | 91 | Retrospective | 67 | 33 | 2 | Urethra: 1 Prostate: 1 | – | – |
| Tsui et al, 2012 [50] | Chang Gung Memorial | 2a | 8 | Retrospective | 75 | 25 | 0 | – | 0 | 0 |
| Yuh et al, 2012 [51] | City of Hope Cancer Center | 2b | 196 | Retrospective | 64 | 36 | 4 | – | – | – |
| Azzouni et al, 2013 [63] | Roswell Park Cancer Institute | 2b | 100 | Retrospective | 35 | 65 | 4 | – | – | – |
| Collins et al, 2013 [52] | Karolinska Institute | 2b | 113 | Prospective | 75 | 25 | 5 | Ureter: 1 | 1 | 18 |
| Johar et al, 2013 [53] | Multicenter | 2b | 939 | Retrospective | 49 | 51 | 9 | – | – | – |
| Maes et al, 2013 [80] | Metro Health Hospital | 2b | 14 | Retrospective | 43 | 57 | 21 | – | – | – |
| Marshall et al, 2013 [54] | IRCC | 2b | 765 | Multi-institutional | 59 | 41 | – | – | – | – |
| Musch et al, 2014 [81] | Klinikin Essen–Mitte | 2b | 100 | Prospective | 61 | 39 | 2 | – | – | – |
| Nazmy et al, 2014 [55] | City of Hope Cancer Center | 2b | 209 | Retrospective | 65 | 35 | 3 | – | – | – |
| Nepple et al, 2013 [82] | Washington University | 2b | 36 | Retrospective | 53 | 47 | 6 | – | 0 | 12 |
| Parekh et al, 2013 [83] | University of Texas Health Sciences Center at San Antonio | 3 | 20 | RCT | 50 | 50 | 5 | – | 0 | 10 |
| Tyritzis et al, 2013 [56] | Karolinska Institute | 2b | 70 | Retrospective | 86 | 14 | 1.5 | Ureter: 1 | 0 | 10 |
| Xylinas et al, 2013 [57] | Cornell | 2b | 175 | Retrospective | 65 | 35 | 5 | – | – | – |
| Phillips et al, 2014 [69] | Seward St. Elizabeth Medical Center | 2b | 23 (aged >80 yr) | Retrospective | 30 | 70 | 26 | – | – | – |
| Raza et al, 2014 [70] | Roswell Park Cancer Institute | 2b | 99 | Retrospective | 48 | 52 | 8 | – | – | |
| Yuh et al, 2014 [58] | City of Hope Cancer Center | 2b | 162 | Retrospective | 67 | 33 | 4 | – | – | – |
| Total | 60 | 40 | 5.6 | |||||||
IRCC = International Robotic Cystectomy Consortium; PSM = positive surgical margin; RCT = randomized controlled trial; UNC = University of North Carolina.
Table 4 summarizes the studies assessing the effects of patient characteristics and particular surgical aspects on PSM rates in RARC series. Notably, Richards et al [38] , Schumacher et al [39] , and the IRCC [68] did not demonstrate decreasing surgical margin rates with sequential case number. In a study of the role of previous robot-assisted radical prostatectomy (RARP) experience on RARC outcomes, there was a trend toward increased positive margins with increasing RARP volumes, but it did not reach statistical significance (p = 0.089) [61] . The authors chiefly attributed this situation to the performance of RARC on patients with higher risk (higher than T3) disease. One study reported that PSMs occurred only in the overweight or obese patients, although pT4 rates were much higher in those patients (26% vs 7%) [65] .
Table 4 Predictors of positive surgical margins in robot-assisted radical cystectomy series
| Reference | Institution | IDEAL stage | Cases | Study design | Pathologic stage, % | Overall PSM rate, % | PSM rate, % | ||
|---|---|---|---|---|---|---|---|---|---|
| ≤pT2 | ≥pT3 | ≤pT2 | ≥pT3 | ||||||
| Case volume | |||||||||
| Guru et al, 2008 [59] | Roswell Park Cancer Institute | 2a | 1–12 | Prospective | 33 | 66 | 17 | – | – |
| 13–24 | 58 | 42 | 25 | ||||||
| 25–36 | 50 | 50 | 0 | ||||||
| 37–47 | 46 | 54 | 9 | ||||||
| 48–58 | 64 | 36 | 0 | ||||||
| Hayn et al, 2011 [64] | Roswell Park Cancer Institute | 2a | 1–50 | Prospective | 51 | 49 | 8 | – | – |
| 51–100 | 12 | ||||||||
| 101–164 | 6 | ||||||||
| Richards et al, 2011 [38] | Wake Forest University | 2b | 60 | Retrospective | 63 | 37 | 10 | – | – |
| 1–20 | 55 | 45 | 5 | ||||||
| 21–40 | 70 | 30 | 5 | ||||||
| 41–60 | 65 | 35 | 20 | ||||||
| Schumacher et al, 2011 [39] | Karolinska Institute | 2b | 45 | Retrospective | 78 | 22 | 2 | 0 | 10 |
| 1–15 | 87 | 13 | 0 | ||||||
| 16–30 | 67 | 33 | 7 | ||||||
| 31–45 | 80 | 20 | 0 | ||||||
| Azzouni et al, 2013 [63] | Roswell Park Cancer Institute | 2b | 100 | Retrospective | 35 | 65 | 4 | – | – |
| 1–25 | 36 | 64 | 4 | ||||||
| 26–50 | 40 | 60 | 4 | ||||||
| 51–75 | 44 | 56 | 4 | ||||||
| 76–100 | 20 | 80 | 4 | ||||||
| Previous RARP experience | |||||||||
| Hayn et al, 2010 [61] | IRCC | 2b | 482 | Retrospective | 64 | 36 | 7 | – | – |
| ≤50 previous RARP: 83 | 68 | 32 | 4 | ||||||
| 51–100 previous RARP: 173 | 76 | 24 | 5 | ||||||
| 101–150 previous RARP: 168 | 54 | 46 | 9.5 | ||||||
| >150 previous RARP: 48 | 42 | 58 | 12.5 | ||||||
| Patient BMI | |||||||||
| Butt et al, 2008 [65] | Roswell Park Cancer Institute | 2a | BMI <25: 14 | Retrospective | 64 | 36 | 0 | 0 | – |
| BMI 25–29: 18 | 28 | 72 | 28 | ||||||
| BMI ≥30: 17 | 42 | 58 | 6 | ||||||
| Poch et al, 2012 [46] | Roswell Park Cancer Institute | 2b | 56 | Retrospective | 55 | 45 | – | – | – |
| BMI <25: 14 | 50 | 50 | |||||||
| BMI 25 to <30: 21 | 57 | 43 | |||||||
| BMI ≥30: 21 | 52 | 48 | |||||||
| Intracorporeal vs extracorporeal diversion | |||||||||
| Kang et al, 2012 [85] | Korea University School of Medicine | 2a | 38 extracorporeal diversion | Retrospective | 76 | 24 | 2.5 | – | – |
| 4 intracorporeal diversion | 100 | 0 | 0 | ||||||
BMI = body mass index; IRCC = International Robotic Cystectomy Consortium; PSM = positive surgical margin; RARP = robot-assisted radical prostatectomy.
Table 5 summarizes the oncologic outcomes of current RARC publications. Neoadjuvant chemotherapy use was reported in 0–31% of patients. Adjuvant chemotherapy use was reported in 4–29% of patients.
Table 5 Survival outcomes in robot-assisted radical cystectomy series
| Reference | Institution | IDEAL stage | Cases, no. | Study design | Follow-up, mo | Neoadjuvant chemotherapy, % | Adjuvant chemotherapy, % | DFS estimates, % | CSS estimates, % | OS estimates, % | ||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 1 yr | 3 yr | 5 yr | 1 yr | 3 yr | 5 yr | 1 yr | 3 yr | 5 yr | ||||||||
| Pruthi et al, 2008 [15] | UNC | 2b | 50 | Retrospective | 13.2 | 0 | 22 | – | – | – | 94 (13 mo) | – | – | 90 (13 mo) | – | – |
| Murphy et al, 2008 [12] | Guy's Hospital | 2a | 23 | Retrospective | 17 | 29 | – | 91 (17 mo) | – | – | – | – | – | – | – | – |
| Josephson et al, 2010 [24] | City of Hope Cancer Center | 2b | 58 | Retrospective | 12 | 22 | – | – | 76 (2 yr) | – | – | 76 (2 yr) | – | – | 54 (2 yr) | – |
| Kang et al, 2010 [25] | Multicenter | 2b | 104 | Retrospective | 12 | – | – | 96 | – | – | – | – | – | – | – | – |
| Kauffman et al, 2011 [27] | Cornell | 2b | 85 | Retrospective | 18 | 20 | 12 | 79 | 73 (2 yr) | 88 | 85 (2 yr) | 83 | 79 (2 yr) | |||
| Martin et al, 2010 [30] | Mayo Arizona | 2b | 59 | Multi-institutional | 21 | 17 | – | 82 | 71 | – | – | – | – | 82 | 72 | – |
| Tulane University | ||||||||||||||||
| Pruthi et al, 2010 [31] | UNC | 2b | 100 | Retrospective | 21.2 | 5 | 18 | – | – | – | 94 (21 mo) | – | – | 91 (21 mo) | – | – |
| Canda et al, 2012 [33] | Ankara Ataturk Training and Research Hospital | 2a | 27 | Not reported | 6 | – | 4 | 85 (6 mo) | – | – | 89 (6 mo) | – | – | 72 (6 mo) | – | – |
| Mmeje et al, 2013 [45] | Mayo Arizona | 2b | 50 | Multi-institutional | 41.5 | 12 | 46 | – | 43 | 39 | – | – | – | – | 55 | 45 |
| UNC | ||||||||||||||||
| Treiyer et al, 2012 [49] | Saarland University | 2b | 91 | Retrospective | 15 | 0 | – | – | – | – | 94 (15 mo) | – | – | 93 (15 mo) | – | – |
| Collins et al, 2013 [52] | Karolinska Institute | 2b | 113 | Prospective | 25 | 31 | – | – | – | – | – | 81 | 67 | – | 80 | 66 |
| Khan et al, 2013 [66] | Guy's & St. Thomas Hospital | 1 | 14 | Prospective | 84 | 28 | 14 | 50 | 75 | 64 | ||||||
| Nepple et al, 2013 [82] | Washington University | 2b | 36 | Retrospective | 12 | 6 | – | – | 67 (2 yr) | – | – | 75 (2 yr) | – | – | 68 (2 yr) | – |
| Snow-Lisy et al, 2014 [67] | Cleveland Clinic | 2b | 17 | Retrospective | 67 | – | – | – | – | – | – | – | 69 | – | – | 39 |
| Tyritzis et al, 2013 [56] | Karolinska Institute | 2b | 70 | Retrospective | 30.3 | 24 | – | – | 81 (2 yr) | – | – | 89 (2 yr) | – | – | 89 (2 yr) | – |
| Xylinas et al, 2013 [57] | Cornell | 2b | 175 | Retrospective | 37 | 19 | – | 67 | 63 | – | 68 | 66 | – | – | – | |
| Raza et al, 2014 [70] | Roswell Park Cancer Institute | 2b | 99 | Retrospective | 73.9 | 6 | 29 | – | – | 53 | – | – | 68 | – | – | 42 |
| Yuh et al, 2014 [58] | City of Hope Cancer Center | 2b | 162 | Retrospective | 52 | 23 | – | – | 76 | 74 | – | 83 | 80 | – | 61 | 54 |
CSS = cancer-specific survival; DFS = disease-free survival; OS = overall survival; UNC = University of North Carolina.
Several studies further analyzed the use of adjuvant chemotherapy after RARC. General indications for selecting patients for adjuvant chemotherapy included pathologic stage pT3–4 or node-positive disease. Pruthi et al described the use of adjuvant chemotherapy in 18 of 100 RARC patients, with mean time to chemotherapy initiation at approximately 7 wk, which was faster than the authors’ historical time to chemotherapy in open cystectomy of 10 wk [31] . In a randomized trial of RARC (n = 21) compared with ORC (n = 20), 7 wk was also the mean time to initiation of chemotherapy after RARC [74] . In one analysis of patients with node-positive disease at the time of RARC, 46% received adjuvant chemotherapy [45] .
Survival represents the gold standard with respect to evaluating effectiveness and risks of treatment; however, RARC reports with 5-yr outcomes have become available only recently. Data remain limited for assessing long-term outcomes, patterns of recurrence, and means for predicting survival. The role of adjuvant treatments after RARC is also poorly defined.
Series detailing cancer control outcomes had a mean follow-up between 6 and 84 mo ( Table 5 ), although only 6 of 18 series (33%) reported a mean follow-up >36 mo. At 1, 2, 3, and 5 yr, DFS was 79–96%, 67–81%, 67–76%, and 53–74%, respectively; CSS was 88–94%, 75–89%, 68–83%, and 66–80%, respectively; and OS was 82–90%, 54–89%, 61–80%, and 39–66%, respectively. In the series with longest follow-up, Khan et al described only 14 patients with ≥5 yr of follow-up, showing DFS of 50%, CSS of 75%, and OS of 64% [66] .
Several series reported on adverse oncologic outcomes associated with increased pathologic stage or lymph node involvement[58] and [93]. In a series of 162 patients with urothelial carcinoma, Yuh et al found that 5-yr survival was worse with higher pathologic stage or lymph node positivity (p < 0.01). Patients with a lymph node density of 1–10% (defined as number of positive nodes divided by number of total nodes) had DFS, CSS, and OS of 34%, 49%, and 31%, respectively, whereas patients with lymph node density >10% had further reduced survival of 30%, 38%, and 20%, respectively. Predictors of DFS were lymph node density, pathologic stage, and age-adjusted Charlson comorbidity index, whereas the same measures plus receipt of transfusion were predictive for OS [58] . Similarly, in an analysis of 99 patients with follow-up >5 yr, pathologic stage and lymph node positivity were independent predictors of DFS, CSS, and OS, whereas positive margin status and Charlson comorbidity index predicted worse OS and CSS [70] .
In series with median follow-up of >36 mo, rates of local recurrence without distant disease ranged between 0% (n = 15) and 9% (n = 99)[57], [58], [66], [67], and [70]. No port-site recurrences occurred in these series. Xylinas et al examined 175 patients with a median follow-up of 37 mo, showing recurrence of disease in 29%. Of these patients, 8 (5%) had local recurrence alone, 11 had local and distant metastases, and 32 had distant metastases alone [57] .
In an analysis of patients with positive lymph nodes (n = 50) at the time of PLND, median time to recurrence was 10 mo after RARC [45] . Estimated OS at 36 and 60 mo was 55% and 45%, respectively, with recurrence-free survival at 36 and 60 mo of 43% and 39%, respectively. Similarly, Tyritzis et al reported recurrence-free survival of 34% and OS of 63% after 24 mo in node-positive patients [56] .
Table 6 presents the RARC series reporting on continence outcomes. Although functional outcomes are a major area of study in patients undergoing RARP, a lack of data remains for evaluation after RARC. Worldwide, the number of patients evaluated for continence after orthotopic bladder substitution is <200 from nine reports at the present time. There are also widespread differences in patient selection, methods of data collection, and outcome assessment.
Table 6 Urinary continence rates in robot-assisted radical cystectomy series
| Reference | Institution | IDEAL stage | Cases, no. | Study design | Nerve-sparing surgery, % | Intracorporeal diversion, % | Follow-up, mo | Method of data collection | Continence definition | Continence rate, % | ||
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 3 mo | 6 mo | 12 mo | ||||||||||
| Mottrie et al, 2007 [9] | O.L.V.–Clinic | 2a | 27 | Retrospective | 29 | 0 | 10.2 | – | – | 86 | – | – |
| Murphy et al, 2008 [12] | Guy's Hospital | 2a | 23 | Retrospective | 20 | 0 | 17 | – | – | – | – | 100 D |
| 75 N (17 mo) |
||||||||||||
| Palou Redorta et al, 2009 [20] | Barcelona Autonomous University | 2a | 9 | Retrospective | 100 | 0 | 7 | – | – | 100 D and N (7 mo) |
||
| Canda et al, 2012 [33] | Ankara Ataturk Training and Research Hospital | 2a | 27 | Not reported | 89 | 100 | 6 | – | D: 0–1 safety pads | – | 48 | – |
| N: dry with no protection | 11 | |||||||||||
| Jonsson et al, 2011 [35] | Karolinska Institute | 2b | 36 | Prospective | 55 | 100 | 25 | – | 0–1 pads | – | – | 83 D 66 N |
| Manoharan et al, 2011 [37] | University of Miami | 2a | 14 | Retrospective | – | 0 | – | – | – | 93 D | ||
| 71 N | ||||||||||||
| Torrey et al, 2012 [41] | City of Hope Cancer Center | 2b | 34 | Retrospective | 0 | 0 (all Indiana pouch) | 12.1 | Physician charting | – | – | – | 97 |
| Goh et al, 2012 [43] | Keck School of Medicine, University of Southern California, Los Angeles | 2a | 15 | Prospective | – | 100 | 3 | – | – | 75 | ||
| Tyritzis et al, 2013 [56] | Karolinska Institute | 2b | 70 | Retrospective | 58 BNS | 100 | 12 | Internally validated questionnaire | 0–1 pads | – | D: 77 men, |
D: 88 men, |
| 8 UNS | 40 women N: 54 men, 40 women |
67 women | ||||||||||
| N: 76 men, 76 women |
||||||||||||
BNS = bilateral nerve sparing; D = daytime; N = nocturnal; UNS = unilateral nerve sparing.
Follow-up for continence evaluation varied widely, from 6 to 25 mo. Nerve-sparing procedures were performed in 20–100% of patients. Only three of six series reported using a distinct definition forcontinence, which was generally no pad or one pad (safety) per day. One of the earliest RARC series reported an 86% continent rate (seven of eight men) after 3.5 mo [9] . More recent series published 6-mo continence rates of 48–100% for daytime continence and 11–100% for nighttime continence. At 12 mo after RARC, continence rates ranged from 83% to 100% in men and were 67% in women for daytime continence and 66–76% for nighttime continence.
Using strict definitions fordaytime continence(no or one security pad per day) andnighttime continence(goodindicates dry with no protection,fairindicates dry with one awakening), Canda et al examined 23 patients with intracorporeal Studer pouch. After excluding patients who died or were lost to follow-up, 11 of 15 men (73%) and 0 of 2 women were continent during the daytime. Three of these 17 patients (18%) had good nighttime continence, and 4 (24%) had fair continence [33] .
Only one series has described continence results in patients undergoing RARC and continent cutaneous diversion. Torrey et al examined 34 patients who had RARC and Indiana pouch continent cutaneous diversion and reported 97% continence at a mean follow-up of 20 mo for both daytime and nighttime. One patient continued to experience daytime and nighttime incontinence requiring the use of pads [41] .
Table 7 summarizes the series examining potency outcomes. Similar to continence outcomes, evaluation of erectile function after RARC is not well described. Early reports suggest that erections sufficient for penetration are achievable, although sample sizes were very small and lacked validated objective evaluations. Follow-up was again too short to form definitive conclusions, with only one study reporting outcomes up to 2 yr after RARC. As noted earlier, nerve-sparing procedures were performed in 20–100% of patients. The data recording used International Index of Erectile Function (IIEF) scores in five of seven series. However, only three series provided a clear definition ofpotency[12], [35], and [56].
Table 7 Erectile function in robot-assisted radical cystectomy series
| Reference | Institution | IDEAL stage | Cases, no. | Nerve-sparing surgery, % | Study design | Follow-up, mo | Method of data collection | Potency definition | Potency rate at follow-up |
|---|---|---|---|---|---|---|---|---|---|
| Mottrie et al, 2007 [9] | O.L.V.–Clinic | 2a | 27 | 29 | Retrospective | 10.2 | – | – | 86% |
| Murphy et al, 2008 [12] | Guy's Hospital | 2a | 23 | 20 | Retrospective | 17 | IIEF | IIEF >21 with or without PDE5-I | 75% |
| Palou Redorta et al, 2009 [20] | Barcelona Autonomous University | 2a | 9 | 100 | Retrospective | 7 | – | – | 100% |
| Akbulut et al, 2011 [32] | Ankara Ataturk Training and Research Hospital | 2a | 12 | 82 bilateral 9 unilateral |
Not reported | 7.1 | IIEF | None provided | A single patient with IIEF >18 |
| Canda et al, 2012 [33] | Ankara Ataturk Training and Research Hospital | 2a | 27 | 89 | Not reported | 6 | IIEF | None provided | A single patient with IIEF >18 |
| Jonsson et al, 2011 [35] | Karolinska Institute | 2b | 36 | 55 | Prospective | 25 | IIEF | Adequate for penetration with or without PDE5-I | 41% at 12 mo 75% of patients having nerve sparing |
| Tyritzis et al, 2013 [56] | Karolinska Institute | 2b | 70 | 58 bilateral 8 unilateral |
Retrospective | 12 | IIEF | Adequate for penetration with or without PDE5-I | 63% at 12 mo |
IIEF = International Index of Erectile Function; PDE5-I = phosphodiesterase type 5 inhibitor.
In some early, small series, Mottrie et al [9] and Murphy et al [12] reported sufficient erections in six of seven and three of four men, respectively. Similar to well-described literature on RARP, phosphodiesterase type 5 inhibitors (PDE5-Is) were frequently administered to patients for penile rehabilitation after RARC; however, no comparative data in this setting have demonstrated a benefit.
Several series with intracorporeal neobladder have evaluated erectile function postoperatively, with varying results. In the experience of the Karolinska Institute, 41 of 62 men (66%) underwent nerve-sparing RARC. Of these 41 men, 26 (63%) were potent with or without the use of PDE5-Is after 12 mo [56] . In contrast, Canda et al found IIEF scores >18 in only 1 of 11 preoperatively potent men, although follow-up was shorter (6 mo) [33] .
Table 8 summarizes comparative studies evaluating lymph node yield after ORC, LRC, and RARC. In two randomized studies of ORC compared with RARC, lymph node yields were not statistically different[74] and [83]. Cumulative analyses showed no significant difference in lymph node yield between RARC and ORC (OR: 2.94; 95% CI, −0.28 to 6.15;p = 0.07) ( Fig. 2 ).
Table 8 Comparative studies evaluating lymph node yield after open, laparoscopic, and robot-assisted radical cystectomy
| Comparison | Level of evidence | Reference | Cases, no. | Study design | Extension of LND | Retrieved nodes, no. | pN+, no. (%) | Metastatic nodes, no., median |
|---|---|---|---|---|---|---|---|---|
| ORC vs RARC | 2 | |||||||
| Nix et al, 2010 [74] | 21 RARC | RCT | Standard | 19 | 4 (19) | – | ||
| 20 ORC | 18 | 7 (35) | ||||||
| Parekh et al, 2013 [83] | 20 RARC | RCT | Standard | 17.2 ± 13 | 4 (20) | – | ||
| 20 ORC | 24.2 ± 16.4 | 4 (20) | ||||||
| 3 | ||||||||
| Pruthi et al, 2008 [71] | 20 RARC | Gender matched | Standard | 19 | 2 (10) | – | ||
| 24 ORC | Retrospective | 16 | 5 (21) | |||||
| Wang et al, 2008 [72] | 33 RARC | Nonmatched | Standard | 17 | 19 | – | ||
| 21 ORC | 20 | 34 * | ||||||
| Ng et al, 2010 [73] | 83 RARC | Nonmatched | Standard | 17.9 ± 10.4 | 13 (16) | – | ||
| 104 ORC | 15.7 ± 13.2 | 24 (23) * | ||||||
| Richards et al, 2010 [75] | 35 RARC | Nonmatched | Extended | 16 | 10 (29) | – | ||
| 35 ORC | 15 | 10 (29) | ||||||
| Martin et al, 2011 [76] | 19 RARC | Nonmatched | – | 16 | – | – | ||
| 14 ORC | 13 | |||||||
| Gondo et al, 2012 [92] | 11 RARC | Nonmatched | Extended | 20.7 ± 8.2 | 9 | – | ||
| 15 ORC | 13.8 ± 6.6 * | 13 | ||||||
| Khan et al, 2012 [87] | 48 RARC | Prospective | Extended | 16 | 5 | – | ||
| 52 ORC | 11 | 15 | ||||||
| Richards et al, 2012 [77] | 20 RARC | Nonmatched | Extended | 17 | 7 (35) | – | ||
| 20 ORC (>75 yr) | 15 | 3 (15) | ||||||
| Styn et al, 2012 [78] | 50 RARC | 1:2 by age, sex, clinical stage, diversion | – | 14.3 ± 9.1 | 6 (12) | – | ||
| 100 ORC | 15.2 ± 9.5 | 19 (19) | ||||||
| Sung et al, 2012 [79] | 35 RARC | Nonmatched | Standard | 19.1 ± 8.2 | 9 (26) | – | ||
| 104 ORC | 12.9 ± 9.0 * | 27 (26) | ||||||
| Knox et al, 2013 [86] | 58 RARC | Nonmatched | Extended | 21 | 1 | – | ||
| 84 ORC | 17 | 3 | ||||||
| Maes et al, 2013 [80] | 14 RARC | Nonmatched | Extended | 11.9 | 1 (7) | – | ||
| 14 ORC | 9.5 | 5 (35) | ||||||
| Musch et al, 2014 [81] | 100 RARC | Nonmatched | – | 27.5 ± 11.0 | 20 (20) | – | ||
| 42 ORC | 19.6 ± 8.8 * | 9 (21) | ||||||
| Nepple et al, 2013 [82] | 36 RARC | Nonmatched | Standard | 17 | 8 (22) | – | ||
| 29 ORC | 14 | 7 (24) | ||||||
| 4 | Abaza et al, 2012 [88] | 35 RARC | Nonmatched | Extended | 37.5 ± 13.2 | 12 (34) | 1.5 | |
| 120 ORC | 36.9 ± 14.8 | 36 (30) | 2 | |||||
| LRC vs RARC | ||||||||
| 3 | Khan et al, 2012 [87] | 48 RARC | Prospective | Extended | 16 | 5 | – | |
| 58 LRC | 10 | 10 | ||||||
| 4 | Abraham et al, 2007 [89] | 14 RARC | Nonmatched | 10 extended | 22.3 | 2 (10) | ||
| 20 LRC | 16 extended | 16.5 | 2 (12.5) | |||||
* Statistically significant.
LND = lymph node dissection; LRC = laparoscopic radical cystectomy: ORC = open radical cystectomy; RARC = robot-assisted radical cystectomy; RCT = randomized controlled trial.
Fig. 2 Comparison of lymph node yields following robot-assisted or open radical cystectomy. CI = confidence interval; ORC = open radical cystectomy; RARC = robot-assisted radical cystectomy; SD = standard deviation.
Table 9 summarizes PSM rates in RARC, ORC, and LRC. In two randomized trials comparing RARC and ORC, Nix et al and Parekh et al did not show any increase in positive margins with RARC[74] and [83]. Cumulative analyses showed no significant difference in rates of surgical margins between RARC and ORC (5% and 7%, respectively; OR: 0.71; 95% CI, 0.46–1.1;p = 0.13) ( Fig. 3 ). In two comparative nonrandomized studies between RARC and LRC, no significant differences in PSM rates were detected (p = 0.86)[87] and [89].
Table 9 Comparative studies evaluating positive surgical margins after open, laparoscopic, and robot-assisted radical cystectomy
| Comparison | Level of evidence | Reference | Cases, no. | Pathologic stage, % | Overall PSM, no. (%) | PSM in pT2 cancer | |
|---|---|---|---|---|---|---|---|
| pT2 | pT3 | ||||||
| ORC vs RARC | 2b | ||||||
| Nix et al, 2010 [74] | 21 RARC | 67 | 14 | 0 | 0 | ||
| 20 ORC | 40 | 25 | 0 | 0 | |||
| Parekh et al, 2013 [83] | 20 RARC | 50 | 50 | 1 (5) | 0 | ||
| 20 ORC | 65 | 35 | 1 (5) | 0 | |||
| 3 | |||||||
| Rhee et al, 2006 [84] | 7 RARC | 86 | 14 | 0 | 0 | ||
| 23 ORC | 43 | 57 | 0 | 0 | |||
| Galich et al, 2006 [90] | 13 RARC | 54 | 46 | 0 | – | ||
| 24 ORC | 37 | 63 | 3 (12) | ||||
| Pruthi et al, 2007 [71] | 20 RARC | 78 | 22 | 0 | 0 | ||
| 24 ORC | 63 | 37 | 0 | ||||
| Wang et al, 2008 [72] | 33 RARC | 72 | 28 | 2 (6) | – | ||
| 21 ORC | 43 | 57 | 3 (14) | ||||
| Ng et al, 2010 [73] | 83 RARC | 61 | 39 | 6 (7) | 0 | ||
| 104 ORC | 58 | 42 | 9 (9) | 0 | |||
| Richards et al, 2010 [75] | 35 RARC | 60 | 40 | 1 (3) | – | ||
| 35 ORC | 57 | 43 | 3 (9) | ||||
| Martin et al, 2011 [76] | 19 RARC | 42 | 58 | – | – | ||
| 14 ORC | 93 | 7 | |||||
| Gondo et al, 2012 [92] | 11 RARC | 91 | 9 | 1 (9) | – | ||
| 15 ORC | 53 | 47 | 2 (13) | ||||
| Khan et al, 2012 [87] | 48 RARC | 75 | 25 | 0 | – | ||
| 52 ORC | 50 | 50 | 6 (10) | ||||
| Richards et al, 2012 [77] | 20 RARC | 60 | 40 | 1 (5) | – | ||
| 20 ORC (>75 yr) | 50 | 50 | 2 (10) | ||||
| Styn et al, 2012 [78] | 50 RARC | 60 | 40 | 1 (2) | – | ||
| 100 ORC | 72 | 28 | 1 (1) | ||||
| Sung et al, 2012 [79] | 35 RARC | 43 | 57 | – | – | ||
| 104 ORC | 38 | 62 | |||||
| Kader et al, 2013 [91] | 100 RARC | 58 | 42 | 12 (12) | – | ||
| 100 ORC | 53 | 47 | 11 (11) | ||||
| Knox et al, 2013 [86] | 58 RARC | 66 | 34 | 4 (7) | – | ||
| 84 ORC | 43 | 57 | 7 (8) | ||||
| Maes et al, 2013 [80] | 14 RARC | 43 | 57 | 3 (21) | – | ||
| 14 ORC | 57 | 43 | 2 (14) | ||||
| Musch et al, 2013 [81] | 100 RARC | 61 | 39 | 2 (2) | – | ||
| 42 ORC | 57 | 43 | 1 (2) | ||||
| Nepple et al, 2013 [82] | 36 RARC | 53 | 47 | 2 (6) | 0 | ||
| 29 ORC | 58 | 42 | 2 (7) | 0 | |||
| 4 | Abaza et al, 2012 [88] | 35 RARC | 60 | 23 | 2 (6) | 0 | |
| 120 ORC | 45 | 42 | 8 (7) | 0 | |||
| LRC vs RARC | |||||||
| 3 | Khan et al, 2012 [87] | 48 RARC | 75 | 25 | 0 | – | |
| 58 LRC | 57 | 43 | 2 (4) | ||||
| 4 | Abraham et al, 2007 [89] | 14 RARC | – | – | 1 (7) | 0 | |
| 20 LRC | 0 | – | |||||
LRC = laparoscopic radical cystectomy; ORC = open radical cystectomy; PSM = positive surgical margin; RARC = robot-assisted radical cystectomy.
Fig. 3 Comparison of positive surgical margin rates following robot-assisted or open radical cystectomy. CI = confidence interval; M-H = Mantel-Haenszel; ORC = open radical cystectomy; RARC = robot-assisted radical cystectomy.
Table 10 summarizes series that emphasized early oncologic comparisons for RARC, LRC, and ORC, though interpretation should be cautious with small series of shorter follow-up and potential bias of patient selection. A nonrandomized comparison of ORC (n = 52) with RARC (n = 48) with a follow-up of 38 mo showed disease-specific survival of 69% in the ORC group compared with 79% in the RARC group [87] . A series by Nepple et al showed similar estimates in DFS, CSS, and OS, although patients were not matched [82] .
Table 10 Comparative studies evaluating recurrence-free, cancer-specific, and overall survival estimates after open, laparoscopic, and robot-assisted radical cystectomy
| Comparison | Level of evidence | Reference | Cases, no. | Study design | Follow–up, mo | Neoadjuvant chemotherapy, % | DFS estimates, % | CSS estimates, % | OS estimates, % |
|---|---|---|---|---|---|---|---|---|---|
| ORC vs RARC | 3 | ||||||||
| Khan et al, 2012 [87] | 48 RARC | Prospective | 38 | – | – | 79 | – | ||
| 52 ORC | 69 | ||||||||
| Nepple et al, 2013 [82] | 36 RARC | Nonmatched | 12 | 6 | 67 (2 yr) | 75 (2 yr) | 68 (2 yr) | ||
| 29 ORC | 14 | 58 (2 yr) | 63 (2 yr) | 63 (2 yr) | |||||
| LRC vs RARC | 3 | ||||||||
| Khan et al, 2012 [87] | 48 RARC | Prospective | 38 | – | – | 79 | – | ||
| 58 LRC | 93 |
CSS = cancer-specific survival; DFS = disease-free survival; LRC = laparoscopic radical cystectomy; ORC = open radical cystectomy; OS = overall survival; RARC = robot-assisted radical cystectomy.
Our systematic review sought to identify and report the current state of the literature for RARC with regard to pathologic, oncologic, and functional outcomes. Various oncologic parameters, including pathologic findings and postoperative survival rates, were examined. With regard to nodal dissection, robotic ELND achieves a similar nodal yield to open ELND when performed by experienced surgeons. Nearly all RARC series reported nodal yields >15. With regard to margin rates, most series reported PSM rates of <10%, with rates of approximately 1% in pT2 disease. Although the IRCC (n = 513) reported a very high positive margin rate of 39% in pT4 patients [68] , other authors have reported rates similar to those noted in ORC series.
Although these immediate pathologic variables may act as surrogates for quality of resection, long-term survival outcomes must be analogous to those of ORC for RARC to be a viable surgical option. Currently, oncologic data are immature, and adequate comparative studies of RARC and ORC are nonexistent. In a few analyses measuring CSS and OS at 5 yr postoperatively, results appear similar to those reported in ORC; however, larger numbers and longer follow-up are needed for adequate comparison. At present, data reporting functional analysis of continence and potency recovery after RARC are inadequate to compare RARC reliably with ORC.
PLND, in conjunction with radical cystectomy, provides a staging benefit as well as a possible advantage for survival in retrospective studies. Stein et al examined 1054 patients treated with radical cystectomy and PLND with a 24% node-positive rate; these patients experienced 5- and 10-yr recurrence-free survival of 35% and 34%, respectively [2] . Although prospective validation is necessary, Leissner et al suggested that ELND improved outcomes in both low-volume node-positive and node-negative patients with greater number of lymph nodes removed [93] . The true survival benefit of ELND must be proven in a prospective fashion to overcome the Will Rogers phenomenon of apparent improved survival that results from stage migration with more thorough dissection.
Early critical concerns of RARC involved whether LND could be performed robotically with the same quality as during ORC. This review suggests that thorough robotic ELND dissection at the time of RARC is possible following a similar template as is performed during ORC. Half of current RARC series describe an extended template dissection, with the average number of lymph nodes removed between 11 and 55. In a small study of open completion LND after robotic LND, only four additional lymph nodes were recovered [34] . Although few series described the time necessary to perform a complete robotic LND, some authors described operative times approaching 2 h for the node dissection alone, suggesting that robotic LND may lengthen operative time. Further study is necessary to determine whether the LND segment of RARC is significantly longer compared with open LND. Complications specific to LND—particularly vascular injuries—were rare, as were lymphoceles, with an incidence <10%. However, complication rates may often be underreported, as reporting guidelines lack standardization.
Assessment of RARC lymph node yields as related to patient characteristics (eg, BMI) or surgeon characteristics (eg, prior RARP experience) has not shown a specific association. Similar to the ORC literature, Bochner et al reported that only extent of LND was associated with lymph node yield when examining variables such as receipt of neoadjuvant chemotherapy, pathologic stage, surgeon, and pathologist [94] . Although performance of LND was associated with higher surgeon volume, analysis of several learning curve evaluations did not find increases in lymph node yield with increasing case number. This result may seem counterintuitive, but it may be that these experienced robotic surgeons were able to translate surgical technique from RARP and PLND and thus reduce the number of cases needed to reach stable lymph node yields. Instead, a reduction in LND time could occur with experience, although it has not been specifically examined. In the IRCC database, patients with pT4 disease had lower nodal yields, possibly related to more difficult dissection or to RARC being performed for palliative intent.
PSM at cystectomy is a measure of disease burden and a predictor of outcome. In a previous study of 1589 patients who underwent radical cystectomy at Memorial Sloan Kettering Cancer Center, the positive margin rate was 4.2%. Risk factors for PSMs were female sex, higher pathologic stage, vascular invasion, mixed histology, and lymph node involvement. Patients with PSMs had a 5-yr CSS of only 32% [95] . In another large multi-institutional analysis of 4400 ORC patients, the incidence of PSMs was 6.3% [96] . A potential challenge of RARC and limitation of current robotic technology is in treating bulkier tumors because of the lack of tactile feedback.
The present systematic review demonstrates that PSMs are uncommon in RARC series and appropriately rare for pT2 disease. No significant difference was found when comparing the surgical margin rate between RARC and ORC. The high variability of positive margins across studies, between 0% and 26%, suggests significant heterogeneity in cancer characteristics, patient selection, and surgical technique and experience, among other variables. From the systematic review, the weighted average of positive margins in RARC series was 5.6%, which is comparable to the large open series cited earlier. In the aforementioned analysis of 4400 ORC patients, margin-positive rates by stage were 2.3% for pT2, 7.6% for pT3, and 24% for pT4 disease [96] . The effects of the learning curve as institutions adopted this new technology and patient selection toward earlier stage disease likely affected reported margin rates and should be considered when interpreting outcomes. Nonetheless, higher reported rates of positive margins in pT4 disease in some RARC series suggest that caution be taken for higher stage disease, with particular attention paid to the risk of margin involvement.
Several RARC series did not show decreasing margin rates with sequential case volume. A few reasons could explain this observation: (1) The positive margin numbers may be too low to detect a subgroup difference; (2) the learning curve for reducing margins at RARC could be extremely high, with a number not yet reached in smaller learning curve assessments; or (3) over time, more experienced surgeons may be more willing to take on bulky or higher stage tumors. This final hypothesis is supported by a multivariate analysis adjusting for pathologic stage that shows that differences in stage of disease accounted for an increase in margin rates with more experienced robotic surgeons [61] .
Chemotherapy use alongside surgery in the treatment of MIBC can be implemented either before or after cystectomy. While neoadjuvant chemotherapy has been shown to confer an OS advantage of 5% in randomized trials [97] , the benefit of adjuvant chemotherapy is less proven. In a recent meta-analysis of nine randomized controlled trials comprising 945 patients that investigated the use of adjuvant chemotherapy, benefits to both OS and DFS were appreciated. Patients receiving adjuvant chemotherapy after cystectomy had 23% relative risk reduction in the risk of death (OS:p = 0.049) and 34% relative decrease in the risk of disease recurrence (DFS:p = 0.014) [98] . For the current systematic review, neoadjuvant chemotherapy use was 0–31%, and adjuvant chemotherapy was delivered to 4–29% of patients. Adjuvant therapy was chiefly administered in patients with advanced-stage pT3 or higher or with positive lymph nodes. Although time to initiation of adjuvant chemotherapy was shorter by 3 wk in the analysis by Pruthi et al, further validation is required [31] .
Long-term freedom from disease recurrence and bladder cancer–related death is the primary measure of treatment efficacy with radical cystectomy. Particularly with assessments of survival, gathering data for comparison with the open standard is challenging secondary to the necessity of controlling for cancer characteristics, additional therapies, and the length of follow-up required to detect significant differences. Shorter-interval examinations of survival may not amply capture events such as local recurrence, distant recurrence, or secondary therapies. Only two series in this systematic review compared survival for RARC and ORC. These studies were not randomized and included sequential series of retrospective groups (LOE 4)[82] and [87]. In a series by Nepple et al, 36 patients who underwent RARC were compared with 29 patients who underwent ORC with a median follow-up of only 12 mo. Estimated 2-yr DFS (67% vs 58%), CSS (75% vs 63%), and OS (68% vs 63%) after RARC and ORC were similar for the two techniques, respectively [82] .
Because of limitations of present studies, comparisons must be made to large historical retrospective open series. A long-term analysis of survival in 1100 chemotherapy-naive cystectomy patients by Hautmann et al demonstrated 10-yr CSS and OS rates of 67% and 44%, respectively [99] . For this systematic review, 5-yr estimates for DFS, CSS, and OS were 53–74%, 66–80%, and 39–66%, respectively. Analogous to stratified outcomes in ORC, survival outcomes were worse in RARC series with increasing pathologic stage and with lymph node metastases. Local control of disease appears to be adequate such that the majority of recurrences after RARC are distant or outside the pelvis. A potential concern for port-site metastases with RARC remains of particular interest. Although no specific published series address this concern and most larger RARC oncologic series did not report any incidents, a few case reports suggest that this concern requires further study.
Since the original description of neurovascular bundle preservation during radical prostatectomy by Walsh et al [104] , techniques to improve functional outcomes through meticulous nerve sparing have been translated to radical cystectomy. Turner et al determined that nerve sparing improved urinary continence after orthotopic urinary diversion [100] , and nerve sparing has been shown to assist with recovery of erectile function objectively based on IIEF [101] . Long-term functional evaluations of ileal neobladder continent diversions have demonstrated daytime continence rates of 92% and nighttime continence rates of 80% [102] .
To date, very limited data are available regarding functional outcomes of continence or potency after RARC. These analyses have chiefly been limited to only a few centers that exhibit significant heterogeneity. The 12-mo reported continence rates were 88–100% in men and 67% in women for daytime continence and between 66% and 75% for nighttime continence. Potency recovery exhibited even greater variation, with sufficient erection rates between 9% and 81%. Functional outcomes are likely influenced by patient factors and selection, comorbidity, prior treatments, surgeon experience, and technique (eg, the use of cautery vs clips). In addition, methodology of reporting, definitions ofcontinence, measurement tools, rehabilitation programs, and inconsistencies in follow-up can affect the actual measurement of continence and potency. Specific functional concerns of RARC related to patient selection are that many patients may be older or have poor baseline erectile function. Moreover, technical concerns for a possible PSM, which portends a dismal outcome, may affect the performance of nerve sparing. The lack of conclusive data regarding functional recovery after RARC is a necessary area for future study.
There is no evidence to date that the results from a recent systematic review on RARP finding slight advantages to continence and potency recovery compared with open radical prostatectomy or laparoscopic radical prostatectomy extrapolate to RARC [103] . Precise definitions ofcontinenceandpotencyare necessary so that future data acquisition can be carried out in a standardized, stringent, and uniform fashion for both ORC and RARC.
From a methodological perspective, the most relevant limitations of this systematic review are the quality of the available studies, the small number of patients in and the retrospective nature of most series, the shorter-term follow-up of these studies, and the lack of standardized definitions. The papers included in the present review included only two small randomized controlled trials; the remaining series are LOE 3 or 4. Comparisons made in these single-institution studies inevitably carry the risk of selection bias. Even in randomized controlled studies, there were unlikely to have been equally experienced open and robotic surgeons operating on comparable patients. Heterogeneity in lymph node templates, sampling methods, specimen handling, and pathologic review may affect lymph node yields. Most cumulative outcomes were weighted by the results of experienced surgeons, which may make conclusions difficult to generalize. The inability to account for surgeon factors or specific technique modifications is another limitation. Most series failed to provide specific information concerning relevant aspects of the reconstructive portions of the operation.
Sufficient lymph node yields are achievable through robotic PLND if an extended template is followed. PSM rates appear similar with RARC and ORC. Conclusive long-term survival outcomes for RARC are limited, although oncologic outcomes of ≤5 yr are similar to those reported for ORC. Initial functional outcomes appear favorable; however, additional research on continence and potency after RARC is needed.
Author contributions: Bertram Yuh had full access to all the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.
Study concept and design: Yuh, Wilson, Bochner, Montorsi, Chan, Thalmann, Palou, Stenzl, Guru, Catto, Novara, Wiklund.
Acquisition of data: Yuh, Wilson, Bochner, Montorsi, Chan, Thalmann, Palou, Stenzl, Guru, Catto, Novara, Wiklund.
Analysis and interpretation of data: Yuh, Novara.
Drafting of the manuscript: Yuh, Wilson, Bochner, Montorsi, Chan, Thalmann, Palou, Stenzl, Guru, Catto, Novara, Wiklund.
Critical revision of the manuscript for important intellectual content: Yuh, Wilson, Bochner, Montorsi, Chan, Thalmann, Palou, Stenzl, Guru, Catto, Novara, Wiklund.
Statistical analysis: Yuh, Novara.
Obtaining funding: Wilson.
Administrative, technical, or material support: Yuh, Wilson, Bochner, Montorsi, Chan, Thalmann, Palou, Stenzl, Guru, Catto, Novara, Wiklund.
Supervision: Yuh, Wilson, Bochner, Montorsi, Chan, Thalmann, Palou, Stenzl, Guru, Catto, Novara, Wiklund.
Other(specify): None.
Financial disclosures: Bertram Yuh certifies that all conflicts of interest, including specific financial interests and relationships and affiliations relevant to the subject matter or materials discussed in the manuscript (eg, employment/affiliation, grants or funding, consultancies, honoraria, stock ownership or options, expert testimony, royalties, or patents filed, received, or pending), are the following: Dr. Wiklund has received proctoring assignments and a research grant from Intuitive Surgical, Sunnyvale, CA, USA; Dr. Wilson has been a consultant and a speaker for Intuitive Surgical, Sunnyvale, CA, USA.
Funding/Support and role of the sponsor: None.
Radical cystectomy and pelvic lymph node dissection (PLND) is the gold standard treatment for muscle-invasive bladder cancer (MIBC) and high-risk non–muscle-invasive disease [1] . Patients undergoing this operation can experience 66% recurrence-free survival at 10 yr after surgery [2] . The addition of neoadjuvant platinum-based chemotherapy has been shown to improve overall survival (OS) rates by approximately 5% [3] . Robot-assisted radical cystectomy (RARC) was initially described by Menon et al in 2003 [4] . Over time, many international centers have adopted RARC.
Oncologic outcomes from large population-based cohorts of RARC with lengthy follow-up are lacking. Early on in RARC history, surrogates for oncologic control were reported using positive surgical margin (PSM) rates and lymph node yields. More recently, 5-yr survival figures have become available. The majority of these outcomes, however, capture institutions early in their learning curves and incorporate patients potentially selected for the robotic technique, thus avoiding more advanced-stage or technically difficult cases. Data on functional consequences of RARC are even more limited; therefore, the quality of nerve sparing and its effect on potency recovery and continence are inadequately understood.
Because of the expanding evidence available in the field of RARC, and in preparation for the Pasadena international consensus meeting on best practice in RARC and urinary reconstruction, we performed a systematic literature review of perioperative, functional, and oncologic outcomes of RARC in comparison with open radical cystectomy (ORC) and laparoscopic radical cystectomy (LRC).
We report on the systematic review and cumulative analysis of oncologic and functional outcomes of RARC. We systematically examined lymph node yields, PSMs, cancer-specific survival (CSS), disease-free survival (DFS), and OS. In addition, functional outcomes after RARC, including urinary continence and erectile function, were systematically examined.
A systematic literature search was initially performed in September 2013 using the Medline, Scopus, and Web of Science databases. The searches included only a free-text protocol using the termsrobot-assisted radical cystectomyorda Vinci radical cystectomyorrobot* radical cystectomyin all the fields of the records for Medline and Scopus searches and in theTitleandTopicfields for the Web of Science search. No limits were applied. A full update of the searches was performed on April 28, 2014.
Two authors (G.N. and B.Y.) separately reviewed the records to select RARC case series and studies that compared RARC with ORC and RARC with LRC. Discrepancies were resolved by open discussion. Other significant studies cited in the reference lists of the selected papers were evaluated, as were studies published after the systematic search.
All noncomparative studies reporting the following data on RARC were collected: intraoperative and perioperative data (operative time, blood loss, transfusion rate, in-hospital stay, readmission, complication rates), functional data (urinary continence, erectile function), and oncologic data (PSMs, lymph node yield, DFS, CSS, OS). The present review included only studies reporting on functional and oncologic data.
Studies reporting on partial cystectomy, prostate-sparing cystectomy, salvage cystectomy, cystectomy for urachal cancer or benign disease, single-case reports, pure laparoscopic (or mixed) series, or laparoendoscopic single-site or natural orifice transluminal endoscopic surgery for radical cystectomy; experimental studies on animal models; congress abstracts; review papers; editorials; population-based studies; and book chapters were not included in the review. All data retrieved from the selected studies were recorded in an electronic database.
All papers were categorized according to the 2011 levels of evidence (LOEs) for therapy studies: LOE 1, systematic review of randomized trials or n-of-1 trials; LOE 2, randomized trial or observational study with dramatic effect; LOE 3, nonrandomized controlled cohort/follow-up study; LOE 4, case series, case–control study, or historically controlled study; or LOE 5, mechanism-based reasoning [5] . Papers were categorized according to the IDEAL recommendations [6] .
Cumulative analysis was conducted using Review Manager v5.2 software designed for composing Cochrane Reviews (Cochrane Collaboration, Oxford, UK). Statistical heterogeneity was tested using the chi-square test. Apvalue <0.10 was used to indicate heterogeneity. Where there was a lack of heterogeneity, fixed-effects models were used for the cumulative analysis. Random-effects models were used in case of heterogeneity. For continuous outcomes, the results were expressed as weighted mean differences and standard deviations (SDs); for dichotomous variables, results were given as odds ratios (ORs) and 95% confidence intervals (CIs). Because of limitations in the Review Manager v5.2 software, meta-analysis of continuous variables was possible only when rough data were presented as mean and SD. Authors of the papers were contacted to provide missing data, whenever necessary. For all statistical analyses, two-sidedp < 0.05 was considered statistically significant.
Figure 1 shows a flowchart of this systematic review of the literature.
In total, 65 surgical series[4], [7], [8], [9], [10], [11], [12], [13], [14], [15], [16], [17], [18], [19], [20], [21], [22], [23], [24], [25], [26], [27], [28], [29], [30], [31], [32], [33], [34], [35], [36], [37], [38], [39], [40], [41], [42], [43], [44], [45], [46], [47], [48], [49], [50], [51], [52], [53], [54], [55], [56], [57], [58], [59], [60], [61], [62], [63], [64], [65], [66], [67], [68], [69], and [70]and 22 comparative studies[71], [72], [73], [74], [75], [76], [77], [78], [79], [80], [81], [82], [83], [84], [85], [86], [87], [88], [89], [90], [91], and [92]reported on pathologic, oncologic (n = 18), or functional (n = 9) outcomes of RARC.
Most surgical series were retrospective, single-center studies (LOE 4). Exceptions included prospective studies[8], [22], [28], [35], [36], [43], [52], [59], [64], [66], [81], and [87]and some multi-institutional retrospective collaboration studies[16], [23], [25], [30], [45], [48], [54], [61], and [68]. Only two of the comparative studies were randomized[74] and [83](LOE 2b); all other comparative studies were nonrandomized, whether prospective or retrospective (LOE 4).
Table 1 summarizes the number of lymph nodes recovered in published RARC series. The majority of studies (86%) reported extent of lymph node dissection (LND), with more centers performing extended LND (ELND) in recent series. Standard LND typically involved the removal of obturator, internal iliac, external iliac, and some portion of the common iliac lymph nodes bilaterally. ELND templates typically brought the proximal extent up to the aortic bifurcation or inferior mesenteric artery. Approximately half of the analyzed studies reported following an extended template of dissection.
Table 1 Lymph node yields in robot-assisted radical cystectomy series
| Reference | Institution | IDEAL stage | Cases, no. | Study design | Extension of LND | Operative time, min | Retrieved nodes, no. | pN+, % | Metastatic nodes, median, no. | Complications due to LND |
|---|---|---|---|---|---|---|---|---|---|---|
| Menon et al, 2003 [4] | Henry Ford Hospital | 1 | 17 | Retrospective | Standard | – | – | 6 | – | – |
| Menon et al, 2004 [7] | Henry Ford Hospital | 1 | 3 female | Retrospective | Standard | – | 12 | 0 | – | – |
| Guru et al, 2007 [8] | Roswell Park Cancer Institute | 1 | 20 | Prospective | Standard | 44 | 13 | 15 | 1 | – |
| Mottrie et al, 2007 [9] | O.L.V.–Clinic | 2a | 27 | Retrospective | Extended | – | 23 | 9 | – | – |
| Pruthi et al, 2008 [71] | UNC | 2a | 20 | Retrospective | Standard | – | 19 | 10 | – | – |
| Hemal et al, 2008 [10] | All India Institute of Medical Sciences | 1 | 6 | Retrospective | Standard | – | 12 | 17 | – | – |
| Lowentritt et al, 2008 [11] | Tulane University | 2a | 4 | Retrospective | Standard | – | 12 | 25 | – | – |
| Murphy et al, 2008 [12] | Guy's Hospital | 2a | 23 | Retrospective | Standard | – | 16 | 9 | – | – |
| Park et al, 2008 [13] | Yonsei | 2a | 4 | Retrospective | Standard | – | 17 | 0 | – | – |
| Pruthi et al, 2008 [14] | UNC | 2a | 12 female | Retrospective | Standard, then extended | – | 19 | 17 | – | – |
| Pruthi et al, 2008 [15] | UNC | 2b | 50 | Retrospective | Standard | – | 19 | 20 | – | – |
| Wang et al, 2008 [72] | Cornell | 2b | 33 | Retrospective | Standard | – | 17 | 19 | – | – |
| Woods et al, 2008 [16] | Mayo Arizona Tulane University |
2b | 27 | Multi-institutional | Extended | – | 12.3 | 33 | 3.1 | 0 |
| Yuh et al, 2008 [17] | Roswell Park Cancer Institute | 2a | 54 | Retrospective | Extended | – | 17 | – | – | – |
| Gamboa et al, 2009 [18] | University of California, Irvine |
2a | 41 | Retrospective | Standard | – | 23 | 14 | 4 | – |
| Pruthi et al, 2009 [19] | UNC | 2b | 50 | Retrospective | Standard, then extended | – | 19 | 16 | – | – |
| 10 female | 19 | |||||||||
| 40 male | 18 | |||||||||
| Palou Redorta et al, 2009 [20] | Barcelona Autonomous University | 2a | 9 | Retrospective | Extended | 60 | 10 | 0 | – | – |
| Yuh et al, 2009 [21] | Roswell Park Cancer Institute | 2b | 73 | Retrospective | Extended | – | 19 | – | – | – |
| Guru et al, 2010 [22] | Roswell Park Cancer Institute | 2a | 26 | Prospective | Extended | – | 21 | 29 | 1 | Internal iliac artery injury: 1 |
| Hellenthal et al, 2011 [23] | IRCC | 2b | 437 | Multi-institutional | – | – | 17 | 20 | – | – |
| Josephson et al, 2010 [24] | City of Hope Cancer Center | 2b | 58 | Retrospective | Extended | – | 27 | 24 | – | – |
| Kang et al, 2010 [25] | Multicenter | 2b | 71 standard LND 33 extended LND |
Retrospective | Standard Extended |
– | 15.7 24.7 |
10 | – | – |
| Kasraeian et al, 2010 [26] | Montsouris Institute | 2a | 9 | Retrospective | Extended | – | 11 | 22 | – | 0 |
| Kauffman et al, 2011 [27] | Cornell | 2b | 85 | Retrospective | Extended | – | 19 | 15 | – | – |
| Kwon et al, 2010 [28] | Kyungpook National University | 2a | 17 | Prospective | Standard | – | 6 | 6 | 1 | 0 |
| Lavery et al, 2011 [29] | Ohio State University | 2a | 15 | Retrospective | Extended | 107 | 41.8 | 20 | – | 0 |
| Martin et al, 2010 [30] | Mayo Arizona Tulane University |
2b | 59 | Multi-institutional | Extended | – | – | 34 | – | – |
| Ng et al, 2010 [73] | Cornell | 2b | 83 | Retrospective | Standard | – | 16 | 16 | – | – |
| Nix et al, 2010 [74] | UNC | 3 | 21 | RCT | Standard | – | 19 | 19 | – | – |
| Pruthi et al, 2010 [31] | UNC | 2b | 100 | Retrospective | Standard, then extended | – | 19 | 20 | – | – |
| Richards et al, 2010 [75] | Wake Forest University | 2b | 35 | Retrospective | Extended | – | 16 | 29 | – | – |
| Akbulut et al, 2011 [32] | Ankara Ataturk Training and Research Hospital | 2a | 12 | Not reported | Extended | – | 21.3 | 42 | – | 8 |
| Canda et al, 2012 [33] | Ankara Ataturk Training and Research Hospital | 2a | 27 | Not reported | Extended | – | 24.8 | 22 | – | – |
| Davis et al, 2011 [34] | University of Texas M.D. Anderson Cancer Center | 2a | 11 | Retrospective | Extended | 117 | 43 | 9 | 1 | – |
| Jonsson et al, 2011 [35] | Karolinska Institute | 2b | 45 36 neobladder 9 ileal conduit |
Prospective | Standard Extended |
– | 19 19 27 |
20 17 33 |
– | – |
| Khan et al, 2011 [36] | Guy's Hospital | 2a | 50 | Prospective | – | – | 17 | – | – | – |
| Manoharan et al, 2011 [37] | University of Miami | 2a | 14 | Retrospective | Standard | – | 12 | – | – | – |
| Martin et al, 2011 [76] | Mayo Arizona | 2b | 19 | Retrospective | – | – | 16 | – | – | – |
| Richards et al, 2011 [38] | Wake Forest University | 2b | 60 | Retrospective | Extended | – | 17 | 30 | – | Lymphocele: 1 |
| Schumacher et al, 2011 [39] | Karolinska Institute | 2b | 45 | Retrospective | Standard 49%, extended 31% | – | 22.5 | – | 1.5 | Lymphocele: 2 |
| Shah et al, 2011 [40] | Ohio State University | 2b | 30 | Retrospective | Extended | – | – | 30 | – | – |
| Torrey et al, 2011 [41] | City of Hope Cancer Center | 2b | 34 | Retrospective | Extended | – | 28.9 | – | – | – |
| Cho et al, 2012 [42] | Hallym University College of Medicine | 2b | 35 | Retrospective | Standard | – | – | 6 | – | – |
| Goh et al, 2012 [43] | Keck School of Medicine, University of Southern California, Los Angeles | 2a | 15 | Prospective | Superextended | – | 55 | 26 | – | – |
| Lau et al, 2012 [44] | City of Hope Cancer Center | 2b | 23 (aged >80 yr) | Retrospective | Extended | – | 20.4 | 22 | – | – |
| Mmeje et al, 2013 [45] | Mayo Arizona UNC |
2b | 50 | Multi-institutional | Extended | – | 18 | 100 | 3 | – |
| Poch et al, 2012 [46] | Roswell Park Cancer Institute | 2b | 56 | Retrospective | – | – | 25 | 16 | – | – |
| Richards et al, 2012 [77] | Wake Forest University | 2b | 20 (aged >75 yr) | Retrospective | Extended | – | 17 | 35 | – | – |
| Saar et al, 2013 [47] | Saarland University | 2b | 62 | Retrospective | – | – | 14.2 | 21 | – | – |
| Smith et al, 2012 [48] | Mayo Arizona UNC, Tulane University |
2b | 227 | Multi-institutional | – | – | 18 | 20 | – | – |
| Styn et al, 2012 [78] | University of Michigan | 2b | 50 | Retrospective | – | – | 14.3 | 12 | – | – |
| Sung et al, 2012 [79] | Samsung Medical Center | 2b | 35 | Retrospective | Standard | – | 19.1 | 26 | – | Lymphocele: 1 |
| Treiyer et al, 2012 [49] | Saarland University | 2b | 91 | Retrospective | Standard | – | 14.5 | 14 | – | – |
| Tsui et al, 2012 [50] | Chang Gung Memorial Taiwan | 2a | 8 | Retrospective | Standard | – | 3 | 12.50 | – | – |
| Yuh et al, 2012 [51] | City of Hope Cancer Center | 2b | 196 | Retrospective | Extended | – | 28 | 22 | – | Lymphocele: 3 |
| Collins et al, 2013 [52] | Karolinska Institute | 2b | 113 | Prospective | Extended 56%, standard 34%, limited 5%, none 5% | – | 21 | 20 | – | Lymphocele: 5 |
| Johar et al, 2013 [53] | IRCC | 2b | 939 | Multi-institutional | – | – | 18.1 | 26 | – | – |
| Maes et al, 2013 [80] | Metro Health Hospital | 2b | 14 | Retrospective | Extended | – | 11.9 | 7 | – | – |
| Marshall et al, 2013 [54] | IRCC | 2b | 765 | Multi-institutional | Extended 58%, standard 40%, no LND 2% | – | 18 | 27 | – | – |
| Musch et al, 2014 [81] | Klinikin Essen–Mitte | 2b | 100 | Prospective | – | – | 26.5 | 20 | – | Lymphocele: 4 |
| Nazmy et al, 2014 [55] | City of Hope Cancer Center | 2b | 209 | Retrospective | Extended | – | – | 22 | – | Lymphocele: 3 |
| Nepple et al, 2013 [82] | Washington University | 2b | 36 | Retrospective | Standard | – | 17 | 22 | – | – |
| Parekh et al, 2013 [83] | University of Texas Health Sciences Center at San Antonio | 3 | 20 | RCT | Standard | – | 11 | 20 | – | – |
| Tyritzis et al, 2013 [56] | Karolinska Institute | 2b | 70 | Retrospective | Standard 43% Extended 48% |
– | 21 | 14 | – | Lymphocele: 6 Lymphedema: 1 |
| Xylinas et al, 2013 [57] | Cornell | 2b | 175 | Retrospective | Standard | – | 19 | 17 | – | Lymphocele: 2 |
| Phillips et al, 2014 [69] | Seward St. Elizabeth Medical Center | 2b | 23 (>80 yr) | Retrospective | Extended | – | 19 | – | – | |
| Raza et al, 2014 [70] | Roswell Park Cancer Institute | 2b | 99 | Retrospective | – | – | 20.7 | 36 | – | – |
| Yuh et al, 2014 [58] | City of Hope Cancer Center | 2b | 162 | Retrospective | Extended | – | 28 | 23 | – | – |
| Total | 19.3 | 23 |
IRCC = International Robotic Cystectomy Consortium; LND = lymph node dissection; RCT = randomized controlled trial; UNC = University of North Carolina.
The lymph node yield from all series was 19 (range: 3–55). Initial descriptions using a standard template of dissection achieved yields of 18 lymph nodes [59] . Number of lymph nodes recovered with an ELND ranged from 11 to 55. Abaza et al adopted a robotic template similar to the open technique, including external iliac, obturator, hypogastric, common iliac, and presacral up to the aortic bifurcation; the mean lymph node yield was 37.5 (SD: 13.2), demonstrating that lymph node counts could mirror those of open dissection if the same template was followed [88] . In a study of open completion LND after robot-assisted ELND in 11 men, Davis et al removed only an additional 4 lymph nodes with an open approach after 43 were removed with robot assistance [34] . Time of LND was rarely reported, although it ranged from 44 min in standard LND to 117 min in ELND[8] and [34]. The lymph node–positive rate was 22%. In series with >20 RARCs, lymph node–positive rates ranged from 6% to 42%. Reports of vascular injuries were rare, and lymphocele rates were 0–9%.
Table 2 summarizes the studies assessing the effects of patient characteristics and particular surgical aspects on lymph node yields in RARC series. Cumulative analysis from the International Robotic Cystectomy Consortium (IRCC) with respect to lymphadenectomy in 437 patients found a median of 17 lymph nodes removed, with a 20% node-positivity rate [23] . Patient age and sex did not affect the performance of lymphadenectomy. In a different series, increasing body mass index (BMI) did not appear to negatively affect lymph node yield, with >20 lymph nodes removed in normal, overweight, and obese patients [46] .
Table 2 Impact of patient characteristics and surgical aspects on lymph node yield in robot-assisted radical cystectomy series
| Reference | Institution | IDEAL stage |
Cases | Study design | Extension of LND | Operative time, min | Retrieved nodes, no. | pN+, % |
|---|---|---|---|---|---|---|---|---|
| Patient BMI | ||||||||
| Poch et al, 2012 [46] | Roswell Park Cancer Institute | 2b | 56 | Retrospective | – | – | 25 | 16 |
| BMI <25: 14 | 22 | 7 | ||||||
| BMI 25 to <30: 21 | 23 | 14 | ||||||
| BMI ≥30: 21 | 20 | 24 | ||||||
| Case volume | ||||||||
| Guru et al, 2008 [59] | Roswell Park Cancer Institute | 2a | 1–12 | Prospective | Extended | 46 | 33% >13 | 8 |
| 13–24 | 44 | 66% >13 | 33 | |||||
| 25–36 | 41 | 83% >13 | 25 | |||||
| 37–47 | 43 | 72% >13 | 64 | |||||
| 48–58 | 56 | 91% >13 | 18 | |||||
| Pruthi et al, 2008 [60] | University of North Carolina | 2b | 50 | Retrospective | Standard | – | 19 | – |
| 1–10 | 21 | |||||||
| 11–20 | 19 | |||||||
| 21–30 | 20 | |||||||
| 31–40 | 17 | |||||||
| 41–50 | 20 | |||||||
| Richards et al, 2011 [38] | Wake Forest University | 2b | 60 | Retrospective | Extended | – | 17 | 30 |
| 1–20 | 17 | |||||||
| 21–40 | 19.1 | |||||||
| 41–60 | 14.4 | |||||||
| Schumacher et al, 2011 [39] | Karolinska Institute | 2b | 45 | Retrospective | Standard 49% | – | 22.5 | – |
| Extended 31% | ||||||||
| 1–15 | Standard 40% | |||||||
| Extended 7% | ||||||||
| 16–30 | Standard 47% | |||||||
| Extended 53% | ||||||||
| 31–45 | Standard 60% | |||||||
| Extended 33% | ||||||||
| Prior RARP experience | ||||||||
| Hayn et al, 2010 [61] | IRCC | 2b | 496 | Retrospective | – | – | 17.8 | – |
| ≤50 previous RARP: 83 | 13.7 | |||||||
| 51–100 previous RARP: 187 | 19.8 | |||||||
| 101–150 previous RARP: 176 | 19.6 | |||||||
| >150 previous RARP: 50 | 11.8 * | |||||||
* Statistically significant.
BMI = body mass index; RCC = International Robotic Cystectomy Consortium; LND = lymph node dissection; RARP = robot-assisted radical prostatectomy.
It is interesting to note that in single-institution series, Richards et al [38] , Schumacher et al [39] , Guru et al [59] , and Pruthi et al [60] did not find higher lymph node yields with increasing sequential case numbers. However, in the IRCC, performance of lymphadenectomy was positively correlated with surgeon and institution volume but was reduced in patients with more advanced disease (pT4 stage), which may reflect operative avoidance of bulky nodal tissue.
Table 3 summarizes the occurrence of PSMs reported in the RARC series. The reported PSM rates were 5.6% (range: 0–26%). However, in series of >100 patients, margin rates ranged between 4% and 9%[48] and [53]. PSMs were reported in 1–1.5% of patients with pT2 disease and 0–25% of patients with pT3 and higher disease. PSM rates from the IRCC in 939 cases were 9% [53] .
Table 3 Positive surgical margins in robot-assisted radical cystectomy series
| Reference | Institution | IDEAL stage | Cases, no. | Study design | Pathologic stage, % | Overall PSM rate, % | PSM location | PSM rate, % | ||
|---|---|---|---|---|---|---|---|---|---|---|
| ≤pT2 | ≥pT3 | ≤pT2 | ≥pT3 | |||||||
| Menon et al, 2003 [4] | Henry Ford Hospital | 1 | 17 | Retrospective | – | – | 0 | – | 0 | 0 |
| Yohannes et al, 2003 [62] | Creighton University | 1 | 2 | Retrospective | 0 | 100 | 0 | – | 0 | 0 |
| Menon et al, 2004 [7] | Henry Ford Hospital | 1 | 3 female | Retrospective | 66 | 33 | 0 | – | 0 | 0 |
| Rhee et al, 2006 [84] | University of Virginia | 1 | 7 | Retrospective | 43 | 57 | 0 | – | 0 | 0 |
| Guru et al, 2007 [8] | Roswell Park Cancer Institute | 1 | 20 | Prospective | 40 | 60 | 15 | Prostate: 1 Ureter: 1 Vagina: 1 |
0 | 25 |
| Mottrie et al, 2007 [9] | O.L.V.–Clinic | 2a | 27 | Retrospective | 78 | 22 | 4 | Ureter: 1 | – | – |
| Pruthi et al, 2008 [71] | UNC | 2a | 20 | Retrospective | 70 | 20 | 0 | – | 0 | 0 |
| Hemal et al, 2008 [10] | All India Institute of Medical Sciences | 1 | 6 | Retrospective | 67 | 33 | 0 | – | 0 | 0 |
| Lowentritt et al, 2008 [11] | Tulane University | 2a | 4 | Retrospective | 25 | 75 | 0 | – | 0 | 0 |
| Murphy et al, 2008 [12] | Guy's Hospital | 2a | 23 | Retrospective | 74 | 17 | 0 | – | 0 | 0 |
| Park et al, 2008 [13] | Yonsei | 2a | 4 | Retrospective | 50 | 50 | 0 | – | 0 | 0 |
| Pruthi et al, 2008 [14] | UNC | 2a | 12 female | Retrospective | 58 | 25 | 0 | – | 0 | 0 |
| Pruthi et al, 2008 [15] | UNC | 2b | 50 | Retrospective | 66 | 14 | 0 | – | 0 | 0 |
| Pruthi et al, 2009 [19] | UNC | 2b | 50 10 female 40 male |
Retrospective | 66 50 70 |
18 30 15 |
0 | – | 0 | 0 |
| Wang et al, 2008 [72] | Cornell | 2b | 33 | Retrospective | 72 | 28 | 6 | Perivesical fat: 2 | 0 | 22 |
| Woods et al, 2008 [16] | Mayo Arizona Tulane University |
2b | 27 | Multi-institutional | – | – | 7 | – | 0 | – |
| Yuh et al, 2008 [17] | Roswell Park Cancer Institute | 2a | 54 | Retrospective | 44 | 56 | 13 | – | 0 | 23 |
| Gamboa et al, 2009 [18] | University of California, Irvine | 2a | 41 | Retrospective | – | – | 5 | – | 0 | – |
| Palou Redorta et al, 2009 [20] | Barcelona Autonomous University | 2a | 9 | Retrospective | 66 | 33 | 11 | – | – | – |
| Yuh et al, 2009 [21] | Roswell Park Cancer Institute | 2b | 73 | Retrospective | 45 | 55 | 10 | – | 0 | 18 |
| Guru et al, 2010 [22] | Roswell Park Cancer Institute | 2a | 20 | Prospective | 62 | 38 | 4 | – | 0 | 9 |
| Hayn et al, 2010 [61] | IRCC | 2b | 482 | Multi-institutional | 64 | 36 | 7 | – | – | – |
| Hellenthal et al, 2010 [68] | IRCC | 2b | 513 | Multi-institutional | 64 | 36 | 7 | – | 1.50 | 17 |
| Kang et al, 2010 [25] | Multicenter | 2b | 104 | Multi-institutional | 70 | 30 | 5 | – | – | – |
| Kasraeian et al, 2010 [26] | Montsouris Institute | 2a | 9 | Retrospective | 44 | 66 | 0 | – | – | – |
| Kauffman et al, 2011 [27] | Cornell | 2b | 85 | Retrospective | 64 | 36 | 6 | – | 0 | 16 |
| Kwon et al, 2010 [28] | Kyungpook National University | 2a | 17 | Prospective | 59 | 41 | 0 | – | 0 | 0 |
| Martin et al, 2010 [30] | Mayo Arizona Tulane University |
2b | 59 | Multi-institutional | 47 | 53 | – | – | – | – |
| Ng et al, 2010 [73] | Cornell | 2b | 83 | Retrospective | 61 | 39 | 7 | – | 0 | 19 |
| Nix et al, 2010 [74] | UNC | 3 | 21 | RCT | 67 | 14 | 0 | – | 0 | 0 |
| Pruthi et al, 2010 [31] | UNC | 2b | 100 | Retrospective | 67 | 13 | 0 | – | 0 | 0 |
| Richards et al, 2010 [75] | Wake Forest University | 2b | 35 | Retrospective | 60 | 40 | 3 | – | – | – |
| Akbulut et al, 2011 [32] | Ankara Ataturk Training and Research Hospital | 2a | 12 | Not reported | 58 | 42 | 0 | – | 0 | 0 |
| Canda et al, 2012 [33] | Ankara Ataturk Rraining and Research Hospital | 2a | 27 | Not reported | 56 | 44 | 4 | – | 0 | 4 |
| Davis et al, 2011 [34] | University of Texas M.D. Anderson Cancer Center | 2a | 11 | Retrospective | 92 | 8 | 0 | – | – | – |
| Jonsson et al, 2011 [35] | Karolinska Institute | 2b | 45 | Prospective | 78 | 22 | 2 | – | 0 | 10 |
| Khan et al, 2011 [36] | Guy's Hospital | 2a | 50 | Prospective | 72 | 28 | 2 | – | 0 | 7 |
| Manoharan et al, 2011 [37] | University of Miami | 2a | 14 | Retrospective | – | – | 0 | – | 0 | 0 |
| Martin et al, 2011 [76] | Mayo Arizona | 2b | 19 | Retrospective | 42 | 58 | – | – | – | – |
| Richards et al, 2011 [38] | Wake Forest University | 2b | 60 | Retrospective | 63 | 37 | 10 | – | – | – |
| Schumacher et al, 2011 [39] | Karolinska Institute | 2b | 45 | Retrospective | 78 | 22 | 2 | Ureter: 1 | 0 | 10 |
| Shah et al, 2011 [40] | Ohio State University | 2b | 30 | Retrospective | 65 | 35 | 7 | – | 0 | 22 |
| Cho et al, 2012 [42] | Hallym University College of Medicine | 2b | 35 | Retrospective | 86 | 14 | 3 | – | – | – |
| Goh et al, 2012 [43] | Keck School of Medicine, University of Southern California, Los Angeles | 2a | 15 | Prospective | 67 | 33 | 0 | – | – | – |
| Lau et al, 2012 [44] | City of Hope Cancer Center | 2b | 23 (aged >80 yr) | Retrospective | 61 | 39 | 13 | Ureter: 1 | – | – |
| Mmeje et al, 2013 [45] | Mayo Arizona | 2b | 50 | Multi-institutional | 34 | 66 | 2 | – | – | – |
| UNC | ||||||||||
| Poch et al, 2012 [46] | Roswell Park Cancer Institute | 2b | 56 | Retrospective | 55 | 45 | – | – | – | – |
| Richards et al, 2012 [77] | Wake Forest University | 2b | 20 (aged >75 yr) | Retrospective | 60 | 40 | 5 | – | – | – |
| Saar et al, 2013 [47] | Saarland University | 2b | 62 | Retrospective | 64 | 36 | 2 | – | – | – |
| Smith et al, 2012 [48] | Mayo Arizona | 2b | 227 | Multi-institutional | – | – | 2 | – | – | – |
| UNC, Tulane University | ||||||||||
| Styn et al, 2012 [78] | University of Michigan | 2b | 50 | Retrospective | 60 | 40 | 2 | – | – | – |
| Sung et al, 2012 [79] | Samsung Medical Center | 2b | 35 | Retrospective | 43 | 57 | – | – | – | – |
| Treiyer et al, 2012 [49] | Saarland University | 2b | 91 | Retrospective | 67 | 33 | 2 | Urethra: 1 Prostate: 1 | – | – |
| Tsui et al, 2012 [50] | Chang Gung Memorial | 2a | 8 | Retrospective | 75 | 25 | 0 | – | 0 | 0 |
| Yuh et al, 2012 [51] | City of Hope Cancer Center | 2b | 196 | Retrospective | 64 | 36 | 4 | – | – | – |
| Azzouni et al, 2013 [63] | Roswell Park Cancer Institute | 2b | 100 | Retrospective | 35 | 65 | 4 | – | – | – |
| Collins et al, 2013 [52] | Karolinska Institute | 2b | 113 | Prospective | 75 | 25 | 5 | Ureter: 1 | 1 | 18 |
| Johar et al, 2013 [53] | Multicenter | 2b | 939 | Retrospective | 49 | 51 | 9 | – | – | – |
| Maes et al, 2013 [80] | Metro Health Hospital | 2b | 14 | Retrospective | 43 | 57 | 21 | – | – | – |
| Marshall et al, 2013 [54] | IRCC | 2b | 765 | Multi-institutional | 59 | 41 | – | – | – | – |
| Musch et al, 2014 [81] | Klinikin Essen–Mitte | 2b | 100 | Prospective | 61 | 39 | 2 | – | – | – |
| Nazmy et al, 2014 [55] | City of Hope Cancer Center | 2b | 209 | Retrospective | 65 | 35 | 3 | – | – | – |
| Nepple et al, 2013 [82] | Washington University | 2b | 36 | Retrospective | 53 | 47 | 6 | – | 0 | 12 |
| Parekh et al, 2013 [83] | University of Texas Health Sciences Center at San Antonio | 3 | 20 | RCT | 50 | 50 | 5 | – | 0 | 10 |
| Tyritzis et al, 2013 [56] | Karolinska Institute | 2b | 70 | Retrospective | 86 | 14 | 1.5 | Ureter: 1 | 0 | 10 |
| Xylinas et al, 2013 [57] | Cornell | 2b | 175 | Retrospective | 65 | 35 | 5 | – | – | – |
| Phillips et al, 2014 [69] | Seward St. Elizabeth Medical Center | 2b | 23 (aged >80 yr) | Retrospective | 30 | 70 | 26 | – | – | – |
| Raza et al, 2014 [70] | Roswell Park Cancer Institute | 2b | 99 | Retrospective | 48 | 52 | 8 | – | – | |
| Yuh et al, 2014 [58] | City of Hope Cancer Center | 2b | 162 | Retrospective | 67 | 33 | 4 | – | – | – |
| Total | 60 | 40 | 5.6 | |||||||
IRCC = International Robotic Cystectomy Consortium; PSM = positive surgical margin; RCT = randomized controlled trial; UNC = University of North Carolina.
Table 4 summarizes the studies assessing the effects of patient characteristics and particular surgical aspects on PSM rates in RARC series. Notably, Richards et al [38] , Schumacher et al [39] , and the IRCC [68] did not demonstrate decreasing surgical margin rates with sequential case number. In a study of the role of previous robot-assisted radical prostatectomy (RARP) experience on RARC outcomes, there was a trend toward increased positive margins with increasing RARP volumes, but it did not reach statistical significance (p = 0.089) [61] . The authors chiefly attributed this situation to the performance of RARC on patients with higher risk (higher than T3) disease. One study reported that PSMs occurred only in the overweight or obese patients, although pT4 rates were much higher in those patients (26% vs 7%) [65] .
Table 4 Predictors of positive surgical margins in robot-assisted radical cystectomy series
| Reference | Institution | IDEAL stage | Cases | Study design | Pathologic stage, % | Overall PSM rate, % | PSM rate, % | ||
|---|---|---|---|---|---|---|---|---|---|
| ≤pT2 | ≥pT3 | ≤pT2 | ≥pT3 | ||||||
| Case volume | |||||||||
| Guru et al, 2008 [59] | Roswell Park Cancer Institute | 2a | 1–12 | Prospective | 33 | 66 | 17 | – | – |
| 13–24 | 58 | 42 | 25 | ||||||
| 25–36 | 50 | 50 | 0 | ||||||
| 37–47 | 46 | 54 | 9 | ||||||
| 48–58 | 64 | 36 | 0 | ||||||
| Hayn et al, 2011 [64] | Roswell Park Cancer Institute | 2a | 1–50 | Prospective | 51 | 49 | 8 | – | – |
| 51–100 | 12 | ||||||||
| 101–164 | 6 | ||||||||
| Richards et al, 2011 [38] | Wake Forest University | 2b | 60 | Retrospective | 63 | 37 | 10 | – | – |
| 1–20 | 55 | 45 | 5 | ||||||
| 21–40 | 70 | 30 | 5 | ||||||
| 41–60 | 65 | 35 | 20 | ||||||
| Schumacher et al, 2011 [39] | Karolinska Institute | 2b | 45 | Retrospective | 78 | 22 | 2 | 0 | 10 |
| 1–15 | 87 | 13 | 0 | ||||||
| 16–30 | 67 | 33 | 7 | ||||||
| 31–45 | 80 | 20 | 0 | ||||||
| Azzouni et al, 2013 [63] | Roswell Park Cancer Institute | 2b | 100 | Retrospective | 35 | 65 | 4 | – | – |
| 1–25 | 36 | 64 | 4 | ||||||
| 26–50 | 40 | 60 | 4 | ||||||
| 51–75 | 44 | 56 | 4 | ||||||
| 76–100 | 20 | 80 | 4 | ||||||
| Previous RARP experience | |||||||||
| Hayn et al, 2010 [61] | IRCC | 2b | 482 | Retrospective | 64 | 36 | 7 | – | – |
| ≤50 previous RARP: 83 | 68 | 32 | 4 | ||||||
| 51–100 previous RARP: 173 | 76 | 24 | 5 | ||||||
| 101–150 previous RARP: 168 | 54 | 46 | 9.5 | ||||||
| >150 previous RARP: 48 | 42 | 58 | 12.5 | ||||||
| Patient BMI | |||||||||
| Butt et al, 2008 [65] | Roswell Park Cancer Institute | 2a | BMI <25: 14 | Retrospective | 64 | 36 | 0 | 0 | – |
| BMI 25–29: 18 | 28 | 72 | 28 | ||||||
| BMI ≥30: 17 | 42 | 58 | 6 | ||||||
| Poch et al, 2012 [46] | Roswell Park Cancer Institute | 2b | 56 | Retrospective | 55 | 45 | – | – | – |
| BMI <25: 14 | 50 | 50 | |||||||
| BMI 25 to <30: 21 | 57 | 43 | |||||||
| BMI ≥30: 21 | 52 | 48 | |||||||
| Intracorporeal vs extracorporeal diversion | |||||||||
| Kang et al, 2012 [85] | Korea University School of Medicine | 2a | 38 extracorporeal diversion | Retrospective | 76 | 24 | 2.5 | – | – |
| 4 intracorporeal diversion | 100 | 0 | 0 | ||||||
BMI = body mass index; IRCC = International Robotic Cystectomy Consortium; PSM = positive surgical margin; RARP = robot-assisted radical prostatectomy.
Table 5 summarizes the oncologic outcomes of current RARC publications. Neoadjuvant chemotherapy use was reported in 0–31% of patients. Adjuvant chemotherapy use was reported in 4–29% of patients.
Table 5 Survival outcomes in robot-assisted radical cystectomy series
| Reference | Institution | IDEAL stage | Cases, no. | Study design | Follow-up, mo | Neoadjuvant chemotherapy, % | Adjuvant chemotherapy, % | DFS estimates, % | CSS estimates, % | OS estimates, % | ||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 1 yr | 3 yr | 5 yr | 1 yr | 3 yr | 5 yr | 1 yr | 3 yr | 5 yr | ||||||||
| Pruthi et al, 2008 [15] | UNC | 2b | 50 | Retrospective | 13.2 | 0 | 22 | – | – | – | 94 (13 mo) | – | – | 90 (13 mo) | – | – |
| Murphy et al, 2008 [12] | Guy's Hospital | 2a | 23 | Retrospective | 17 | 29 | – | 91 (17 mo) | – | – | – | – | – | – | – | – |
| Josephson et al, 2010 [24] | City of Hope Cancer Center | 2b | 58 | Retrospective | 12 | 22 | – | – | 76 (2 yr) | – | – | 76 (2 yr) | – | – | 54 (2 yr) | – |
| Kang et al, 2010 [25] | Multicenter | 2b | 104 | Retrospective | 12 | – | – | 96 | – | – | – | – | – | – | – | – |
| Kauffman et al, 2011 [27] | Cornell | 2b | 85 | Retrospective | 18 | 20 | 12 | 79 | 73 (2 yr) | 88 | 85 (2 yr) | 83 | 79 (2 yr) | |||
| Martin et al, 2010 [30] | Mayo Arizona | 2b | 59 | Multi-institutional | 21 | 17 | – | 82 | 71 | – | – | – | – | 82 | 72 | – |
| Tulane University | ||||||||||||||||
| Pruthi et al, 2010 [31] | UNC | 2b | 100 | Retrospective | 21.2 | 5 | 18 | – | – | – | 94 (21 mo) | – | – | 91 (21 mo) | – | – |
| Canda et al, 2012 [33] | Ankara Ataturk Training and Research Hospital | 2a | 27 | Not reported | 6 | – | 4 | 85 (6 mo) | – | – | 89 (6 mo) | – | – | 72 (6 mo) | – | – |
| Mmeje et al, 2013 [45] | Mayo Arizona | 2b | 50 | Multi-institutional | 41.5 | 12 | 46 | – | 43 | 39 | – | – | – | – | 55 | 45 |
| UNC | ||||||||||||||||
| Treiyer et al, 2012 [49] | Saarland University | 2b | 91 | Retrospective | 15 | 0 | – | – | – | – | 94 (15 mo) | – | – | 93 (15 mo) | – | – |
| Collins et al, 2013 [52] | Karolinska Institute | 2b | 113 | Prospective | 25 | 31 | – | – | – | – | – | 81 | 67 | – | 80 | 66 |
| Khan et al, 2013 [66] | Guy's & St. Thomas Hospital | 1 | 14 | Prospective | 84 | 28 | 14 | 50 | 75 | 64 | ||||||
| Nepple et al, 2013 [82] | Washington University | 2b | 36 | Retrospective | 12 | 6 | – | – | 67 (2 yr) | – | – | 75 (2 yr) | – | – | 68 (2 yr) | – |
| Snow-Lisy et al, 2014 [67] | Cleveland Clinic | 2b | 17 | Retrospective | 67 | – | – | – | – | – | – | – | 69 | – | – | 39 |
| Tyritzis et al, 2013 [56] | Karolinska Institute | 2b | 70 | Retrospective | 30.3 | 24 | – | – | 81 (2 yr) | – | – | 89 (2 yr) | – | – | 89 (2 yr) | – |
| Xylinas et al, 2013 [57] | Cornell | 2b | 175 | Retrospective | 37 | 19 | – | 67 | 63 | – | 68 | 66 | – | – | – | |
| Raza et al, 2014 [70] | Roswell Park Cancer Institute | 2b | 99 | Retrospective | 73.9 | 6 | 29 | – | – | 53 | – | – | 68 | – | – | 42 |
| Yuh et al, 2014 [58] | City of Hope Cancer Center | 2b | 162 | Retrospective | 52 | 23 | – | – | 76 | 74 | – | 83 | 80 | – | 61 | 54 |
CSS = cancer-specific survival; DFS = disease-free survival; OS = overall survival; UNC = University of North Carolina.
Several studies further analyzed the use of adjuvant chemotherapy after RARC. General indications for selecting patients for adjuvant chemotherapy included pathologic stage pT3–4 or node-positive disease. Pruthi et al described the use of adjuvant chemotherapy in 18 of 100 RARC patients, with mean time to chemotherapy initiation at approximately 7 wk, which was faster than the authors’ historical time to chemotherapy in open cystectomy of 10 wk [31] . In a randomized trial of RARC (n = 21) compared with ORC (n = 20), 7 wk was also the mean time to initiation of chemotherapy after RARC [74] . In one analysis of patients with node-positive disease at the time of RARC, 46% received adjuvant chemotherapy [45] .
Survival represents the gold standard with respect to evaluating effectiveness and risks of treatment; however, RARC reports with 5-yr outcomes have become available only recently. Data remain limited for assessing long-term outcomes, patterns of recurrence, and means for predicting survival. The role of adjuvant treatments after RARC is also poorly defined.
Series detailing cancer control outcomes had a mean follow-up between 6 and 84 mo ( Table 5 ), although only 6 of 18 series (33%) reported a mean follow-up >36 mo. At 1, 2, 3, and 5 yr, DFS was 79–96%, 67–81%, 67–76%, and 53–74%, respectively; CSS was 88–94%, 75–89%, 68–83%, and 66–80%, respectively; and OS was 82–90%, 54–89%, 61–80%, and 39–66%, respectively. In the series with longest follow-up, Khan et al described only 14 patients with ≥5 yr of follow-up, showing DFS of 50%, CSS of 75%, and OS of 64% [66] .
Several series reported on adverse oncologic outcomes associated with increased pathologic stage or lymph node involvement[58] and [93]. In a series of 162 patients with urothelial carcinoma, Yuh et al found that 5-yr survival was worse with higher pathologic stage or lymph node positivity (p < 0.01). Patients with a lymph node density of 1–10% (defined as number of positive nodes divided by number of total nodes) had DFS, CSS, and OS of 34%, 49%, and 31%, respectively, whereas patients with lymph node density >10% had further reduced survival of 30%, 38%, and 20%, respectively. Predictors of DFS were lymph node density, pathologic stage, and age-adjusted Charlson comorbidity index, whereas the same measures plus receipt of transfusion were predictive for OS [58] . Similarly, in an analysis of 99 patients with follow-up >5 yr, pathologic stage and lymph node positivity were independent predictors of DFS, CSS, and OS, whereas positive margin status and Charlson comorbidity index predicted worse OS and CSS [70] .
In series with median follow-up of >36 mo, rates of local recurrence without distant disease ranged between 0% (n = 15) and 9% (n = 99)[57], [58], [66], [67], and [70]. No port-site recurrences occurred in these series. Xylinas et al examined 175 patients with a median follow-up of 37 mo, showing recurrence of disease in 29%. Of these patients, 8 (5%) had local recurrence alone, 11 had local and distant metastases, and 32 had distant metastases alone [57] .
In an analysis of patients with positive lymph nodes (n = 50) at the time of PLND, median time to recurrence was 10 mo after RARC [45] . Estimated OS at 36 and 60 mo was 55% and 45%, respectively, with recurrence-free survival at 36 and 60 mo of 43% and 39%, respectively. Similarly, Tyritzis et al reported recurrence-free survival of 34% and OS of 63% after 24 mo in node-positive patients [56] .
Table 6 presents the RARC series reporting on continence outcomes. Although functional outcomes are a major area of study in patients undergoing RARP, a lack of data remains for evaluation after RARC. Worldwide, the number of patients evaluated for continence after orthotopic bladder substitution is <200 from nine reports at the present time. There are also widespread differences in patient selection, methods of data collection, and outcome assessment.
Table 6 Urinary continence rates in robot-assisted radical cystectomy series
| Reference | Institution | IDEAL stage | Cases, no. | Study design | Nerve-sparing surgery, % | Intracorporeal diversion, % | Follow-up, mo | Method of data collection | Continence definition | Continence rate, % | ||
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 3 mo | 6 mo | 12 mo | ||||||||||
| Mottrie et al, 2007 [9] | O.L.V.–Clinic | 2a | 27 | Retrospective | 29 | 0 | 10.2 | – | – | 86 | – | – |
| Murphy et al, 2008 [12] | Guy's Hospital | 2a | 23 | Retrospective | 20 | 0 | 17 | – | – | – | – | 100 D |
| 75 N (17 mo) |
||||||||||||
| Palou Redorta et al, 2009 [20] | Barcelona Autonomous University | 2a | 9 | Retrospective | 100 | 0 | 7 | – | – | 100 D and N (7 mo) |
||
| Canda et al, 2012 [33] | Ankara Ataturk Training and Research Hospital | 2a | 27 | Not reported | 89 | 100 | 6 | – | D: 0–1 safety pads | – | 48 | – |
| N: dry with no protection | 11 | |||||||||||
| Jonsson et al, 2011 [35] | Karolinska Institute | 2b | 36 | Prospective | 55 | 100 | 25 | – | 0–1 pads | – | – | 83 D 66 N |
| Manoharan et al, 2011 [37] | University of Miami | 2a | 14 | Retrospective | – | 0 | – | – | – | 93 D | ||
| 71 N | ||||||||||||
| Torrey et al, 2012 [41] | City of Hope Cancer Center | 2b | 34 | Retrospective | 0 | 0 (all Indiana pouch) | 12.1 | Physician charting | – | – | – | 97 |
| Goh et al, 2012 [43] | Keck School of Medicine, University of Southern California, Los Angeles | 2a | 15 | Prospective | – | 100 | 3 | – | – | 75 | ||
| Tyritzis et al, 2013 [56] | Karolinska Institute | 2b | 70 | Retrospective | 58 BNS | 100 | 12 | Internally validated questionnaire | 0–1 pads | – | D: 77 men, |
D: 88 men, |
| 8 UNS | 40 women N: 54 men, 40 women |
67 women | ||||||||||
| N: 76 men, 76 women |
||||||||||||
BNS = bilateral nerve sparing; D = daytime; N = nocturnal; UNS = unilateral nerve sparing.
Follow-up for continence evaluation varied widely, from 6 to 25 mo. Nerve-sparing procedures were performed in 20–100% of patients. Only three of six series reported using a distinct definition forcontinence, which was generally no pad or one pad (safety) per day. One of the earliest RARC series reported an 86% continent rate (seven of eight men) after 3.5 mo [9] . More recent series published 6-mo continence rates of 48–100% for daytime continence and 11–100% for nighttime continence. At 12 mo after RARC, continence rates ranged from 83% to 100% in men and were 67% in women for daytime continence and 66–76% for nighttime continence.
Using strict definitions fordaytime continence(no or one security pad per day) andnighttime continence(goodindicates dry with no protection,fairindicates dry with one awakening), Canda et al examined 23 patients with intracorporeal Studer pouch. After excluding patients who died or were lost to follow-up, 11 of 15 men (73%) and 0 of 2 women were continent during the daytime. Three of these 17 patients (18%) had good nighttime continence, and 4 (24%) had fair continence [33] .
Only one series has described continence results in patients undergoing RARC and continent cutaneous diversion. Torrey et al examined 34 patients who had RARC and Indiana pouch continent cutaneous diversion and reported 97% continence at a mean follow-up of 20 mo for both daytime and nighttime. One patient continued to experience daytime and nighttime incontinence requiring the use of pads [41] .
Table 7 summarizes the series examining potency outcomes. Similar to continence outcomes, evaluation of erectile function after RARC is not well described. Early reports suggest that erections sufficient for penetration are achievable, although sample sizes were very small and lacked validated objective evaluations. Follow-up was again too short to form definitive conclusions, with only one study reporting outcomes up to 2 yr after RARC. As noted earlier, nerve-sparing procedures were performed in 20–100% of patients. The data recording used International Index of Erectile Function (IIEF) scores in five of seven series. However, only three series provided a clear definition ofpotency[12], [35], and [56].
Table 7 Erectile function in robot-assisted radical cystectomy series
| Reference | Institution | IDEAL stage | Cases, no. | Nerve-sparing surgery, % | Study design | Follow-up, mo | Method of data collection | Potency definition | Potency rate at follow-up |
|---|---|---|---|---|---|---|---|---|---|
| Mottrie et al, 2007 [9] | O.L.V.–Clinic | 2a | 27 | 29 | Retrospective | 10.2 | – | – | 86% |
| Murphy et al, 2008 [12] | Guy's Hospital | 2a | 23 | 20 | Retrospective | 17 | IIEF | IIEF >21 with or without PDE5-I | 75% |
| Palou Redorta et al, 2009 [20] | Barcelona Autonomous University | 2a | 9 | 100 | Retrospective | 7 | – | – | 100% |
| Akbulut et al, 2011 [32] | Ankara Ataturk Training and Research Hospital | 2a | 12 | 82 bilateral 9 unilateral |
Not reported | 7.1 | IIEF | None provided | A single patient with IIEF >18 |
| Canda et al, 2012 [33] | Ankara Ataturk Training and Research Hospital | 2a | 27 | 89 | Not reported | 6 | IIEF | None provided | A single patient with IIEF >18 |
| Jonsson et al, 2011 [35] | Karolinska Institute | 2b | 36 | 55 | Prospective | 25 | IIEF | Adequate for penetration with or without PDE5-I | 41% at 12 mo 75% of patients having nerve sparing |
| Tyritzis et al, 2013 [56] | Karolinska Institute | 2b | 70 | 58 bilateral 8 unilateral |
Retrospective | 12 | IIEF | Adequate for penetration with or without PDE5-I | 63% at 12 mo |
IIEF = International Index of Erectile Function; PDE5-I = phosphodiesterase type 5 inhibitor.
In some early, small series, Mottrie et al [9] and Murphy et al [12] reported sufficient erections in six of seven and three of four men, respectively. Similar to well-described literature on RARP, phosphodiesterase type 5 inhibitors (PDE5-Is) were frequently administered to patients for penile rehabilitation after RARC; however, no comparative data in this setting have demonstrated a benefit.
Several series with intracorporeal neobladder have evaluated erectile function postoperatively, with varying results. In the experience of the Karolinska Institute, 41 of 62 men (66%) underwent nerve-sparing RARC. Of these 41 men, 26 (63%) were potent with or without the use of PDE5-Is after 12 mo [56] . In contrast, Canda et al found IIEF scores >18 in only 1 of 11 preoperatively potent men, although follow-up was shorter (6 mo) [33] .
Table 8 summarizes comparative studies evaluating lymph node yield after ORC, LRC, and RARC. In two randomized studies of ORC compared with RARC, lymph node yields were not statistically different[74] and [83]. Cumulative analyses showed no significant difference in lymph node yield between RARC and ORC (OR: 2.94; 95% CI, −0.28 to 6.15;p = 0.07) ( Fig. 2 ).
Table 8 Comparative studies evaluating lymph node yield after open, laparoscopic, and robot-assisted radical cystectomy
| Comparison | Level of evidence | Reference | Cases, no. | Study design | Extension of LND | Retrieved nodes, no. | pN+, no. (%) | Metastatic nodes, no., median |
|---|---|---|---|---|---|---|---|---|
| ORC vs RARC | 2 | |||||||
| Nix et al, 2010 [74] | 21 RARC | RCT | Standard | 19 | 4 (19) | – | ||
| 20 ORC | 18 | 7 (35) | ||||||
| Parekh et al, 2013 [83] | 20 RARC | RCT | Standard | 17.2 ± 13 | 4 (20) | – | ||
| 20 ORC | 24.2 ± 16.4 | 4 (20) | ||||||
| 3 | ||||||||
| Pruthi et al, 2008 [71] | 20 RARC | Gender matched | Standard | 19 | 2 (10) | – | ||
| 24 ORC | Retrospective | 16 | 5 (21) | |||||
| Wang et al, 2008 [72] | 33 RARC | Nonmatched | Standard | 17 | 19 | – | ||
| 21 ORC | 20 | 34 * | ||||||
| Ng et al, 2010 [73] | 83 RARC | Nonmatched | Standard | 17.9 ± 10.4 | 13 (16) | – | ||
| 104 ORC | 15.7 ± 13.2 | 24 (23) * | ||||||
| Richards et al, 2010 [75] | 35 RARC | Nonmatched | Extended | 16 | 10 (29) | – | ||
| 35 ORC | 15 | 10 (29) | ||||||
| Martin et al, 2011 [76] | 19 RARC | Nonmatched | – | 16 | – | – | ||
| 14 ORC | 13 | |||||||
| Gondo et al, 2012 [92] | 11 RARC | Nonmatched | Extended | 20.7 ± 8.2 | 9 | – | ||
| 15 ORC | 13.8 ± 6.6 * | 13 | ||||||
| Khan et al, 2012 [87] | 48 RARC | Prospective | Extended | 16 | 5 | – | ||
| 52 ORC | 11 | 15 | ||||||
| Richards et al, 2012 [77] | 20 RARC | Nonmatched | Extended | 17 | 7 (35) | – | ||
| 20 ORC (>75 yr) | 15 | 3 (15) | ||||||
| Styn et al, 2012 [78] | 50 RARC | 1:2 by age, sex, clinical stage, diversion | – | 14.3 ± 9.1 | 6 (12) | – | ||
| 100 ORC | 15.2 ± 9.5 | 19 (19) | ||||||
| Sung et al, 2012 [79] | 35 RARC | Nonmatched | Standard | 19.1 ± 8.2 | 9 (26) | – | ||
| 104 ORC | 12.9 ± 9.0 * | 27 (26) | ||||||
| Knox et al, 2013 [86] | 58 RARC | Nonmatched | Extended | 21 | 1 | – | ||
| 84 ORC | 17 | 3 | ||||||
| Maes et al, 2013 [80] | 14 RARC | Nonmatched | Extended | 11.9 | 1 (7) | – | ||
| 14 ORC | 9.5 | 5 (35) | ||||||
| Musch et al, 2014 [81] | 100 RARC | Nonmatched | – | 27.5 ± 11.0 | 20 (20) | – | ||
| 42 ORC | 19.6 ± 8.8 * | 9 (21) | ||||||
| Nepple et al, 2013 [82] | 36 RARC | Nonmatched | Standard | 17 | 8 (22) | – | ||
| 29 ORC | 14 | 7 (24) | ||||||
| 4 | Abaza et al, 2012 [88] | 35 RARC | Nonmatched | Extended | 37.5 ± 13.2 | 12 (34) | 1.5 | |
| 120 ORC | 36.9 ± 14.8 | 36 (30) | 2 | |||||
| LRC vs RARC | ||||||||
| 3 | Khan et al, 2012 [87] | 48 RARC | Prospective | Extended | 16 | 5 | – | |
| 58 LRC | 10 | 10 | ||||||
| 4 | Abraham et al, 2007 [89] | 14 RARC | Nonmatched | 10 extended | 22.3 | 2 (10) | ||
| 20 LRC | 16 extended | 16.5 | 2 (12.5) | |||||
* Statistically significant.
LND = lymph node dissection; LRC = laparoscopic radical cystectomy: ORC = open radical cystectomy; RARC = robot-assisted radical cystectomy; RCT = randomized controlled trial.
Fig. 2 Comparison of lymph node yields following robot-assisted or open radical cystectomy. CI = confidence interval; ORC = open radical cystectomy; RARC = robot-assisted radical cystectomy; SD = standard deviation.
Table 9 summarizes PSM rates in RARC, ORC, and LRC. In two randomized trials comparing RARC and ORC, Nix et al and Parekh et al did not show any increase in positive margins with RARC[74] and [83]. Cumulative analyses showed no significant difference in rates of surgical margins between RARC and ORC (5% and 7%, respectively; OR: 0.71; 95% CI, 0.46–1.1;p = 0.13) ( Fig. 3 ). In two comparative nonrandomized studies between RARC and LRC, no significant differences in PSM rates were detected (p = 0.86)[87] and [89].
Table 9 Comparative studies evaluating positive surgical margins after open, laparoscopic, and robot-assisted radical cystectomy
| Comparison | Level of evidence | Reference | Cases, no. | Pathologic stage, % | Overall PSM, no. (%) | PSM in pT2 cancer | |
|---|---|---|---|---|---|---|---|
| pT2 | pT3 | ||||||
| ORC vs RARC | 2b | ||||||
| Nix et al, 2010 [74] | 21 RARC | 67 | 14 | 0 | 0 | ||
| 20 ORC | 40 | 25 | 0 | 0 | |||
| Parekh et al, 2013 [83] | 20 RARC | 50 | 50 | 1 (5) | 0 | ||
| 20 ORC | 65 | 35 | 1 (5) | 0 | |||
| 3 | |||||||
| Rhee et al, 2006 [84] | 7 RARC | 86 | 14 | 0 | 0 | ||
| 23 ORC | 43 | 57 | 0 | 0 | |||
| Galich et al, 2006 [90] | 13 RARC | 54 | 46 | 0 | – | ||
| 24 ORC | 37 | 63 | 3 (12) | ||||
| Pruthi et al, 2007 [71] | 20 RARC | 78 | 22 | 0 | 0 | ||
| 24 ORC | 63 | 37 | 0 | ||||
| Wang et al, 2008 [72] | 33 RARC | 72 | 28 | 2 (6) | – | ||
| 21 ORC | 43 | 57 | 3 (14) | ||||
| Ng et al, 2010 [73] | 83 RARC | 61 | 39 | 6 (7) | 0 | ||
| 104 ORC | 58 | 42 | 9 (9) | 0 | |||
| Richards et al, 2010 [75] | 35 RARC | 60 | 40 | 1 (3) | – | ||
| 35 ORC | 57 | 43 | 3 (9) | ||||
| Martin et al, 2011 [76] | 19 RARC | 42 | 58 | – | – | ||
| 14 ORC | 93 | 7 | |||||
| Gondo et al, 2012 [92] | 11 RARC | 91 | 9 | 1 (9) | – | ||
| 15 ORC | 53 | 47 | 2 (13) | ||||
| Khan et al, 2012 [87] | 48 RARC | 75 | 25 | 0 | – | ||
| 52 ORC | 50 | 50 | 6 (10) | ||||
| Richards et al, 2012 [77] | 20 RARC | 60 | 40 | 1 (5) | – | ||
| 20 ORC (>75 yr) | 50 | 50 | 2 (10) | ||||
| Styn et al, 2012 [78] | 50 RARC | 60 | 40 | 1 (2) | – | ||
| 100 ORC | 72 | 28 | 1 (1) | ||||
| Sung et al, 2012 [79] | 35 RARC | 43 | 57 | – | – | ||
| 104 ORC | 38 | 62 | |||||
| Kader et al, 2013 [91] | 100 RARC | 58 | 42 | 12 (12) | – | ||
| 100 ORC | 53 | 47 | 11 (11) | ||||
| Knox et al, 2013 [86] | 58 RARC | 66 | 34 | 4 (7) | – | ||
| 84 ORC | 43 | 57 | 7 (8) | ||||
| Maes et al, 2013 [80] | 14 RARC | 43 | 57 | 3 (21) | – | ||
| 14 ORC | 57 | 43 | 2 (14) | ||||
| Musch et al, 2013 [81] | 100 RARC | 61 | 39 | 2 (2) | – | ||
| 42 ORC | 57 | 43 | 1 (2) | ||||
| Nepple et al, 2013 [82] | 36 RARC | 53 | 47 | 2 (6) | 0 | ||
| 29 ORC | 58 | 42 | 2 (7) | 0 | |||
| 4 | Abaza et al, 2012 [88] | 35 RARC | 60 | 23 | 2 (6) | 0 | |
| 120 ORC | 45 | 42 | 8 (7) | 0 | |||
| LRC vs RARC | |||||||
| 3 | Khan et al, 2012 [87] | 48 RARC | 75 | 25 | 0 | – | |
| 58 LRC | 57 | 43 | 2 (4) | ||||
| 4 | Abraham et al, 2007 [89] | 14 RARC | – | – | 1 (7) | 0 | |
| 20 LRC | 0 | – | |||||
LRC = laparoscopic radical cystectomy; ORC = open radical cystectomy; PSM = positive surgical margin; RARC = robot-assisted radical cystectomy.
Fig. 3 Comparison of positive surgical margin rates following robot-assisted or open radical cystectomy. CI = confidence interval; M-H = Mantel-Haenszel; ORC = open radical cystectomy; RARC = robot-assisted radical cystectomy.
Table 10 summarizes series that emphasized early oncologic comparisons for RARC, LRC, and ORC, though interpretation should be cautious with small series of shorter follow-up and potential bias of patient selection. A nonrandomized comparison of ORC (n = 52) with RARC (n = 48) with a follow-up of 38 mo showed disease-specific survival of 69% in the ORC group compared with 79% in the RARC group [87] . A series by Nepple et al showed similar estimates in DFS, CSS, and OS, although patients were not matched [82] .
Table 10 Comparative studies evaluating recurrence-free, cancer-specific, and overall survival estimates after open, laparoscopic, and robot-assisted radical cystectomy
| Comparison | Level of evidence | Reference | Cases, no. | Study design | Follow–up, mo | Neoadjuvant chemotherapy, % | DFS estimates, % | CSS estimates, % | OS estimates, % |
|---|---|---|---|---|---|---|---|---|---|
| ORC vs RARC | 3 | ||||||||
| Khan et al, 2012 [87] | 48 RARC | Prospective | 38 | – | – | 79 | – | ||
| 52 ORC | 69 | ||||||||
| Nepple et al, 2013 [82] | 36 RARC | Nonmatched | 12 | 6 | 67 (2 yr) | 75 (2 yr) | 68 (2 yr) | ||
| 29 ORC | 14 | 58 (2 yr) | 63 (2 yr) | 63 (2 yr) | |||||
| LRC vs RARC | 3 | ||||||||
| Khan et al, 2012 [87] | 48 RARC | Prospective | 38 | – | – | 79 | – | ||
| 58 LRC | 93 |
CSS = cancer-specific survival; DFS = disease-free survival; LRC = laparoscopic radical cystectomy; ORC = open radical cystectomy; OS = overall survival; RARC = robot-assisted radical cystectomy.
Our systematic review sought to identify and report the current state of the literature for RARC with regard to pathologic, oncologic, and functional outcomes. Various oncologic parameters, including pathologic findings and postoperative survival rates, were examined. With regard to nodal dissection, robotic ELND achieves a similar nodal yield to open ELND when performed by experienced surgeons. Nearly all RARC series reported nodal yields >15. With regard to margin rates, most series reported PSM rates of <10%, with rates of approximately 1% in pT2 disease. Although the IRCC (n = 513) reported a very high positive margin rate of 39% in pT4 patients [68] , other authors have reported rates similar to those noted in ORC series.
Although these immediate pathologic variables may act as surrogates for quality of resection, long-term survival outcomes must be analogous to those of ORC for RARC to be a viable surgical option. Currently, oncologic data are immature, and adequate comparative studies of RARC and ORC are nonexistent. In a few analyses measuring CSS and OS at 5 yr postoperatively, results appear similar to those reported in ORC; however, larger numbers and longer follow-up are needed for adequate comparison. At present, data reporting functional analysis of continence and potency recovery after RARC are inadequate to compare RARC reliably with ORC.
PLND, in conjunction with radical cystectomy, provides a staging benefit as well as a possible advantage for survival in retrospective studies. Stein et al examined 1054 patients treated with radical cystectomy and PLND with a 24% node-positive rate; these patients experienced 5- and 10-yr recurrence-free survival of 35% and 34%, respectively [2] . Although prospective validation is necessary, Leissner et al suggested that ELND improved outcomes in both low-volume node-positive and node-negative patients with greater number of lymph nodes removed [93] . The true survival benefit of ELND must be proven in a prospective fashion to overcome the Will Rogers phenomenon of apparent improved survival that results from stage migration with more thorough dissection.
Early critical concerns of RARC involved whether LND could be performed robotically with the same quality as during ORC. This review suggests that thorough robotic ELND dissection at the time of RARC is possible following a similar template as is performed during ORC. Half of current RARC series describe an extended template dissection, with the average number of lymph nodes removed between 11 and 55. In a small study of open completion LND after robotic LND, only four additional lymph nodes were recovered [34] . Although few series described the time necessary to perform a complete robotic LND, some authors described operative times approaching 2 h for the node dissection alone, suggesting that robotic LND may lengthen operative time. Further study is necessary to determine whether the LND segment of RARC is significantly longer compared with open LND. Complications specific to LND—particularly vascular injuries—were rare, as were lymphoceles, with an incidence <10%. However, complication rates may often be underreported, as reporting guidelines lack standardization.
Assessment of RARC lymph node yields as related to patient characteristics (eg, BMI) or surgeon characteristics (eg, prior RARP experience) has not shown a specific association. Similar to the ORC literature, Bochner et al reported that only extent of LND was associated with lymph node yield when examining variables such as receipt of neoadjuvant chemotherapy, pathologic stage, surgeon, and pathologist [94] . Although performance of LND was associated with higher surgeon volume, analysis of several learning curve evaluations did not find increases in lymph node yield with increasing case number. This result may seem counterintuitive, but it may be that these experienced robotic surgeons were able to translate surgical technique from RARP and PLND and thus reduce the number of cases needed to reach stable lymph node yields. Instead, a reduction in LND time could occur with experience, although it has not been specifically examined. In the IRCC database, patients with pT4 disease had lower nodal yields, possibly related to more difficult dissection or to RARC being performed for palliative intent.
PSM at cystectomy is a measure of disease burden and a predictor of outcome. In a previous study of 1589 patients who underwent radical cystectomy at Memorial Sloan Kettering Cancer Center, the positive margin rate was 4.2%. Risk factors for PSMs were female sex, higher pathologic stage, vascular invasion, mixed histology, and lymph node involvement. Patients with PSMs had a 5-yr CSS of only 32% [95] . In another large multi-institutional analysis of 4400 ORC patients, the incidence of PSMs was 6.3% [96] . A potential challenge of RARC and limitation of current robotic technology is in treating bulkier tumors because of the lack of tactile feedback.
The present systematic review demonstrates that PSMs are uncommon in RARC series and appropriately rare for pT2 disease. No significant difference was found when comparing the surgical margin rate between RARC and ORC. The high variability of positive margins across studies, between 0% and 26%, suggests significant heterogeneity in cancer characteristics, patient selection, and surgical technique and experience, among other variables. From the systematic review, the weighted average of positive margins in RARC series was 5.6%, which is comparable to the large open series cited earlier. In the aforementioned analysis of 4400 ORC patients, margin-positive rates by stage were 2.3% for pT2, 7.6% for pT3, and 24% for pT4 disease [96] . The effects of the learning curve as institutions adopted this new technology and patient selection toward earlier stage disease likely affected reported margin rates and should be considered when interpreting outcomes. Nonetheless, higher reported rates of positive margins in pT4 disease in some RARC series suggest that caution be taken for higher stage disease, with particular attention paid to the risk of margin involvement.
Several RARC series did not show decreasing margin rates with sequential case volume. A few reasons could explain this observation: (1) The positive margin numbers may be too low to detect a subgroup difference; (2) the learning curve for reducing margins at RARC could be extremely high, with a number not yet reached in smaller learning curve assessments; or (3) over time, more experienced surgeons may be more willing to take on bulky or higher stage tumors. This final hypothesis is supported by a multivariate analysis adjusting for pathologic stage that shows that differences in stage of disease accounted for an increase in margin rates with more experienced robotic surgeons [61] .
Chemotherapy use alongside surgery in the treatment of MIBC can be implemented either before or after cystectomy. While neoadjuvant chemotherapy has been shown to confer an OS advantage of 5% in randomized trials [97] , the benefit of adjuvant chemotherapy is less proven. In a recent meta-analysis of nine randomized controlled trials comprising 945 patients that investigated the use of adjuvant chemotherapy, benefits to both OS and DFS were appreciated. Patients receiving adjuvant chemotherapy after cystectomy had 23% relative risk reduction in the risk of death (OS:p = 0.049) and 34% relative decrease in the risk of disease recurrence (DFS:p = 0.014) [98] . For the current systematic review, neoadjuvant chemotherapy use was 0–31%, and adjuvant chemotherapy was delivered to 4–29% of patients. Adjuvant therapy was chiefly administered in patients with advanced-stage pT3 or higher or with positive lymph nodes. Although time to initiation of adjuvant chemotherapy was shorter by 3 wk in the analysis by Pruthi et al, further validation is required [31] .
Long-term freedom from disease recurrence and bladder cancer–related death is the primary measure of treatment efficacy with radical cystectomy. Particularly with assessments of survival, gathering data for comparison with the open standard is challenging secondary to the necessity of controlling for cancer characteristics, additional therapies, and the length of follow-up required to detect significant differences. Shorter-interval examinations of survival may not amply capture events such as local recurrence, distant recurrence, or secondary therapies. Only two series in this systematic review compared survival for RARC and ORC. These studies were not randomized and included sequential series of retrospective groups (LOE 4)[82] and [87]. In a series by Nepple et al, 36 patients who underwent RARC were compared with 29 patients who underwent ORC with a median follow-up of only 12 mo. Estimated 2-yr DFS (67% vs 58%), CSS (75% vs 63%), and OS (68% vs 63%) after RARC and ORC were similar for the two techniques, respectively [82] .
Because of limitations of present studies, comparisons must be made to large historical retrospective open series. A long-term analysis of survival in 1100 chemotherapy-naive cystectomy patients by Hautmann et al demonstrated 10-yr CSS and OS rates of 67% and 44%, respectively [99] . For this systematic review, 5-yr estimates for DFS, CSS, and OS were 53–74%, 66–80%, and 39–66%, respectively. Analogous to stratified outcomes in ORC, survival outcomes were worse in RARC series with increasing pathologic stage and with lymph node metastases. Local control of disease appears to be adequate such that the majority of recurrences after RARC are distant or outside the pelvis. A potential concern for port-site metastases with RARC remains of particular interest. Although no specific published series address this concern and most larger RARC oncologic series did not report any incidents, a few case reports suggest that this concern requires further study.
Since the original description of neurovascular bundle preservation during radical prostatectomy by Walsh et al [104] , techniques to improve functional outcomes through meticulous nerve sparing have been translated to radical cystectomy. Turner et al determined that nerve sparing improved urinary continence after orthotopic urinary diversion [100] , and nerve sparing has been shown to assist with recovery of erectile function objectively based on IIEF [101] . Long-term functional evaluations of ileal neobladder continent diversions have demonstrated daytime continence rates of 92% and nighttime continence rates of 80% [102] .
To date, very limited data are available regarding functional outcomes of continence or potency after RARC. These analyses have chiefly been limited to only a few centers that exhibit significant heterogeneity. The 12-mo reported continence rates were 88–100% in men and 67% in women for daytime continence and between 66% and 75% for nighttime continence. Potency recovery exhibited even greater variation, with sufficient erection rates between 9% and 81%. Functional outcomes are likely influenced by patient factors and selection, comorbidity, prior treatments, surgeon experience, and technique (eg, the use of cautery vs clips). In addition, methodology of reporting, definitions ofcontinence, measurement tools, rehabilitation programs, and inconsistencies in follow-up can affect the actual measurement of continence and potency. Specific functional concerns of RARC related to patient selection are that many patients may be older or have poor baseline erectile function. Moreover, technical concerns for a possible PSM, which portends a dismal outcome, may affect the performance of nerve sparing. The lack of conclusive data regarding functional recovery after RARC is a necessary area for future study.
There is no evidence to date that the results from a recent systematic review on RARP finding slight advantages to continence and potency recovery compared with open radical prostatectomy or laparoscopic radical prostatectomy extrapolate to RARC [103] . Precise definitions ofcontinenceandpotencyare necessary so that future data acquisition can be carried out in a standardized, stringent, and uniform fashion for both ORC and RARC.
From a methodological perspective, the most relevant limitations of this systematic review are the quality of the available studies, the small number of patients in and the retrospective nature of most series, the shorter-term follow-up of these studies, and the lack of standardized definitions. The papers included in the present review included only two small randomized controlled trials; the remaining series are LOE 3 or 4. Comparisons made in these single-institution studies inevitably carry the risk of selection bias. Even in randomized controlled studies, there were unlikely to have been equally experienced open and robotic surgeons operating on comparable patients. Heterogeneity in lymph node templates, sampling methods, specimen handling, and pathologic review may affect lymph node yields. Most cumulative outcomes were weighted by the results of experienced surgeons, which may make conclusions difficult to generalize. The inability to account for surgeon factors or specific technique modifications is another limitation. Most series failed to provide specific information concerning relevant aspects of the reconstructive portions of the operation.
Sufficient lymph node yields are achievable through robotic PLND if an extended template is followed. PSM rates appear similar with RARC and ORC. Conclusive long-term survival outcomes for RARC are limited, although oncologic outcomes of ≤5 yr are similar to those reported for ORC. Initial functional outcomes appear favorable; however, additional research on continence and potency after RARC is needed.
Author contributions: Bertram Yuh had full access to all the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.
Study concept and design: Yuh, Wilson, Bochner, Montorsi, Chan, Thalmann, Palou, Stenzl, Guru, Catto, Novara, Wiklund.
Acquisition of data: Yuh, Wilson, Bochner, Montorsi, Chan, Thalmann, Palou, Stenzl, Guru, Catto, Novara, Wiklund.
Analysis and interpretation of data: Yuh, Novara.
Drafting of the manuscript: Yuh, Wilson, Bochner, Montorsi, Chan, Thalmann, Palou, Stenzl, Guru, Catto, Novara, Wiklund.
Critical revision of the manuscript for important intellectual content: Yuh, Wilson, Bochner, Montorsi, Chan, Thalmann, Palou, Stenzl, Guru, Catto, Novara, Wiklund.
Statistical analysis: Yuh, Novara.
Obtaining funding: Wilson.
Administrative, technical, or material support: Yuh, Wilson, Bochner, Montorsi, Chan, Thalmann, Palou, Stenzl, Guru, Catto, Novara, Wiklund.
Supervision: Yuh, Wilson, Bochner, Montorsi, Chan, Thalmann, Palou, Stenzl, Guru, Catto, Novara, Wiklund.
Other(specify): None.
Financial disclosures: Bertram Yuh certifies that all conflicts of interest, including specific financial interests and relationships and affiliations relevant to the subject matter or materials discussed in the manuscript (eg, employment/affiliation, grants or funding, consultancies, honoraria, stock ownership or options, expert testimony, royalties, or patents filed, received, or pending), are the following: Dr. Wiklund has received proctoring assignments and a research grant from Intuitive Surgical, Sunnyvale, CA, USA; Dr. Wilson has been a consultant and a speaker for Intuitive Surgical, Sunnyvale, CA, USA.
Funding/Support and role of the sponsor: None.
Radical cystectomy and pelvic lymph node dissection (PLND) is the gold standard treatment for muscle-invasive bladder cancer (MIBC) and high-risk non–muscle-invasive disease [1] . Patients undergoing this operation can experience 66% recurrence-free survival at 10 yr after surgery [2] . The addition of neoadjuvant platinum-based chemotherapy has been shown to improve overall survival (OS) rates by approximately 5% [3] . Robot-assisted radical cystectomy (RARC) was initially described by Menon et al in 2003 [4] . Over time, many international centers have adopted RARC.
Oncologic outcomes from large population-based cohorts of RARC with lengthy follow-up are lacking. Early on in RARC history, surrogates for oncologic control were reported using positive surgical margin (PSM) rates and lymph node yields. More recently, 5-yr survival figures have become available. The majority of these outcomes, however, capture institutions early in their learning curves and incorporate patients potentially selected for the robotic technique, thus avoiding more advanced-stage or technically difficult cases. Data on functional consequences of RARC are even more limited; therefore, the quality of nerve sparing and its effect on potency recovery and continence are inadequately understood.
Because of the expanding evidence available in the field of RARC, and in preparation for the Pasadena international consensus meeting on best practice in RARC and urinary reconstruction, we performed a systematic literature review of perioperative, functional, and oncologic outcomes of RARC in comparison with open radical cystectomy (ORC) and laparoscopic radical cystectomy (LRC).
We report on the systematic review and cumulative analysis of oncologic and functional outcomes of RARC. We systematically examined lymph node yields, PSMs, cancer-specific survival (CSS), disease-free survival (DFS), and OS. In addition, functional outcomes after RARC, including urinary continence and erectile function, were systematically examined.
A systematic literature search was initially performed in September 2013 using the Medline, Scopus, and Web of Science databases. The searches included only a free-text protocol using the termsrobot-assisted radical cystectomyorda Vinci radical cystectomyorrobot* radical cystectomyin all the fields of the records for Medline and Scopus searches and in theTitleandTopicfields for the Web of Science search. No limits were applied. A full update of the searches was performed on April 28, 2014.
Two authors (G.N. and B.Y.) separately reviewed the records to select RARC case series and studies that compared RARC with ORC and RARC with LRC. Discrepancies were resolved by open discussion. Other significant studies cited in the reference lists of the selected papers were evaluated, as were studies published after the systematic search.
All noncomparative studies reporting the following data on RARC were collected: intraoperative and perioperative data (operative time, blood loss, transfusion rate, in-hospital stay, readmission, complication rates), functional data (urinary continence, erectile function), and oncologic data (PSMs, lymph node yield, DFS, CSS, OS). The present review included only studies reporting on functional and oncologic data.
Studies reporting on partial cystectomy, prostate-sparing cystectomy, salvage cystectomy, cystectomy for urachal cancer or benign disease, single-case reports, pure laparoscopic (or mixed) series, or laparoendoscopic single-site or natural orifice transluminal endoscopic surgery for radical cystectomy; experimental studies on animal models; congress abstracts; review papers; editorials; population-based studies; and book chapters were not included in the review. All data retrieved from the selected studies were recorded in an electronic database.
All papers were categorized according to the 2011 levels of evidence (LOEs) for therapy studies: LOE 1, systematic review of randomized trials or n-of-1 trials; LOE 2, randomized trial or observational study with dramatic effect; LOE 3, nonrandomized controlled cohort/follow-up study; LOE 4, case series, case–control study, or historically controlled study; or LOE 5, mechanism-based reasoning [5] . Papers were categorized according to the IDEAL recommendations [6] .
Cumulative analysis was conducted using Review Manager v5.2 software designed for composing Cochrane Reviews (Cochrane Collaboration, Oxford, UK). Statistical heterogeneity was tested using the chi-square test. Apvalue <0.10 was used to indicate heterogeneity. Where there was a lack of heterogeneity, fixed-effects models were used for the cumulative analysis. Random-effects models were used in case of heterogeneity. For continuous outcomes, the results were expressed as weighted mean differences and standard deviations (SDs); for dichotomous variables, results were given as odds ratios (ORs) and 95% confidence intervals (CIs). Because of limitations in the Review Manager v5.2 software, meta-analysis of continuous variables was possible only when rough data were presented as mean and SD. Authors of the papers were contacted to provide missing data, whenever necessary. For all statistical analyses, two-sidedp < 0.05 was considered statistically significant.
Figure 1 shows a flowchart of this systematic review of the literature.
In total, 65 surgical series[4], [7], [8], [9], [10], [11], [12], [13], [14], [15], [16], [17], [18], [19], [20], [21], [22], [23], [24], [25], [26], [27], [28], [29], [30], [31], [32], [33], [34], [35], [36], [37], [38], [39], [40], [41], [42], [43], [44], [45], [46], [47], [48], [49], [50], [51], [52], [53], [54], [55], [56], [57], [58], [59], [60], [61], [62], [63], [64], [65], [66], [67], [68], [69], and [70]and 22 comparative studies[71], [72], [73], [74], [75], [76], [77], [78], [79], [80], [81], [82], [83], [84], [85], [86], [87], [88], [89], [90], [91], and [92]reported on pathologic, oncologic (n = 18), or functional (n = 9) outcomes of RARC.
Most surgical series were retrospective, single-center studies (LOE 4). Exceptions included prospective studies[8], [22], [28], [35], [36], [43], [52], [59], [64], [66], [81], and [87]and some multi-institutional retrospective collaboration studies[16], [23], [25], [30], [45], [48], [54], [61], and [68]. Only two of the comparative studies were randomized[74] and [83](LOE 2b); all other comparative studies were nonrandomized, whether prospective or retrospective (LOE 4).
Table 1 summarizes the number of lymph nodes recovered in published RARC series. The majority of studies (86%) reported extent of lymph node dissection (LND), with more centers performing extended LND (ELND) in recent series. Standard LND typically involved the removal of obturator, internal iliac, external iliac, and some portion of the common iliac lymph nodes bilaterally. ELND templates typically brought the proximal extent up to the aortic bifurcation or inferior mesenteric artery. Approximately half of the analyzed studies reported following an extended template of dissection.
Table 1 Lymph node yields in robot-assisted radical cystectomy series
| Reference | Institution | IDEAL stage | Cases, no. | Study design | Extension of LND | Operative time, min | Retrieved nodes, no. | pN+, % | Metastatic nodes, median, no. | Complications due to LND |
|---|---|---|---|---|---|---|---|---|---|---|
| Menon et al, 2003 [4] | Henry Ford Hospital | 1 | 17 | Retrospective | Standard | – | – | 6 | – | – |
| Menon et al, 2004 [7] | Henry Ford Hospital | 1 | 3 female | Retrospective | Standard | – | 12 | 0 | – | – |
| Guru et al, 2007 [8] | Roswell Park Cancer Institute | 1 | 20 | Prospective | Standard | 44 | 13 | 15 | 1 | – |
| Mottrie et al, 2007 [9] | O.L.V.–Clinic | 2a | 27 | Retrospective | Extended | – | 23 | 9 | – | – |
| Pruthi et al, 2008 [71] | UNC | 2a | 20 | Retrospective | Standard | – | 19 | 10 | – | – |
| Hemal et al, 2008 [10] | All India Institute of Medical Sciences | 1 | 6 | Retrospective | Standard | – | 12 | 17 | – | – |
| Lowentritt et al, 2008 [11] | Tulane University | 2a | 4 | Retrospective | Standard | – | 12 | 25 | – | – |
| Murphy et al, 2008 [12] | Guy's Hospital | 2a | 23 | Retrospective | Standard | – | 16 | 9 | – | – |
| Park et al, 2008 [13] | Yonsei | 2a | 4 | Retrospective | Standard | – | 17 | 0 | – | – |
| Pruthi et al, 2008 [14] | UNC | 2a | 12 female | Retrospective | Standard, then extended | – | 19 | 17 | – | – |
| Pruthi et al, 2008 [15] | UNC | 2b | 50 | Retrospective | Standard | – | 19 | 20 | – | – |
| Wang et al, 2008 [72] | Cornell | 2b | 33 | Retrospective | Standard | – | 17 | 19 | – | – |
| Woods et al, 2008 [16] | Mayo Arizona Tulane University |
2b | 27 | Multi-institutional | Extended | – | 12.3 | 33 | 3.1 | 0 |
| Yuh et al, 2008 [17] | Roswell Park Cancer Institute | 2a | 54 | Retrospective | Extended | – | 17 | – | – | – |
| Gamboa et al, 2009 [18] | University of California, Irvine |
2a | 41 | Retrospective | Standard | – | 23 | 14 | 4 | – |
| Pruthi et al, 2009 [19] | UNC | 2b | 50 | Retrospective | Standard, then extended | – | 19 | 16 | – | – |
| 10 female | 19 | |||||||||
| 40 male | 18 | |||||||||
| Palou Redorta et al, 2009 [20] | Barcelona Autonomous University | 2a | 9 | Retrospective | Extended | 60 | 10 | 0 | – | – |
| Yuh et al, 2009 [21] | Roswell Park Cancer Institute | 2b | 73 | Retrospective | Extended | – | 19 | – | – | – |
| Guru et al, 2010 [22] | Roswell Park Cancer Institute | 2a | 26 | Prospective | Extended | – | 21 | 29 | 1 | Internal iliac artery injury: 1 |
| Hellenthal et al, 2011 [23] | IRCC | 2b | 437 | Multi-institutional | – | – | 17 | 20 | – | – |
| Josephson et al, 2010 [24] | City of Hope Cancer Center | 2b | 58 | Retrospective | Extended | – | 27 | 24 | – | – |
| Kang et al, 2010 [25] | Multicenter | 2b | 71 standard LND 33 extended LND |
Retrospective | Standard Extended |
– | 15.7 24.7 |
10 | – | – |
| Kasraeian et al, 2010 [26] | Montsouris Institute | 2a | 9 | Retrospective | Extended | – | 11 | 22 | – | 0 |
| Kauffman et al, 2011 [27] | Cornell | 2b | 85 | Retrospective | Extended | – | 19 | 15 | – | – |
| Kwon et al, 2010 [28] | Kyungpook National University | 2a | 17 | Prospective | Standard | – | 6 | 6 | 1 | 0 |
| Lavery et al, 2011 [29] | Ohio State University | 2a | 15 | Retrospective | Extended | 107 | 41.8 | 20 | – | 0 |
| Martin et al, 2010 [30] | Mayo Arizona Tulane University |
2b | 59 | Multi-institutional | Extended | – | – | 34 | – | – |
| Ng et al, 2010 [73] | Cornell | 2b | 83 | Retrospective | Standard | – | 16 | 16 | – | – |
| Nix et al, 2010 [74] | UNC | 3 | 21 | RCT | Standard | – | 19 | 19 | – | – |
| Pruthi et al, 2010 [31] | UNC | 2b | 100 | Retrospective | Standard, then extended | – | 19 | 20 | – | – |
| Richards et al, 2010 [75] | Wake Forest University | 2b | 35 | Retrospective | Extended | – | 16 | 29 | – | – |
| Akbulut et al, 2011 [32] | Ankara Ataturk Training and Research Hospital | 2a | 12 | Not reported | Extended | – | 21.3 | 42 | – | 8 |
| Canda et al, 2012 [33] | Ankara Ataturk Training and Research Hospital | 2a | 27 | Not reported | Extended | – | 24.8 | 22 | – | – |
| Davis et al, 2011 [34] | University of Texas M.D. Anderson Cancer Center | 2a | 11 | Retrospective | Extended | 117 | 43 | 9 | 1 | – |
| Jonsson et al, 2011 [35] | Karolinska Institute | 2b | 45 36 neobladder 9 ileal conduit |
Prospective | Standard Extended |
– | 19 19 27 |
20 17 33 |
– | – |
| Khan et al, 2011 [36] | Guy's Hospital | 2a | 50 | Prospective | – | – | 17 | – | – | – |
| Manoharan et al, 2011 [37] | University of Miami | 2a | 14 | Retrospective | Standard | – | 12 | – | – | – |
| Martin et al, 2011 [76] | Mayo Arizona | 2b | 19 | Retrospective | – | – | 16 | – | – | – |
| Richards et al, 2011 [38] | Wake Forest University | 2b | 60 | Retrospective | Extended | – | 17 | 30 | – | Lymphocele: 1 |
| Schumacher et al, 2011 [39] | Karolinska Institute | 2b | 45 | Retrospective | Standard 49%, extended 31% | – | 22.5 | – | 1.5 | Lymphocele: 2 |
| Shah et al, 2011 [40] | Ohio State University | 2b | 30 | Retrospective | Extended | – | – | 30 | – | – |
| Torrey et al, 2011 [41] | City of Hope Cancer Center | 2b | 34 | Retrospective | Extended | – | 28.9 | – | – | – |
| Cho et al, 2012 [42] | Hallym University College of Medicine | 2b | 35 | Retrospective | Standard | – | – | 6 | – | – |
| Goh et al, 2012 [43] | Keck School of Medicine, University of Southern California, Los Angeles | 2a | 15 | Prospective | Superextended | – | 55 | 26 | – | – |
| Lau et al, 2012 [44] | City of Hope Cancer Center | 2b | 23 (aged >80 yr) | Retrospective | Extended | – | 20.4 | 22 | – | – |
| Mmeje et al, 2013 [45] | Mayo Arizona UNC |
2b | 50 | Multi-institutional | Extended | – | 18 | 100 | 3 | – |
| Poch et al, 2012 [46] | Roswell Park Cancer Institute | 2b | 56 | Retrospective | – | – | 25 | 16 | – | – |
| Richards et al, 2012 [77] | Wake Forest University | 2b | 20 (aged >75 yr) | Retrospective | Extended | – | 17 | 35 | – | – |
| Saar et al, 2013 [47] | Saarland University | 2b | 62 | Retrospective | – | – | 14.2 | 21 | – | – |
| Smith et al, 2012 [48] | Mayo Arizona UNC, Tulane University |
2b | 227 | Multi-institutional | – | – | 18 | 20 | – | – |
| Styn et al, 2012 [78] | University of Michigan | 2b | 50 | Retrospective | – | – | 14.3 | 12 | – | – |
| Sung et al, 2012 [79] | Samsung Medical Center | 2b | 35 | Retrospective | Standard | – | 19.1 | 26 | – | Lymphocele: 1 |
| Treiyer et al, 2012 [49] | Saarland University | 2b | 91 | Retrospective | Standard | – | 14.5 | 14 | – | – |
| Tsui et al, 2012 [50] | Chang Gung Memorial Taiwan | 2a | 8 | Retrospective | Standard | – | 3 | 12.50 | – | – |
| Yuh et al, 2012 [51] | City of Hope Cancer Center | 2b | 196 | Retrospective | Extended | – | 28 | 22 | – | Lymphocele: 3 |
| Collins et al, 2013 [52] | Karolinska Institute | 2b | 113 | Prospective | Extended 56%, standard 34%, limited 5%, none 5% | – | 21 | 20 | – | Lymphocele: 5 |
| Johar et al, 2013 [53] | IRCC | 2b | 939 | Multi-institutional | – | – | 18.1 | 26 | – | – |
| Maes et al, 2013 [80] | Metro Health Hospital | 2b | 14 | Retrospective | Extended | – | 11.9 | 7 | – | – |
| Marshall et al, 2013 [54] | IRCC | 2b | 765 | Multi-institutional | Extended 58%, standard 40%, no LND 2% | – | 18 | 27 | – | – |
| Musch et al, 2014 [81] | Klinikin Essen–Mitte | 2b | 100 | Prospective | – | – | 26.5 | 20 | – | Lymphocele: 4 |
| Nazmy et al, 2014 [55] | City of Hope Cancer Center | 2b | 209 | Retrospective | Extended | – | – | 22 | – | Lymphocele: 3 |
| Nepple et al, 2013 [82] | Washington University | 2b | 36 | Retrospective | Standard | – | 17 | 22 | – | – |
| Parekh et al, 2013 [83] | University of Texas Health Sciences Center at San Antonio | 3 | 20 | RCT | Standard | – | 11 | 20 | – | – |
| Tyritzis et al, 2013 [56] | Karolinska Institute | 2b | 70 | Retrospective | Standard 43% Extended 48% |
– | 21 | 14 | – | Lymphocele: 6 Lymphedema: 1 |
| Xylinas et al, 2013 [57] | Cornell | 2b | 175 | Retrospective | Standard | – | 19 | 17 | – | Lymphocele: 2 |
| Phillips et al, 2014 [69] | Seward St. Elizabeth Medical Center | 2b | 23 (>80 yr) | Retrospective | Extended | – | 19 | – | – | |
| Raza et al, 2014 [70] | Roswell Park Cancer Institute | 2b | 99 | Retrospective | – | – | 20.7 | 36 | – | – |
| Yuh et al, 2014 [58] | City of Hope Cancer Center | 2b | 162 | Retrospective | Extended | – | 28 | 23 | – | – |
| Total | 19.3 | 23 |
IRCC = International Robotic Cystectomy Consortium; LND = lymph node dissection; RCT = randomized controlled trial; UNC = University of North Carolina.
The lymph node yield from all series was 19 (range: 3–55). Initial descriptions using a standard template of dissection achieved yields of 18 lymph nodes [59] . Number of lymph nodes recovered with an ELND ranged from 11 to 55. Abaza et al adopted a robotic template similar to the open technique, including external iliac, obturator, hypogastric, common iliac, and presacral up to the aortic bifurcation; the mean lymph node yield was 37.5 (SD: 13.2), demonstrating that lymph node counts could mirror those of open dissection if the same template was followed [88] . In a study of open completion LND after robot-assisted ELND in 11 men, Davis et al removed only an additional 4 lymph nodes with an open approach after 43 were removed with robot assistance [34] . Time of LND was rarely reported, although it ranged from 44 min in standard LND to 117 min in ELND[8] and [34]. The lymph node–positive rate was 22%. In series with >20 RARCs, lymph node–positive rates ranged from 6% to 42%. Reports of vascular injuries were rare, and lymphocele rates were 0–9%.
Table 2 summarizes the studies assessing the effects of patient characteristics and particular surgical aspects on lymph node yields in RARC series. Cumulative analysis from the International Robotic Cystectomy Consortium (IRCC) with respect to lymphadenectomy in 437 patients found a median of 17 lymph nodes removed, with a 20% node-positivity rate [23] . Patient age and sex did not affect the performance of lymphadenectomy. In a different series, increasing body mass index (BMI) did not appear to negatively affect lymph node yield, with >20 lymph nodes removed in normal, overweight, and obese patients [46] .
Table 2 Impact of patient characteristics and surgical aspects on lymph node yield in robot-assisted radical cystectomy series
| Reference | Institution | IDEAL stage |
Cases | Study design | Extension of LND | Operative time, min | Retrieved nodes, no. | pN+, % |
|---|---|---|---|---|---|---|---|---|
| Patient BMI | ||||||||
| Poch et al, 2012 [46] | Roswell Park Cancer Institute | 2b | 56 | Retrospective | – | – | 25 | 16 |
| BMI <25: 14 | 22 | 7 | ||||||
| BMI 25 to <30: 21 | 23 | 14 | ||||||
| BMI ≥30: 21 | 20 | 24 | ||||||
| Case volume | ||||||||
| Guru et al, 2008 [59] | Roswell Park Cancer Institute | 2a | 1–12 | Prospective | Extended | 46 | 33% >13 | 8 |
| 13–24 | 44 | 66% >13 | 33 | |||||
| 25–36 | 41 | 83% >13 | 25 | |||||
| 37–47 | 43 | 72% >13 | 64 | |||||
| 48–58 | 56 | 91% >13 | 18 | |||||
| Pruthi et al, 2008 [60] | University of North Carolina | 2b | 50 | Retrospective | Standard | – | 19 | – |
| 1–10 | 21 | |||||||
| 11–20 | 19 | |||||||
| 21–30 | 20 | |||||||
| 31–40 | 17 | |||||||
| 41–50 | 20 | |||||||
| Richards et al, 2011 [38] | Wake Forest University | 2b | 60 | Retrospective | Extended | – | 17 | 30 |
| 1–20 | 17 | |||||||
| 21–40 | 19.1 | |||||||
| 41–60 | 14.4 | |||||||
| Schumacher et al, 2011 [39] | Karolinska Institute | 2b | 45 | Retrospective | Standard 49% | – | 22.5 | – |
| Extended 31% | ||||||||
| 1–15 | Standard 40% | |||||||
| Extended 7% | ||||||||
| 16–30 | Standard 47% | |||||||
| Extended 53% | ||||||||
| 31–45 | Standard 60% | |||||||
| Extended 33% | ||||||||
| Prior RARP experience | ||||||||
| Hayn et al, 2010 [61] | IRCC | 2b | 496 | Retrospective | – | – | 17.8 | – |
| ≤50 previous RARP: 83 | 13.7 | |||||||
| 51–100 previous RARP: 187 | 19.8 | |||||||
| 101–150 previous RARP: 176 | 19.6 | |||||||
| >150 previous RARP: 50 | 11.8 * | |||||||
* Statistically significant.
BMI = body mass index; RCC = International Robotic Cystectomy Consortium; LND = lymph node dissection; RARP = robot-assisted radical prostatectomy.
It is interesting to note that in single-institution series, Richards et al [38] , Schumacher et al [39] , Guru et al [59] , and Pruthi et al [60] did not find higher lymph node yields with increasing sequential case numbers. However, in the IRCC, performance of lymphadenectomy was positively correlated with surgeon and institution volume but was reduced in patients with more advanced disease (pT4 stage), which may reflect operative avoidance of bulky nodal tissue.
Table 3 summarizes the occurrence of PSMs reported in the RARC series. The reported PSM rates were 5.6% (range: 0–26%). However, in series of >100 patients, margin rates ranged between 4% and 9%[48] and [53]. PSMs were reported in 1–1.5% of patients with pT2 disease and 0–25% of patients with pT3 and higher disease. PSM rates from the IRCC in 939 cases were 9% [53] .
Table 3 Positive surgical margins in robot-assisted radical cystectomy series
| Reference | Institution | IDEAL stage | Cases, no. | Study design | Pathologic stage, % | Overall PSM rate, % | PSM location | PSM rate, % | ||
|---|---|---|---|---|---|---|---|---|---|---|
| ≤pT2 | ≥pT3 | ≤pT2 | ≥pT3 | |||||||
| Menon et al, 2003 [4] | Henry Ford Hospital | 1 | 17 | Retrospective | – | – | 0 | – | 0 | 0 |
| Yohannes et al, 2003 [62] | Creighton University | 1 | 2 | Retrospective | 0 | 100 | 0 | – | 0 | 0 |
| Menon et al, 2004 [7] | Henry Ford Hospital | 1 | 3 female | Retrospective | 66 | 33 | 0 | – | 0 | 0 |
| Rhee et al, 2006 [84] | University of Virginia | 1 | 7 | Retrospective | 43 | 57 | 0 | – | 0 | 0 |
| Guru et al, 2007 [8] | Roswell Park Cancer Institute | 1 | 20 | Prospective | 40 | 60 | 15 | Prostate: 1 Ureter: 1 Vagina: 1 |
0 | 25 |
| Mottrie et al, 2007 [9] | O.L.V.–Clinic | 2a | 27 | Retrospective | 78 | 22 | 4 | Ureter: 1 | – | – |
| Pruthi et al, 2008 [71] | UNC | 2a | 20 | Retrospective | 70 | 20 | 0 | – | 0 | 0 |
| Hemal et al, 2008 [10] | All India Institute of Medical Sciences | 1 | 6 | Retrospective | 67 | 33 | 0 | – | 0 | 0 |
| Lowentritt et al, 2008 [11] | Tulane University | 2a | 4 | Retrospective | 25 | 75 | 0 | – | 0 | 0 |
| Murphy et al, 2008 [12] | Guy's Hospital | 2a | 23 | Retrospective | 74 | 17 | 0 | – | 0 | 0 |
| Park et al, 2008 [13] | Yonsei | 2a | 4 | Retrospective | 50 | 50 | 0 | – | 0 | 0 |
| Pruthi et al, 2008 [14] | UNC | 2a | 12 female | Retrospective | 58 | 25 | 0 | – | 0 | 0 |
| Pruthi et al, 2008 [15] | UNC | 2b | 50 | Retrospective | 66 | 14 | 0 | – | 0 | 0 |
| Pruthi et al, 2009 [19] | UNC | 2b | 50 10 female 40 male |
Retrospective | 66 50 70 |
18 30 15 |
0 | – | 0 | 0 |
| Wang et al, 2008 [72] | Cornell | 2b | 33 | Retrospective | 72 | 28 | 6 | Perivesical fat: 2 | 0 | 22 |
| Woods et al, 2008 [16] | Mayo Arizona Tulane University |
2b | 27 | Multi-institutional | – | – | 7 | – | 0 | – |
| Yuh et al, 2008 [17] | Roswell Park Cancer Institute | 2a | 54 | Retrospective | 44 | 56 | 13 | – | 0 | 23 |
| Gamboa et al, 2009 [18] | University of California, Irvine | 2a | 41 | Retrospective | – | – | 5 | – | 0 | – |
| Palou Redorta et al, 2009 [20] | Barcelona Autonomous University | 2a | 9 | Retrospective | 66 | 33 | 11 | – | – | – |
| Yuh et al, 2009 [21] | Roswell Park Cancer Institute | 2b | 73 | Retrospective | 45 | 55 | 10 | – | 0 | 18 |
| Guru et al, 2010 [22] | Roswell Park Cancer Institute | 2a | 20 | Prospective | 62 | 38 | 4 | – | 0 | 9 |
| Hayn et al, 2010 [61] | IRCC | 2b | 482 | Multi-institutional | 64 | 36 | 7 | – | – | – |
| Hellenthal et al, 2010 [68] | IRCC | 2b | 513 | Multi-institutional | 64 | 36 | 7 | – | 1.50 | 17 |
| Kang et al, 2010 [25] | Multicenter | 2b | 104 | Multi-institutional | 70 | 30 | 5 | – | – | – |
| Kasraeian et al, 2010 [26] | Montsouris Institute | 2a | 9 | Retrospective | 44 | 66 | 0 | – | – | – |
| Kauffman et al, 2011 [27] | Cornell | 2b | 85 | Retrospective | 64 | 36 | 6 | – | 0 | 16 |
| Kwon et al, 2010 [28] | Kyungpook National University | 2a | 17 | Prospective | 59 | 41 | 0 | – | 0 | 0 |
| Martin et al, 2010 [30] | Mayo Arizona Tulane University |
2b | 59 | Multi-institutional | 47 | 53 | – | – | – | – |
| Ng et al, 2010 [73] | Cornell | 2b | 83 | Retrospective | 61 | 39 | 7 | – | 0 | 19 |
| Nix et al, 2010 [74] | UNC | 3 | 21 | RCT | 67 | 14 | 0 | – | 0 | 0 |
| Pruthi et al, 2010 [31] | UNC | 2b | 100 | Retrospective | 67 | 13 | 0 | – | 0 | 0 |
| Richards et al, 2010 [75] | Wake Forest University | 2b | 35 | Retrospective | 60 | 40 | 3 | – | – | – |
| Akbulut et al, 2011 [32] | Ankara Ataturk Training and Research Hospital | 2a | 12 | Not reported | 58 | 42 | 0 | – | 0 | 0 |
| Canda et al, 2012 [33] | Ankara Ataturk Rraining and Research Hospital | 2a | 27 | Not reported | 56 | 44 | 4 | – | 0 | 4 |
| Davis et al, 2011 [34] | University of Texas M.D. Anderson Cancer Center | 2a | 11 | Retrospective | 92 | 8 | 0 | – | – | – |
| Jonsson et al, 2011 [35] | Karolinska Institute | 2b | 45 | Prospective | 78 | 22 | 2 | – | 0 | 10 |
| Khan et al, 2011 [36] | Guy's Hospital | 2a | 50 | Prospective | 72 | 28 | 2 | – | 0 | 7 |
| Manoharan et al, 2011 [37] | University of Miami | 2a | 14 | Retrospective | – | – | 0 | – | 0 | 0 |
| Martin et al, 2011 [76] | Mayo Arizona | 2b | 19 | Retrospective | 42 | 58 | – | – | – | – |
| Richards et al, 2011 [38] | Wake Forest University | 2b | 60 | Retrospective | 63 | 37 | 10 | – | – | – |
| Schumacher et al, 2011 [39] | Karolinska Institute | 2b | 45 | Retrospective | 78 | 22 | 2 | Ureter: 1 | 0 | 10 |
| Shah et al, 2011 [40] | Ohio State University | 2b | 30 | Retrospective | 65 | 35 | 7 | – | 0 | 22 |
| Cho et al, 2012 [42] | Hallym University College of Medicine | 2b | 35 | Retrospective | 86 | 14 | 3 | – | – | – |
| Goh et al, 2012 [43] | Keck School of Medicine, University of Southern California, Los Angeles | 2a | 15 | Prospective | 67 | 33 | 0 | – | – | – |
| Lau et al, 2012 [44] | City of Hope Cancer Center | 2b | 23 (aged >80 yr) | Retrospective | 61 | 39 | 13 | Ureter: 1 | – | – |
| Mmeje et al, 2013 [45] | Mayo Arizona | 2b | 50 | Multi-institutional | 34 | 66 | 2 | – | – | – |
| UNC | ||||||||||
| Poch et al, 2012 [46] | Roswell Park Cancer Institute | 2b | 56 | Retrospective | 55 | 45 | – | – | – | – |
| Richards et al, 2012 [77] | Wake Forest University | 2b | 20 (aged >75 yr) | Retrospective | 60 | 40 | 5 | – | – | – |
| Saar et al, 2013 [47] | Saarland University | 2b | 62 | Retrospective | 64 | 36 | 2 | – | – | – |
| Smith et al, 2012 [48] | Mayo Arizona | 2b | 227 | Multi-institutional | – | – | 2 | – | – | – |
| UNC, Tulane University | ||||||||||
| Styn et al, 2012 [78] | University of Michigan | 2b | 50 | Retrospective | 60 | 40 | 2 | – | – | – |
| Sung et al, 2012 [79] | Samsung Medical Center | 2b | 35 | Retrospective | 43 | 57 | – | – | – | – |
| Treiyer et al, 2012 [49] | Saarland University | 2b | 91 | Retrospective | 67 | 33 | 2 | Urethra: 1 Prostate: 1 | – | – |
| Tsui et al, 2012 [50] | Chang Gung Memorial | 2a | 8 | Retrospective | 75 | 25 | 0 | – | 0 | 0 |
| Yuh et al, 2012 [51] | City of Hope Cancer Center | 2b | 196 | Retrospective | 64 | 36 | 4 | – | – | – |
| Azzouni et al, 2013 [63] | Roswell Park Cancer Institute | 2b | 100 | Retrospective | 35 | 65 | 4 | – | – | – |
| Collins et al, 2013 [52] | Karolinska Institute | 2b | 113 | Prospective | 75 | 25 | 5 | Ureter: 1 | 1 | 18 |
| Johar et al, 2013 [53] | Multicenter | 2b | 939 | Retrospective | 49 | 51 | 9 | – | – | – |
| Maes et al, 2013 [80] | Metro Health Hospital | 2b | 14 | Retrospective | 43 | 57 | 21 | – | – | – |
| Marshall et al, 2013 [54] | IRCC | 2b | 765 | Multi-institutional | 59 | 41 | – | – | – | – |
| Musch et al, 2014 [81] | Klinikin Essen–Mitte | 2b | 100 | Prospective | 61 | 39 | 2 | – | – | – |
| Nazmy et al, 2014 [55] | City of Hope Cancer Center | 2b | 209 | Retrospective | 65 | 35 | 3 | – | – | – |
| Nepple et al, 2013 [82] | Washington University | 2b | 36 | Retrospective | 53 | 47 | 6 | – | 0 | 12 |
| Parekh et al, 2013 [83] | University of Texas Health Sciences Center at San Antonio | 3 | 20 | RCT | 50 | 50 | 5 | – | 0 | 10 |
| Tyritzis et al, 2013 [56] | Karolinska Institute | 2b | 70 | Retrospective | 86 | 14 | 1.5 | Ureter: 1 | 0 | 10 |
| Xylinas et al, 2013 [57] | Cornell | 2b | 175 | Retrospective | 65 | 35 | 5 | – | – | – |
| Phillips et al, 2014 [69] | Seward St. Elizabeth Medical Center | 2b | 23 (aged >80 yr) | Retrospective | 30 | 70 | 26 | – | – | – |
| Raza et al, 2014 [70] | Roswell Park Cancer Institute | 2b | 99 | Retrospective | 48 | 52 | 8 | – | – | |
| Yuh et al, 2014 [58] | City of Hope Cancer Center | 2b | 162 | Retrospective | 67 | 33 | 4 | – | – | – |
| Total | 60 | 40 | 5.6 | |||||||
IRCC = International Robotic Cystectomy Consortium; PSM = positive surgical margin; RCT = randomized controlled trial; UNC = University of North Carolina.
Table 4 summarizes the studies assessing the effects of patient characteristics and particular surgical aspects on PSM rates in RARC series. Notably, Richards et al [38] , Schumacher et al [39] , and the IRCC [68] did not demonstrate decreasing surgical margin rates with sequential case number. In a study of the role of previous robot-assisted radical prostatectomy (RARP) experience on RARC outcomes, there was a trend toward increased positive margins with increasing RARP volumes, but it did not reach statistical significance (p = 0.089) [61] . The authors chiefly attributed this situation to the performance of RARC on patients with higher risk (higher than T3) disease. One study reported that PSMs occurred only in the overweight or obese patients, although pT4 rates were much higher in those patients (26% vs 7%) [65] .
Table 4 Predictors of positive surgical margins in robot-assisted radical cystectomy series
| Reference | Institution | IDEAL stage | Cases | Study design | Pathologic stage, % | Overall PSM rate, % | PSM rate, % | ||
|---|---|---|---|---|---|---|---|---|---|
| ≤pT2 | ≥pT3 | ≤pT2 | ≥pT3 | ||||||
| Case volume | |||||||||
| Guru et al, 2008 [59] | Roswell Park Cancer Institute | 2a | 1–12 | Prospective | 33 | 66 | 17 | – | – |
| 13–24 | 58 | 42 | 25 | ||||||
| 25–36 | 50 | 50 | 0 | ||||||
| 37–47 | 46 | 54 | 9 | ||||||
| 48–58 | 64 | 36 | 0 | ||||||
| Hayn et al, 2011 [64] | Roswell Park Cancer Institute | 2a | 1–50 | Prospective | 51 | 49 | 8 | – | – |
| 51–100 | 12 | ||||||||
| 101–164 | 6 | ||||||||
| Richards et al, 2011 [38] | Wake Forest University | 2b | 60 | Retrospective | 63 | 37 | 10 | – | – |
| 1–20 | 55 | 45 | 5 | ||||||
| 21–40 | 70 | 30 | 5 | ||||||
| 41–60 | 65 | 35 | 20 | ||||||
| Schumacher et al, 2011 [39] | Karolinska Institute | 2b | 45 | Retrospective | 78 | 22 | 2 | 0 | 10 |
| 1–15 | 87 | 13 | 0 | ||||||
| 16–30 | 67 | 33 | 7 | ||||||
| 31–45 | 80 | 20 | 0 | ||||||
| Azzouni et al, 2013 [63] | Roswell Park Cancer Institute | 2b | 100 | Retrospective | 35 | 65 | 4 | – | – |
| 1–25 | 36 | 64 | 4 | ||||||
| 26–50 | 40 | 60 | 4 | ||||||
| 51–75 | 44 | 56 | 4 | ||||||
| 76–100 | 20 | 80 | 4 | ||||||
| Previous RARP experience | |||||||||
| Hayn et al, 2010 [61] | IRCC | 2b | 482 | Retrospective | 64 | 36 | 7 | – | – |
| ≤50 previous RARP: 83 | 68 | 32 | 4 | ||||||
| 51–100 previous RARP: 173 | 76 | 24 | 5 | ||||||
| 101–150 previous RARP: 168 | 54 | 46 | 9.5 | ||||||
| >150 previous RARP: 48 | 42 | 58 | 12.5 | ||||||
| Patient BMI | |||||||||
| Butt et al, 2008 [65] | Roswell Park Cancer Institute | 2a | BMI <25: 14 | Retrospective | 64 | 36 | 0 | 0 | – |
| BMI 25–29: 18 | 28 | 72 | 28 | ||||||
| BMI ≥30: 17 | 42 | 58 | 6 | ||||||
| Poch et al, 2012 [46] | Roswell Park Cancer Institute | 2b | 56 | Retrospective | 55 | 45 | – | – | – |
| BMI <25: 14 | 50 | 50 | |||||||
| BMI 25 to <30: 21 | 57 | 43 | |||||||
| BMI ≥30: 21 | 52 | 48 | |||||||
| Intracorporeal vs extracorporeal diversion | |||||||||
| Kang et al, 2012 [85] | Korea University School of Medicine | 2a | 38 extracorporeal diversion | Retrospective | 76 | 24 | 2.5 | – | – |
| 4 intracorporeal diversion | 100 | 0 | 0 | ||||||
BMI = body mass index; IRCC = International Robotic Cystectomy Consortium; PSM = positive surgical margin; RARP = robot-assisted radical prostatectomy.
Table 5 summarizes the oncologic outcomes of current RARC publications. Neoadjuvant chemotherapy use was reported in 0–31% of patients. Adjuvant chemotherapy use was reported in 4–29% of patients.
Table 5 Survival outcomes in robot-assisted radical cystectomy series
| Reference | Institution | IDEAL stage | Cases, no. | Study design | Follow-up, mo | Neoadjuvant chemotherapy, % | Adjuvant chemotherapy, % | DFS estimates, % | CSS estimates, % | OS estimates, % | ||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 1 yr | 3 yr | 5 yr | 1 yr | 3 yr | 5 yr | 1 yr | 3 yr | 5 yr | ||||||||
| Pruthi et al, 2008 [15] | UNC | 2b | 50 | Retrospective | 13.2 | 0 | 22 | – | – | – | 94 (13 mo) | – | – | 90 (13 mo) | – | – |
| Murphy et al, 2008 [12] | Guy's Hospital | 2a | 23 | Retrospective | 17 | 29 | – | 91 (17 mo) | – | – | – | – | – | – | – | – |
| Josephson et al, 2010 [24] | City of Hope Cancer Center | 2b | 58 | Retrospective | 12 | 22 | – | – | 76 (2 yr) | – | – | 76 (2 yr) | – | – | 54 (2 yr) | – |
| Kang et al, 2010 [25] | Multicenter | 2b | 104 | Retrospective | 12 | – | – | 96 | – | – | – | – | – | – | – | – |
| Kauffman et al, 2011 [27] | Cornell | 2b | 85 | Retrospective | 18 | 20 | 12 | 79 | 73 (2 yr) | 88 | 85 (2 yr) | 83 | 79 (2 yr) | |||
| Martin et al, 2010 [30] | Mayo Arizona | 2b | 59 | Multi-institutional | 21 | 17 | – | 82 | 71 | – | – | – | – | 82 | 72 | – |
| Tulane University | ||||||||||||||||
| Pruthi et al, 2010 [31] | UNC | 2b | 100 | Retrospective | 21.2 | 5 | 18 | – | – | – | 94 (21 mo) | – | – | 91 (21 mo) | – | – |
| Canda et al, 2012 [33] | Ankara Ataturk Training and Research Hospital | 2a | 27 | Not reported | 6 | – | 4 | 85 (6 mo) | – | – | 89 (6 mo) | – | – | 72 (6 mo) | – | – |
| Mmeje et al, 2013 [45] | Mayo Arizona | 2b | 50 | Multi-institutional | 41.5 | 12 | 46 | – | 43 | 39 | – | – | – | – | 55 | 45 |
| UNC | ||||||||||||||||
| Treiyer et al, 2012 [49] | Saarland University | 2b | 91 | Retrospective | 15 | 0 | – | – | – | – | 94 (15 mo) | – | – | 93 (15 mo) | – | – |
| Collins et al, 2013 [52] | Karolinska Institute | 2b | 113 | Prospective | 25 | 31 | – | – | – | – | – | 81 | 67 | – | 80 | 66 |
| Khan et al, 2013 [66] | Guy's & St. Thomas Hospital | 1 | 14 | Prospective | 84 | 28 | 14 | 50 | 75 | 64 | ||||||
| Nepple et al, 2013 [82] | Washington University | 2b | 36 | Retrospective | 12 | 6 | – | – | 67 (2 yr) | – | – | 75 (2 yr) | – | – | 68 (2 yr) | – |
| Snow-Lisy et al, 2014 [67] | Cleveland Clinic | 2b | 17 | Retrospective | 67 | – | – | – | – | – | – | – | 69 | – | – | 39 |
| Tyritzis et al, 2013 [56] | Karolinska Institute | 2b | 70 | Retrospective | 30.3 | 24 | – | – | 81 (2 yr) | – | – | 89 (2 yr) | – | – | 89 (2 yr) | – |
| Xylinas et al, 2013 [57] | Cornell | 2b | 175 | Retrospective | 37 | 19 | – | 67 | 63 | – | 68 | 66 | – | – | – | |
| Raza et al, 2014 [70] | Roswell Park Cancer Institute | 2b | 99 | Retrospective | 73.9 | 6 | 29 | – | – | 53 | – | – | 68 | – | – | 42 |
| Yuh et al, 2014 [58] | City of Hope Cancer Center | 2b | 162 | Retrospective | 52 | 23 | – | – | 76 | 74 | – | 83 | 80 | – | 61 | 54 |
CSS = cancer-specific survival; DFS = disease-free survival; OS = overall survival; UNC = University of North Carolina.
Several studies further analyzed the use of adjuvant chemotherapy after RARC. General indications for selecting patients for adjuvant chemotherapy included pathologic stage pT3–4 or node-positive disease. Pruthi et al described the use of adjuvant chemotherapy in 18 of 100 RARC patients, with mean time to chemotherapy initiation at approximately 7 wk, which was faster than the authors’ historical time to chemotherapy in open cystectomy of 10 wk [31] . In a randomized trial of RARC (n = 21) compared with ORC (n = 20), 7 wk was also the mean time to initiation of chemotherapy after RARC [74] . In one analysis of patients with node-positive disease at the time of RARC, 46% received adjuvant chemotherapy [45] .
Survival represents the gold standard with respect to evaluating effectiveness and risks of treatment; however, RARC reports with 5-yr outcomes have become available only recently. Data remain limited for assessing long-term outcomes, patterns of recurrence, and means for predicting survival. The role of adjuvant treatments after RARC is also poorly defined.
Series detailing cancer control outcomes had a mean follow-up between 6 and 84 mo ( Table 5 ), although only 6 of 18 series (33%) reported a mean follow-up >36 mo. At 1, 2, 3, and 5 yr, DFS was 79–96%, 67–81%, 67–76%, and 53–74%, respectively; CSS was 88–94%, 75–89%, 68–83%, and 66–80%, respectively; and OS was 82–90%, 54–89%, 61–80%, and 39–66%, respectively. In the series with longest follow-up, Khan et al described only 14 patients with ≥5 yr of follow-up, showing DFS of 50%, CSS of 75%, and OS of 64% [66] .
Several series reported on adverse oncologic outcomes associated with increased pathologic stage or lymph node involvement[58] and [93]. In a series of 162 patients with urothelial carcinoma, Yuh et al found that 5-yr survival was worse with higher pathologic stage or lymph node positivity (p < 0.01). Patients with a lymph node density of 1–10% (defined as number of positive nodes divided by number of total nodes) had DFS, CSS, and OS of 34%, 49%, and 31%, respectively, whereas patients with lymph node density >10% had further reduced survival of 30%, 38%, and 20%, respectively. Predictors of DFS were lymph node density, pathologic stage, and age-adjusted Charlson comorbidity index, whereas the same measures plus receipt of transfusion were predictive for OS [58] . Similarly, in an analysis of 99 patients with follow-up >5 yr, pathologic stage and lymph node positivity were independent predictors of DFS, CSS, and OS, whereas positive margin status and Charlson comorbidity index predicted worse OS and CSS [70] .
In series with median follow-up of >36 mo, rates of local recurrence without distant disease ranged between 0% (n = 15) and 9% (n = 99)[57], [58], [66], [67], and [70]. No port-site recurrences occurred in these series. Xylinas et al examined 175 patients with a median follow-up of 37 mo, showing recurrence of disease in 29%. Of these patients, 8 (5%) had local recurrence alone, 11 had local and distant metastases, and 32 had distant metastases alone [57] .
In an analysis of patients with positive lymph nodes (n = 50) at the time of PLND, median time to recurrence was 10 mo after RARC [45] . Estimated OS at 36 and 60 mo was 55% and 45%, respectively, with recurrence-free survival at 36 and 60 mo of 43% and 39%, respectively. Similarly, Tyritzis et al reported recurrence-free survival of 34% and OS of 63% after 24 mo in node-positive patients [56] .
Table 6 presents the RARC series reporting on continence outcomes. Although functional outcomes are a major area of study in patients undergoing RARP, a lack of data remains for evaluation after RARC. Worldwide, the number of patients evaluated for continence after orthotopic bladder substitution is <200 from nine reports at the present time. There are also widespread differences in patient selection, methods of data collection, and outcome assessment.
Table 6 Urinary continence rates in robot-assisted radical cystectomy series
| Reference | Institution | IDEAL stage | Cases, no. | Study design | Nerve-sparing surgery, % | Intracorporeal diversion, % | Follow-up, mo | Method of data collection | Continence definition | Continence rate, % | ||
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 3 mo | 6 mo | 12 mo | ||||||||||
| Mottrie et al, 2007 [9] | O.L.V.–Clinic | 2a | 27 | Retrospective | 29 | 0 | 10.2 | – | – | 86 | – | – |
| Murphy et al, 2008 [12] | Guy's Hospital | 2a | 23 | Retrospective | 20 | 0 | 17 | – | – | – | – | 100 D |
| 75 N (17 mo) |
||||||||||||
| Palou Redorta et al, 2009 [20] | Barcelona Autonomous University | 2a | 9 | Retrospective | 100 | 0 | 7 | – | – | 100 D and N (7 mo) |
||
| Canda et al, 2012 [33] | Ankara Ataturk Training and Research Hospital | 2a | 27 | Not reported | 89 | 100 | 6 | – | D: 0–1 safety pads | – | 48 | – |
| N: dry with no protection | 11 | |||||||||||
| Jonsson et al, 2011 [35] | Karolinska Institute | 2b | 36 | Prospective | 55 | 100 | 25 | – | 0–1 pads | – | – | 83 D 66 N |
| Manoharan et al, 2011 [37] | University of Miami | 2a | 14 | Retrospective | – | 0 | – | – | – | 93 D | ||
| 71 N | ||||||||||||
| Torrey et al, 2012 [41] | City of Hope Cancer Center | 2b | 34 | Retrospective | 0 | 0 (all Indiana pouch) | 12.1 | Physician charting | – | – | – | 97 |
| Goh et al, 2012 [43] | Keck School of Medicine, University of Southern California, Los Angeles | 2a | 15 | Prospective | – | 100 | 3 | – | – | 75 | ||
| Tyritzis et al, 2013 [56] | Karolinska Institute | 2b | 70 | Retrospective | 58 BNS | 100 | 12 | Internally validated questionnaire | 0–1 pads | – | D: 77 men, |
D: 88 men, |
| 8 UNS | 40 women N: 54 men, 40 women |
67 women | ||||||||||
| N: 76 men, 76 women |
||||||||||||
BNS = bilateral nerve sparing; D = daytime; N = nocturnal; UNS = unilateral nerve sparing.
Follow-up for continence evaluation varied widely, from 6 to 25 mo. Nerve-sparing procedures were performed in 20–100% of patients. Only three of six series reported using a distinct definition forcontinence, which was generally no pad or one pad (safety) per day. One of the earliest RARC series reported an 86% continent rate (seven of eight men) after 3.5 mo [9] . More recent series published 6-mo continence rates of 48–100% for daytime continence and 11–100% for nighttime continence. At 12 mo after RARC, continence rates ranged from 83% to 100% in men and were 67% in women for daytime continence and 66–76% for nighttime continence.
Using strict definitions fordaytime continence(no or one security pad per day) andnighttime continence(goodindicates dry with no protection,fairindicates dry with one awakening), Canda et al examined 23 patients with intracorporeal Studer pouch. After excluding patients who died or were lost to follow-up, 11 of 15 men (73%) and 0 of 2 women were continent during the daytime. Three of these 17 patients (18%) had good nighttime continence, and 4 (24%) had fair continence [33] .
Only one series has described continence results in patients undergoing RARC and continent cutaneous diversion. Torrey et al examined 34 patients who had RARC and Indiana pouch continent cutaneous diversion and reported 97% continence at a mean follow-up of 20 mo for both daytime and nighttime. One patient continued to experience daytime and nighttime incontinence requiring the use of pads [41] .
Table 7 summarizes the series examining potency outcomes. Similar to continence outcomes, evaluation of erectile function after RARC is not well described. Early reports suggest that erections sufficient for penetration are achievable, although sample sizes were very small and lacked validated objective evaluations. Follow-up was again too short to form definitive conclusions, with only one study reporting outcomes up to 2 yr after RARC. As noted earlier, nerve-sparing procedures were performed in 20–100% of patients. The data recording used International Index of Erectile Function (IIEF) scores in five of seven series. However, only three series provided a clear definition ofpotency[12], [35], and [56].
Table 7 Erectile function in robot-assisted radical cystectomy series
| Reference | Institution | IDEAL stage | Cases, no. | Nerve-sparing surgery, % | Study design | Follow-up, mo | Method of data collection | Potency definition | Potency rate at follow-up |
|---|---|---|---|---|---|---|---|---|---|
| Mottrie et al, 2007 [9] | O.L.V.–Clinic | 2a | 27 | 29 | Retrospective | 10.2 | – | – | 86% |
| Murphy et al, 2008 [12] | Guy's Hospital | 2a | 23 | 20 | Retrospective | 17 | IIEF | IIEF >21 with or without PDE5-I | 75% |
| Palou Redorta et al, 2009 [20] | Barcelona Autonomous University | 2a | 9 | 100 | Retrospective | 7 | – | – | 100% |
| Akbulut et al, 2011 [32] | Ankara Ataturk Training and Research Hospital | 2a | 12 | 82 bilateral 9 unilateral |
Not reported | 7.1 | IIEF | None provided | A single patient with IIEF >18 |
| Canda et al, 2012 [33] | Ankara Ataturk Training and Research Hospital | 2a | 27 | 89 | Not reported | 6 | IIEF | None provided | A single patient with IIEF >18 |
| Jonsson et al, 2011 [35] | Karolinska Institute | 2b | 36 | 55 | Prospective | 25 | IIEF | Adequate for penetration with or without PDE5-I | 41% at 12 mo 75% of patients having nerve sparing |
| Tyritzis et al, 2013 [56] | Karolinska Institute | 2b | 70 | 58 bilateral 8 unilateral |
Retrospective | 12 | IIEF | Adequate for penetration with or without PDE5-I | 63% at 12 mo |
IIEF = International Index of Erectile Function; PDE5-I = phosphodiesterase type 5 inhibitor.
In some early, small series, Mottrie et al [9] and Murphy et al [12] reported sufficient erections in six of seven and three of four men, respectively. Similar to well-described literature on RARP, phosphodiesterase type 5 inhibitors (PDE5-Is) were frequently administered to patients for penile rehabilitation after RARC; however, no comparative data in this setting have demonstrated a benefit.
Several series with intracorporeal neobladder have evaluated erectile function postoperatively, with varying results. In the experience of the Karolinska Institute, 41 of 62 men (66%) underwent nerve-sparing RARC. Of these 41 men, 26 (63%) were potent with or without the use of PDE5-Is after 12 mo [56] . In contrast, Canda et al found IIEF scores >18 in only 1 of 11 preoperatively potent men, although follow-up was shorter (6 mo) [33] .
Table 8 summarizes comparative studies evaluating lymph node yield after ORC, LRC, and RARC. In two randomized studies of ORC compared with RARC, lymph node yields were not statistically different[74] and [83]. Cumulative analyses showed no significant difference in lymph node yield between RARC and ORC (OR: 2.94; 95% CI, −0.28 to 6.15;p = 0.07) ( Fig. 2 ).
Table 8 Comparative studies evaluating lymph node yield after open, laparoscopic, and robot-assisted radical cystectomy
| Comparison | Level of evidence | Reference | Cases, no. | Study design | Extension of LND | Retrieved nodes, no. | pN+, no. (%) | Metastatic nodes, no., median |
|---|---|---|---|---|---|---|---|---|
| ORC vs RARC | 2 | |||||||
| Nix et al, 2010 [74] | 21 RARC | RCT | Standard | 19 | 4 (19) | – | ||
| 20 ORC | 18 | 7 (35) | ||||||
| Parekh et al, 2013 [83] | 20 RARC | RCT | Standard | 17.2 ± 13 | 4 (20) | – | ||
| 20 ORC | 24.2 ± 16.4 | 4 (20) | ||||||
| 3 | ||||||||
| Pruthi et al, 2008 [71] | 20 RARC | Gender matched | Standard | 19 | 2 (10) | – | ||
| 24 ORC | Retrospective | 16 | 5 (21) | |||||
| Wang et al, 2008 [72] | 33 RARC | Nonmatched | Standard | 17 | 19 | – | ||
| 21 ORC | 20 | 34 * | ||||||
| Ng et al, 2010 [73] | 83 RARC | Nonmatched | Standard | 17.9 ± 10.4 | 13 (16) | – | ||
| 104 ORC | 15.7 ± 13.2 | 24 (23) * | ||||||
| Richards et al, 2010 [75] | 35 RARC | Nonmatched | Extended | 16 | 10 (29) | – | ||
| 35 ORC | 15 | 10 (29) | ||||||
| Martin et al, 2011 [76] | 19 RARC | Nonmatched | – | 16 | – | – | ||
| 14 ORC | 13 | |||||||
| Gondo et al, 2012 [92] | 11 RARC | Nonmatched | Extended | 20.7 ± 8.2 | 9 | – | ||
| 15 ORC | 13.8 ± 6.6 * | 13 | ||||||
| Khan et al, 2012 [87] | 48 RARC | Prospective | Extended | 16 | 5 | – | ||
| 52 ORC | 11 | 15 | ||||||
| Richards et al, 2012 [77] | 20 RARC | Nonmatched | Extended | 17 | 7 (35) | – | ||
| 20 ORC (>75 yr) | 15 | 3 (15) | ||||||
| Styn et al, 2012 [78] | 50 RARC | 1:2 by age, sex, clinical stage, diversion | – | 14.3 ± 9.1 | 6 (12) | – | ||
| 100 ORC | 15.2 ± 9.5 | 19 (19) | ||||||
| Sung et al, 2012 [79] | 35 RARC | Nonmatched | Standard | 19.1 ± 8.2 | 9 (26) | – | ||
| 104 ORC | 12.9 ± 9.0 * | 27 (26) | ||||||
| Knox et al, 2013 [86] | 58 RARC | Nonmatched | Extended | 21 | 1 | – | ||
| 84 ORC | 17 | 3 | ||||||
| Maes et al, 2013 [80] | 14 RARC | Nonmatched | Extended | 11.9 | 1 (7) | – | ||
| 14 ORC | 9.5 | 5 (35) | ||||||
| Musch et al, 2014 [81] | 100 RARC | Nonmatched | – | 27.5 ± 11.0 | 20 (20) | – | ||
| 42 ORC | 19.6 ± 8.8 * | 9 (21) | ||||||
| Nepple et al, 2013 [82] | 36 RARC | Nonmatched | Standard | 17 | 8 (22) | – | ||
| 29 ORC | 14 | 7 (24) | ||||||
| 4 | Abaza et al, 2012 [88] | 35 RARC | Nonmatched | Extended | 37.5 ± 13.2 | 12 (34) | 1.5 | |
| 120 ORC | 36.9 ± 14.8 | 36 (30) | 2 | |||||
| LRC vs RARC | ||||||||
| 3 | Khan et al, 2012 [87] | 48 RARC | Prospective | Extended | 16 | 5 | – | |
| 58 LRC | 10 | 10 | ||||||
| 4 | Abraham et al, 2007 [89] | 14 RARC | Nonmatched | 10 extended | 22.3 | 2 (10) | ||
| 20 LRC | 16 extended | 16.5 | 2 (12.5) | |||||
* Statistically significant.
LND = lymph node dissection; LRC = laparoscopic radical cystectomy: ORC = open radical cystectomy; RARC = robot-assisted radical cystectomy; RCT = randomized controlled trial.
Fig. 2 Comparison of lymph node yields following robot-assisted or open radical cystectomy. CI = confidence interval; ORC = open radical cystectomy; RARC = robot-assisted radical cystectomy; SD = standard deviation.
Table 9 summarizes PSM rates in RARC, ORC, and LRC. In two randomized trials comparing RARC and ORC, Nix et al and Parekh et al did not show any increase in positive margins with RARC[74] and [83]. Cumulative analyses showed no significant difference in rates of surgical margins between RARC and ORC (5% and 7%, respectively; OR: 0.71; 95% CI, 0.46–1.1;p = 0.13) ( Fig. 3 ). In two comparative nonrandomized studies between RARC and LRC, no significant differences in PSM rates were detected (p = 0.86)[87] and [89].
Table 9 Comparative studies evaluating positive surgical margins after open, laparoscopic, and robot-assisted radical cystectomy
| Comparison | Level of evidence | Reference | Cases, no. | Pathologic stage, % | Overall PSM, no. (%) | PSM in pT2 cancer | |
|---|---|---|---|---|---|---|---|
| pT2 | pT3 | ||||||
| ORC vs RARC | 2b | ||||||
| Nix et al, 2010 [74] | 21 RARC | 67 | 14 | 0 | 0 | ||
| 20 ORC | 40 | 25 | 0 | 0 | |||
| Parekh et al, 2013 [83] | 20 RARC | 50 | 50 | 1 (5) | 0 | ||
| 20 ORC | 65 | 35 | 1 (5) | 0 | |||
| 3 | |||||||
| Rhee et al, 2006 [84] | 7 RARC | 86 | 14 | 0 | 0 | ||
| 23 ORC | 43 | 57 | 0 | 0 | |||
| Galich et al, 2006 [90] | 13 RARC | 54 | 46 | 0 | – | ||
| 24 ORC | 37 | 63 | 3 (12) | ||||
| Pruthi et al, 2007 [71] | 20 RARC | 78 | 22 | 0 | 0 | ||
| 24 ORC | 63 | 37 | 0 | ||||
| Wang et al, 2008 [72] | 33 RARC | 72 | 28 | 2 (6) | – | ||
| 21 ORC | 43 | 57 | 3 (14) | ||||
| Ng et al, 2010 [73] | 83 RARC | 61 | 39 | 6 (7) | 0 | ||
| 104 ORC | 58 | 42 | 9 (9) | 0 | |||
| Richards et al, 2010 [75] | 35 RARC | 60 | 40 | 1 (3) | – | ||
| 35 ORC | 57 | 43 | 3 (9) | ||||
| Martin et al, 2011 [76] | 19 RARC | 42 | 58 | – | – | ||
| 14 ORC | 93 | 7 | |||||
| Gondo et al, 2012 [92] | 11 RARC | 91 | 9 | 1 (9) | – | ||
| 15 ORC | 53 | 47 | 2 (13) | ||||
| Khan et al, 2012 [87] | 48 RARC | 75 | 25 | 0 | – | ||
| 52 ORC | 50 | 50 | 6 (10) | ||||
| Richards et al, 2012 [77] | 20 RARC | 60 | 40 | 1 (5) | – | ||
| 20 ORC (>75 yr) | 50 | 50 | 2 (10) | ||||
| Styn et al, 2012 [78] | 50 RARC | 60 | 40 | 1 (2) | – | ||
| 100 ORC | 72 | 28 | 1 (1) | ||||
| Sung et al, 2012 [79] | 35 RARC | 43 | 57 | – | – | ||
| 104 ORC | 38 | 62 | |||||
| Kader et al, 2013 [91] | 100 RARC | 58 | 42 | 12 (12) | – | ||
| 100 ORC | 53 | 47 | 11 (11) | ||||
| Knox et al, 2013 [86] | 58 RARC | 66 | 34 | 4 (7) | – | ||
| 84 ORC | 43 | 57 | 7 (8) | ||||
| Maes et al, 2013 [80] | 14 RARC | 43 | 57 | 3 (21) | – | ||
| 14 ORC | 57 | 43 | 2 (14) | ||||
| Musch et al, 2013 [81] | 100 RARC | 61 | 39 | 2 (2) | – | ||
| 42 ORC | 57 | 43 | 1 (2) | ||||
| Nepple et al, 2013 [82] | 36 RARC | 53 | 47 | 2 (6) | 0 | ||
| 29 ORC | 58 | 42 | 2 (7) | 0 | |||
| 4 | Abaza et al, 2012 [88] | 35 RARC | 60 | 23 | 2 (6) | 0 | |
| 120 ORC | 45 | 42 | 8 (7) | 0 | |||
| LRC vs RARC | |||||||
| 3 | Khan et al, 2012 [87] | 48 RARC | 75 | 25 | 0 | – | |
| 58 LRC | 57 | 43 | 2 (4) | ||||
| 4 | Abraham et al, 2007 [89] | 14 RARC | – | – | 1 (7) | 0 | |
| 20 LRC | 0 | – | |||||
LRC = laparoscopic radical cystectomy; ORC = open radical cystectomy; PSM = positive surgical margin; RARC = robot-assisted radical cystectomy.
Fig. 3 Comparison of positive surgical margin rates following robot-assisted or open radical cystectomy. CI = confidence interval; M-H = Mantel-Haenszel; ORC = open radical cystectomy; RARC = robot-assisted radical cystectomy.
Table 10 summarizes series that emphasized early oncologic comparisons for RARC, LRC, and ORC, though interpretation should be cautious with small series of shorter follow-up and potential bias of patient selection. A nonrandomized comparison of ORC (n = 52) with RARC (n = 48) with a follow-up of 38 mo showed disease-specific survival of 69% in the ORC group compared with 79% in the RARC group [87] . A series by Nepple et al showed similar estimates in DFS, CSS, and OS, although patients were not matched [82] .
Table 10 Comparative studies evaluating recurrence-free, cancer-specific, and overall survival estimates after open, laparoscopic, and robot-assisted radical cystectomy
| Comparison | Level of evidence | Reference | Cases, no. | Study design | Follow–up, mo | Neoadjuvant chemotherapy, % | DFS estimates, % | CSS estimates, % | OS estimates, % |
|---|---|---|---|---|---|---|---|---|---|
| ORC vs RARC | 3 | ||||||||
| Khan et al, 2012 [87] | 48 RARC | Prospective | 38 | – | – | 79 | – | ||
| 52 ORC | 69 | ||||||||
| Nepple et al, 2013 [82] | 36 RARC | Nonmatched | 12 | 6 | 67 (2 yr) | 75 (2 yr) | 68 (2 yr) | ||
| 29 ORC | 14 | 58 (2 yr) | 63 (2 yr) | 63 (2 yr) | |||||
| LRC vs RARC | 3 | ||||||||
| Khan et al, 2012 [87] | 48 RARC | Prospective | 38 | – | – | 79 | – | ||
| 58 LRC | 93 |
CSS = cancer-specific survival; DFS = disease-free survival; LRC = laparoscopic radical cystectomy; ORC = open radical cystectomy; OS = overall survival; RARC = robot-assisted radical cystectomy.
Our systematic review sought to identify and report the current state of the literature for RARC with regard to pathologic, oncologic, and functional outcomes. Various oncologic parameters, including pathologic findings and postoperative survival rates, were examined. With regard to nodal dissection, robotic ELND achieves a similar nodal yield to open ELND when performed by experienced surgeons. Nearly all RARC series reported nodal yields >15. With regard to margin rates, most series reported PSM rates of <10%, with rates of approximately 1% in pT2 disease. Although the IRCC (n = 513) reported a very high positive margin rate of 39% in pT4 patients [68] , other authors have reported rates similar to those noted in ORC series.
Although these immediate pathologic variables may act as surrogates for quality of resection, long-term survival outcomes must be analogous to those of ORC for RARC to be a viable surgical option. Currently, oncologic data are immature, and adequate comparative studies of RARC and ORC are nonexistent. In a few analyses measuring CSS and OS at 5 yr postoperatively, results appear similar to those reported in ORC; however, larger numbers and longer follow-up are needed for adequate comparison. At present, data reporting functional analysis of continence and potency recovery after RARC are inadequate to compare RARC reliably with ORC.
PLND, in conjunction with radical cystectomy, provides a staging benefit as well as a possible advantage for survival in retrospective studies. Stein et al examined 1054 patients treated with radical cystectomy and PLND with a 24% node-positive rate; these patients experienced 5- and 10-yr recurrence-free survival of 35% and 34%, respectively [2] . Although prospective validation is necessary, Leissner et al suggested that ELND improved outcomes in both low-volume node-positive and node-negative patients with greater number of lymph nodes removed [93] . The true survival benefit of ELND must be proven in a prospective fashion to overcome the Will Rogers phenomenon of apparent improved survival that results from stage migration with more thorough dissection.
Early critical concerns of RARC involved whether LND could be performed robotically with the same quality as during ORC. This review suggests that thorough robotic ELND dissection at the time of RARC is possible following a similar template as is performed during ORC. Half of current RARC series describe an extended template dissection, with the average number of lymph nodes removed between 11 and 55. In a small study of open completion LND after robotic LND, only four additional lymph nodes were recovered [34] . Although few series described the time necessary to perform a complete robotic LND, some authors described operative times approaching 2 h for the node dissection alone, suggesting that robotic LND may lengthen operative time. Further study is necessary to determine whether the LND segment of RARC is significantly longer compared with open LND. Complications specific to LND—particularly vascular injuries—were rare, as were lymphoceles, with an incidence <10%. However, complication rates may often be underreported, as reporting guidelines lack standardization.
Assessment of RARC lymph node yields as related to patient characteristics (eg, BMI) or surgeon characteristics (eg, prior RARP experience) has not shown a specific association. Similar to the ORC literature, Bochner et al reported that only extent of LND was associated with lymph node yield when examining variables such as receipt of neoadjuvant chemotherapy, pathologic stage, surgeon, and pathologist [94] . Although performance of LND was associated with higher surgeon volume, analysis of several learning curve evaluations did not find increases in lymph node yield with increasing case number. This result may seem counterintuitive, but it may be that these experienced robotic surgeons were able to translate surgical technique from RARP and PLND and thus reduce the number of cases needed to reach stable lymph node yields. Instead, a reduction in LND time could occur with experience, although it has not been specifically examined. In the IRCC database, patients with pT4 disease had lower nodal yields, possibly related to more difficult dissection or to RARC being performed for palliative intent.
PSM at cystectomy is a measure of disease burden and a predictor of outcome. In a previous study of 1589 patients who underwent radical cystectomy at Memorial Sloan Kettering Cancer Center, the positive margin rate was 4.2%. Risk factors for PSMs were female sex, higher pathologic stage, vascular invasion, mixed histology, and lymph node involvement. Patients with PSMs had a 5-yr CSS of only 32% [95] . In another large multi-institutional analysis of 4400 ORC patients, the incidence of PSMs was 6.3% [96] . A potential challenge of RARC and limitation of current robotic technology is in treating bulkier tumors because of the lack of tactile feedback.
The present systematic review demonstrates that PSMs are uncommon in RARC series and appropriately rare for pT2 disease. No significant difference was found when comparing the surgical margin rate between RARC and ORC. The high variability of positive margins across studies, between 0% and 26%, suggests significant heterogeneity in cancer characteristics, patient selection, and surgical technique and experience, among other variables. From the systematic review, the weighted average of positive margins in RARC series was 5.6%, which is comparable to the large open series cited earlier. In the aforementioned analysis of 4400 ORC patients, margin-positive rates by stage were 2.3% for pT2, 7.6% for pT3, and 24% for pT4 disease [96] . The effects of the learning curve as institutions adopted this new technology and patient selection toward earlier stage disease likely affected reported margin rates and should be considered when interpreting outcomes. Nonetheless, higher reported rates of positive margins in pT4 disease in some RARC series suggest that caution be taken for higher stage disease, with particular attention paid to the risk of margin involvement.
Several RARC series did not show decreasing margin rates with sequential case volume. A few reasons could explain this observation: (1) The positive margin numbers may be too low to detect a subgroup difference; (2) the learning curve for reducing margins at RARC could be extremely high, with a number not yet reached in smaller learning curve assessments; or (3) over time, more experienced surgeons may be more willing to take on bulky or higher stage tumors. This final hypothesis is supported by a multivariate analysis adjusting for pathologic stage that shows that differences in stage of disease accounted for an increase in margin rates with more experienced robotic surgeons [61] .
Chemotherapy use alongside surgery in the treatment of MIBC can be implemented either before or after cystectomy. While neoadjuvant chemotherapy has been shown to confer an OS advantage of 5% in randomized trials [97] , the benefit of adjuvant chemotherapy is less proven. In a recent meta-analysis of nine randomized controlled trials comprising 945 patients that investigated the use of adjuvant chemotherapy, benefits to both OS and DFS were appreciated. Patients receiving adjuvant chemotherapy after cystectomy had 23% relative risk reduction in the risk of death (OS:p = 0.049) and 34% relative decrease in the risk of disease recurrence (DFS:p = 0.014) [98] . For the current systematic review, neoadjuvant chemotherapy use was 0–31%, and adjuvant chemotherapy was delivered to 4–29% of patients. Adjuvant therapy was chiefly administered in patients with advanced-stage pT3 or higher or with positive lymph nodes. Although time to initiation of adjuvant chemotherapy was shorter by 3 wk in the analysis by Pruthi et al, further validation is required [31] .
Long-term freedom from disease recurrence and bladder cancer–related death is the primary measure of treatment efficacy with radical cystectomy. Particularly with assessments of survival, gathering data for comparison with the open standard is challenging secondary to the necessity of controlling for cancer characteristics, additional therapies, and the length of follow-up required to detect significant differences. Shorter-interval examinations of survival may not amply capture events such as local recurrence, distant recurrence, or secondary therapies. Only two series in this systematic review compared survival for RARC and ORC. These studies were not randomized and included sequential series of retrospective groups (LOE 4)[82] and [87]. In a series by Nepple et al, 36 patients who underwent RARC were compared with 29 patients who underwent ORC with a median follow-up of only 12 mo. Estimated 2-yr DFS (67% vs 58%), CSS (75% vs 63%), and OS (68% vs 63%) after RARC and ORC were similar for the two techniques, respectively [82] .
Because of limitations of present studies, comparisons must be made to large historical retrospective open series. A long-term analysis of survival in 1100 chemotherapy-naive cystectomy patients by Hautmann et al demonstrated 10-yr CSS and OS rates of 67% and 44%, respectively [99] . For this systematic review, 5-yr estimates for DFS, CSS, and OS were 53–74%, 66–80%, and 39–66%, respectively. Analogous to stratified outcomes in ORC, survival outcomes were worse in RARC series with increasing pathologic stage and with lymph node metastases. Local control of disease appears to be adequate such that the majority of recurrences after RARC are distant or outside the pelvis. A potential concern for port-site metastases with RARC remains of particular interest. Although no specific published series address this concern and most larger RARC oncologic series did not report any incidents, a few case reports suggest that this concern requires further study.
Since the original description of neurovascular bundle preservation during radical prostatectomy by Walsh et al [104] , techniques to improve functional outcomes through meticulous nerve sparing have been translated to radical cystectomy. Turner et al determined that nerve sparing improved urinary continence after orthotopic urinary diversion [100] , and nerve sparing has been shown to assist with recovery of erectile function objectively based on IIEF [101] . Long-term functional evaluations of ileal neobladder continent diversions have demonstrated daytime continence rates of 92% and nighttime continence rates of 80% [102] .
To date, very limited data are available regarding functional outcomes of continence or potency after RARC. These analyses have chiefly been limited to only a few centers that exhibit significant heterogeneity. The 12-mo reported continence rates were 88–100% in men and 67% in women for daytime continence and between 66% and 75% for nighttime continence. Potency recovery exhibited even greater variation, with sufficient erection rates between 9% and 81%. Functional outcomes are likely influenced by patient factors and selection, comorbidity, prior treatments, surgeon experience, and technique (eg, the use of cautery vs clips). In addition, methodology of reporting, definitions ofcontinence, measurement tools, rehabilitation programs, and inconsistencies in follow-up can affect the actual measurement of continence and potency. Specific functional concerns of RARC related to patient selection are that many patients may be older or have poor baseline erectile function. Moreover, technical concerns for a possible PSM, which portends a dismal outcome, may affect the performance of nerve sparing. The lack of conclusive data regarding functional recovery after RARC is a necessary area for future study.
There is no evidence to date that the results from a recent systematic review on RARP finding slight advantages to continence and potency recovery compared with open radical prostatectomy or laparoscopic radical prostatectomy extrapolate to RARC [103] . Precise definitions ofcontinenceandpotencyare necessary so that future data acquisition can be carried out in a standardized, stringent, and uniform fashion for both ORC and RARC.
From a methodological perspective, the most relevant limitations of this systematic review are the quality of the available studies, the small number of patients in and the retrospective nature of most series, the shorter-term follow-up of these studies, and the lack of standardized definitions. The papers included in the present review included only two small randomized controlled trials; the remaining series are LOE 3 or 4. Comparisons made in these single-institution studies inevitably carry the risk of selection bias. Even in randomized controlled studies, there were unlikely to have been equally experienced open and robotic surgeons operating on comparable patients. Heterogeneity in lymph node templates, sampling methods, specimen handling, and pathologic review may affect lymph node yields. Most cumulative outcomes were weighted by the results of experienced surgeons, which may make conclusions difficult to generalize. The inability to account for surgeon factors or specific technique modifications is another limitation. Most series failed to provide specific information concerning relevant aspects of the reconstructive portions of the operation.
Sufficient lymph node yields are achievable through robotic PLND if an extended template is followed. PSM rates appear similar with RARC and ORC. Conclusive long-term survival outcomes for RARC are limited, although oncologic outcomes of ≤5 yr are similar to those reported for ORC. Initial functional outcomes appear favorable; however, additional research on continence and potency after RARC is needed.
Author contributions: Bertram Yuh had full access to all the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.
Study concept and design: Yuh, Wilson, Bochner, Montorsi, Chan, Thalmann, Palou, Stenzl, Guru, Catto, Novara, Wiklund.
Acquisition of data: Yuh, Wilson, Bochner, Montorsi, Chan, Thalmann, Palou, Stenzl, Guru, Catto, Novara, Wiklund.
Analysis and interpretation of data: Yuh, Novara.
Drafting of the manuscript: Yuh, Wilson, Bochner, Montorsi, Chan, Thalmann, Palou, Stenzl, Guru, Catto, Novara, Wiklund.
Critical revision of the manuscript for important intellectual content: Yuh, Wilson, Bochner, Montorsi, Chan, Thalmann, Palou, Stenzl, Guru, Catto, Novara, Wiklund.
Statistical analysis: Yuh, Novara.
Obtaining funding: Wilson.
Administrative, technical, or material support: Yuh, Wilson, Bochner, Montorsi, Chan, Thalmann, Palou, Stenzl, Guru, Catto, Novara, Wiklund.
Supervision: Yuh, Wilson, Bochner, Montorsi, Chan, Thalmann, Palou, Stenzl, Guru, Catto, Novara, Wiklund.
Other(specify): None.
Financial disclosures: Bertram Yuh certifies that all conflicts of interest, including specific financial interests and relationships and affiliations relevant to the subject matter or materials discussed in the manuscript (eg, employment/affiliation, grants or funding, consultancies, honoraria, stock ownership or options, expert testimony, royalties, or patents filed, received, or pending), are the following: Dr. Wiklund has received proctoring assignments and a research grant from Intuitive Surgical, Sunnyvale, CA, USA; Dr. Wilson has been a consultant and a speaker for Intuitive Surgical, Sunnyvale, CA, USA.
Funding/Support and role of the sponsor: None.
Radical cystectomy and pelvic lymph node dissection (PLND) is the gold standard treatment for muscle-invasive bladder cancer (MIBC) and high-risk non–muscle-invasive disease [1] . Patients undergoing this operation can experience 66% recurrence-free survival at 10 yr after surgery [2] . The addition of neoadjuvant platinum-based chemotherapy has been shown to improve overall survival (OS) rates by approximately 5% [3] . Robot-assisted radical cystectomy (RARC) was initially described by Menon et al in 2003 [4] . Over time, many international centers have adopted RARC.
Oncologic outcomes from large population-based cohorts of RARC with lengthy follow-up are lacking. Early on in RARC history, surrogates for oncologic control were reported using positive surgical margin (PSM) rates and lymph node yields. More recently, 5-yr survival figures have become available. The majority of these outcomes, however, capture institutions early in their learning curves and incorporate patients potentially selected for the robotic technique, thus avoiding more advanced-stage or technically difficult cases. Data on functional consequences of RARC are even more limited; therefore, the quality of nerve sparing and its effect on potency recovery and continence are inadequately understood.
Because of the expanding evidence available in the field of RARC, and in preparation for the Pasadena international consensus meeting on best practice in RARC and urinary reconstruction, we performed a systematic literature review of perioperative, functional, and oncologic outcomes of RARC in comparison with open radical cystectomy (ORC) and laparoscopic radical cystectomy (LRC).
We report on the systematic review and cumulative analysis of oncologic and functional outcomes of RARC. We systematically examined lymph node yields, PSMs, cancer-specific survival (CSS), disease-free survival (DFS), and OS. In addition, functional outcomes after RARC, including urinary continence and erectile function, were systematically examined.
A systematic literature search was initially performed in September 2013 using the Medline, Scopus, and Web of Science databases. The searches included only a free-text protocol using the termsrobot-assisted radical cystectomyorda Vinci radical cystectomyorrobot* radical cystectomyin all the fields of the records for Medline and Scopus searches and in theTitleandTopicfields for the Web of Science search. No limits were applied. A full update of the searches was performed on April 28, 2014.
Two authors (G.N. and B.Y.) separately reviewed the records to select RARC case series and studies that compared RARC with ORC and RARC with LRC. Discrepancies were resolved by open discussion. Other significant studies cited in the reference lists of the selected papers were evaluated, as were studies published after the systematic search.
All noncomparative studies reporting the following data on RARC were collected: intraoperative and perioperative data (operative time, blood loss, transfusion rate, in-hospital stay, readmission, complication rates), functional data (urinary continence, erectile function), and oncologic data (PSMs, lymph node yield, DFS, CSS, OS). The present review included only studies reporting on functional and oncologic data.
Studies reporting on partial cystectomy, prostate-sparing cystectomy, salvage cystectomy, cystectomy for urachal cancer or benign disease, single-case reports, pure laparoscopic (or mixed) series, or laparoendoscopic single-site or natural orifice transluminal endoscopic surgery for radical cystectomy; experimental studies on animal models; congress abstracts; review papers; editorials; population-based studies; and book chapters were not included in the review. All data retrieved from the selected studies were recorded in an electronic database.
All papers were categorized according to the 2011 levels of evidence (LOEs) for therapy studies: LOE 1, systematic review of randomized trials or n-of-1 trials; LOE 2, randomized trial or observational study with dramatic effect; LOE 3, nonrandomized controlled cohort/follow-up study; LOE 4, case series, case–control study, or historically controlled study; or LOE 5, mechanism-based reasoning [5] . Papers were categorized according to the IDEAL recommendations [6] .
Cumulative analysis was conducted using Review Manager v5.2 software designed for composing Cochrane Reviews (Cochrane Collaboration, Oxford, UK). Statistical heterogeneity was tested using the chi-square test. Apvalue <0.10 was used to indicate heterogeneity. Where there was a lack of heterogeneity, fixed-effects models were used for the cumulative analysis. Random-effects models were used in case of heterogeneity. For continuous outcomes, the results were expressed as weighted mean differences and standard deviations (SDs); for dichotomous variables, results were given as odds ratios (ORs) and 95% confidence intervals (CIs). Because of limitations in the Review Manager v5.2 software, meta-analysis of continuous variables was possible only when rough data were presented as mean and SD. Authors of the papers were contacted to provide missing data, whenever necessary. For all statistical analyses, two-sidedp < 0.05 was considered statistically significant.
Figure 1 shows a flowchart of this systematic review of the literature.
In total, 65 surgical series[4], [7], [8], [9], [10], [11], [12], [13], [14], [15], [16], [17], [18], [19], [20], [21], [22], [23], [24], [25], [26], [27], [28], [29], [30], [31], [32], [33], [34], [35], [36], [37], [38], [39], [40], [41], [42], [43], [44], [45], [46], [47], [48], [49], [50], [51], [52], [53], [54], [55], [56], [57], [58], [59], [60], [61], [62], [63], [64], [65], [66], [67], [68], [69], and [70]and 22 comparative studies[71], [72], [73], [74], [75], [76], [77], [78], [79], [80], [81], [82], [83], [84], [85], [86], [87], [88], [89], [90], [91], and [92]reported on pathologic, oncologic (n = 18), or functional (n = 9) outcomes of RARC.
Most surgical series were retrospective, single-center studies (LOE 4). Exceptions included prospective studies[8], [22], [28], [35], [36], [43], [52], [59], [64], [66], [81], and [87]and some multi-institutional retrospective collaboration studies[16], [23], [25], [30], [45], [48], [54], [61], and [68]. Only two of the comparative studies were randomized[74] and [83](LOE 2b); all other comparative studies were nonrandomized, whether prospective or retrospective (LOE 4).
Table 1 summarizes the number of lymph nodes recovered in published RARC series. The majority of studies (86%) reported extent of lymph node dissection (LND), with more centers performing extended LND (ELND) in recent series. Standard LND typically involved the removal of obturator, internal iliac, external iliac, and some portion of the common iliac lymph nodes bilaterally. ELND templates typically brought the proximal extent up to the aortic bifurcation or inferior mesenteric artery. Approximately half of the analyzed studies reported following an extended template of dissection.
Table 1 Lymph node yields in robot-assisted radical cystectomy series
| Reference | Institution | IDEAL stage | Cases, no. | Study design | Extension of LND | Operative time, min | Retrieved nodes, no. | pN+, % | Metastatic nodes, median, no. | Complications due to LND |
|---|---|---|---|---|---|---|---|---|---|---|
| Menon et al, 2003 [4] | Henry Ford Hospital | 1 | 17 | Retrospective | Standard | – | – | 6 | – | – |
| Menon et al, 2004 [7] | Henry Ford Hospital | 1 | 3 female | Retrospective | Standard | – | 12 | 0 | – | – |
| Guru et al, 2007 [8] | Roswell Park Cancer Institute | 1 | 20 | Prospective | Standard | 44 | 13 | 15 | 1 | – |
| Mottrie et al, 2007 [9] | O.L.V.–Clinic | 2a | 27 | Retrospective | Extended | – | 23 | 9 | – | – |
| Pruthi et al, 2008 [71] | UNC | 2a | 20 | Retrospective | Standard | – | 19 | 10 | – | – |
| Hemal et al, 2008 [10] | All India Institute of Medical Sciences | 1 | 6 | Retrospective | Standard | – | 12 | 17 | – | – |
| Lowentritt et al, 2008 [11] | Tulane University | 2a | 4 | Retrospective | Standard | – | 12 | 25 | – | – |
| Murphy et al, 2008 [12] | Guy's Hospital | 2a | 23 | Retrospective | Standard | – | 16 | 9 | – | – |
| Park et al, 2008 [13] | Yonsei | 2a | 4 | Retrospective | Standard | – | 17 | 0 | – | – |
| Pruthi et al, 2008 [14] | UNC | 2a | 12 female | Retrospective | Standard, then extended | – | 19 | 17 | – | – |
| Pruthi et al, 2008 [15] | UNC | 2b | 50 | Retrospective | Standard | – | 19 | 20 | – | – |
| Wang et al, 2008 [72] | Cornell | 2b | 33 | Retrospective | Standard | – | 17 | 19 | – | – |
| Woods et al, 2008 [16] | Mayo Arizona Tulane University |
2b | 27 | Multi-institutional | Extended | – | 12.3 | 33 | 3.1 | 0 |
| Yuh et al, 2008 [17] | Roswell Park Cancer Institute | 2a | 54 | Retrospective | Extended | – | 17 | – | – | – |
| Gamboa et al, 2009 [18] | University of California, Irvine |
2a | 41 | Retrospective | Standard | – | 23 | 14 | 4 | – |
| Pruthi et al, 2009 [19] | UNC | 2b | 50 | Retrospective | Standard, then extended | – | 19 | 16 | – | – |
| 10 female | 19 | |||||||||
| 40 male | 18 | |||||||||
| Palou Redorta et al, 2009 [20] | Barcelona Autonomous University | 2a | 9 | Retrospective | Extended | 60 | 10 | 0 | – | – |
| Yuh et al, 2009 [21] | Roswell Park Cancer Institute | 2b | 73 | Retrospective | Extended | – | 19 | – | – | – |
| Guru et al, 2010 [22] | Roswell Park Cancer Institute | 2a | 26 | Prospective | Extended | – | 21 | 29 | 1 | Internal iliac artery injury: 1 |
| Hellenthal et al, 2011 [23] | IRCC | 2b | 437 | Multi-institutional | – | – | 17 | 20 | – | – |
| Josephson et al, 2010 [24] | City of Hope Cancer Center | 2b | 58 | Retrospective | Extended | – | 27 | 24 | – | – |
| Kang et al, 2010 [25] | Multicenter | 2b | 71 standard LND 33 extended LND |
Retrospective | Standard Extended |
– | 15.7 24.7 |
10 | – | – |
| Kasraeian et al, 2010 [26] | Montsouris Institute | 2a | 9 | Retrospective | Extended | – | 11 | 22 | – | 0 |
| Kauffman et al, 2011 [27] | Cornell | 2b | 85 | Retrospective | Extended | – | 19 | 15 | – | – |
| Kwon et al, 2010 [28] | Kyungpook National University | 2a | 17 | Prospective | Standard | – | 6 | 6 | 1 | 0 |
| Lavery et al, 2011 [29] | Ohio State University | 2a | 15 | Retrospective | Extended | 107 | 41.8 | 20 | – | 0 |
| Martin et al, 2010 [30] | Mayo Arizona Tulane University |
2b | 59 | Multi-institutional | Extended | – | – | 34 | – | – |
| Ng et al, 2010 [73] | Cornell | 2b | 83 | Retrospective | Standard | – | 16 | 16 | – | – |
| Nix et al, 2010 [74] | UNC | 3 | 21 | RCT | Standard | – | 19 | 19 | – | – |
| Pruthi et al, 2010 [31] | UNC | 2b | 100 | Retrospective | Standard, then extended | – | 19 | 20 | – | – |
| Richards et al, 2010 [75] | Wake Forest University | 2b | 35 | Retrospective | Extended | – | 16 | 29 | – | – |
| Akbulut et al, 2011 [32] | Ankara Ataturk Training and Research Hospital | 2a | 12 | Not reported | Extended | – | 21.3 | 42 | – | 8 |
| Canda et al, 2012 [33] | Ankara Ataturk Training and Research Hospital | 2a | 27 | Not reported | Extended | – | 24.8 | 22 | – | – |
| Davis et al, 2011 [34] | University of Texas M.D. Anderson Cancer Center | 2a | 11 | Retrospective | Extended | 117 | 43 | 9 | 1 | – |
| Jonsson et al, 2011 [35] | Karolinska Institute | 2b | 45 36 neobladder 9 ileal conduit |
Prospective | Standard Extended |
– | 19 19 27 |
20 17 33 |
– | – |
| Khan et al, 2011 [36] | Guy's Hospital | 2a | 50 | Prospective | – | – | 17 | – | – | – |
| Manoharan et al, 2011 [37] | University of Miami | 2a | 14 | Retrospective | Standard | – | 12 | – | – | – |
| Martin et al, 2011 [76] | Mayo Arizona | 2b | 19 | Retrospective | – | – | 16 | – | – | – |
| Richards et al, 2011 [38] | Wake Forest University | 2b | 60 | Retrospective | Extended | – | 17 | 30 | – | Lymphocele: 1 |
| Schumacher et al, 2011 [39] | Karolinska Institute | 2b | 45 | Retrospective | Standard 49%, extended 31% | – | 22.5 | – | 1.5 | Lymphocele: 2 |
| Shah et al, 2011 [40] | Ohio State University | 2b | 30 | Retrospective | Extended | – | – | 30 | – | – |
| Torrey et al, 2011 [41] | City of Hope Cancer Center | 2b | 34 | Retrospective | Extended | – | 28.9 | – | – | – |
| Cho et al, 2012 [42] | Hallym University College of Medicine | 2b | 35 | Retrospective | Standard | – | – | 6 | – | – |
| Goh et al, 2012 [43] | Keck School of Medicine, University of Southern California, Los Angeles | 2a | 15 | Prospective | Superextended | – | 55 | 26 | – | – |
| Lau et al, 2012 [44] | City of Hope Cancer Center | 2b | 23 (aged >80 yr) | Retrospective | Extended | – | 20.4 | 22 | – | – |
| Mmeje et al, 2013 [45] | Mayo Arizona UNC |
2b | 50 | Multi-institutional | Extended | – | 18 | 100 | 3 | – |
| Poch et al, 2012 [46] | Roswell Park Cancer Institute | 2b | 56 | Retrospective | – | – | 25 | 16 | – | – |
| Richards et al, 2012 [77] | Wake Forest University | 2b | 20 (aged >75 yr) | Retrospective | Extended | – | 17 | 35 | – | – |
| Saar et al, 2013 [47] | Saarland University | 2b | 62 | Retrospective | – | – | 14.2 | 21 | – | – |
| Smith et al, 2012 [48] | Mayo Arizona UNC, Tulane University |
2b | 227 | Multi-institutional | – | – | 18 | 20 | – | – |
| Styn et al, 2012 [78] | University of Michigan | 2b | 50 | Retrospective | – | – | 14.3 | 12 | – | – |
| Sung et al, 2012 [79] | Samsung Medical Center | 2b | 35 | Retrospective | Standard | – | 19.1 | 26 | – | Lymphocele: 1 |
| Treiyer et al, 2012 [49] | Saarland University | 2b | 91 | Retrospective | Standard | – | 14.5 | 14 | – | – |
| Tsui et al, 2012 [50] | Chang Gung Memorial Taiwan | 2a | 8 | Retrospective | Standard | – | 3 | 12.50 | – | – |
| Yuh et al, 2012 [51] | City of Hope Cancer Center | 2b | 196 | Retrospective | Extended | – | 28 | 22 | – | Lymphocele: 3 |
| Collins et al, 2013 [52] | Karolinska Institute | 2b | 113 | Prospective | Extended 56%, standard 34%, limited 5%, none 5% | – | 21 | 20 | – | Lymphocele: 5 |
| Johar et al, 2013 [53] | IRCC | 2b | 939 | Multi-institutional | – | – | 18.1 | 26 | – | – |
| Maes et al, 2013 [80] | Metro Health Hospital | 2b | 14 | Retrospective | Extended | – | 11.9 | 7 | – | – |
| Marshall et al, 2013 [54] | IRCC | 2b | 765 | Multi-institutional | Extended 58%, standard 40%, no LND 2% | – | 18 | 27 | – | – |
| Musch et al, 2014 [81] | Klinikin Essen–Mitte | 2b | 100 | Prospective | – | – | 26.5 | 20 | – | Lymphocele: 4 |
| Nazmy et al, 2014 [55] | City of Hope Cancer Center | 2b | 209 | Retrospective | Extended | – | – | 22 | – | Lymphocele: 3 |
| Nepple et al, 2013 [82] | Washington University | 2b | 36 | Retrospective | Standard | – | 17 | 22 | – | – |
| Parekh et al, 2013 [83] | University of Texas Health Sciences Center at San Antonio | 3 | 20 | RCT | Standard | – | 11 | 20 | – | – |
| Tyritzis et al, 2013 [56] | Karolinska Institute | 2b | 70 | Retrospective | Standard 43% Extended 48% |
– | 21 | 14 | – | Lymphocele: 6 Lymphedema: 1 |
| Xylinas et al, 2013 [57] | Cornell | 2b | 175 | Retrospective | Standard | – | 19 | 17 | – | Lymphocele: 2 |
| Phillips et al, 2014 [69] | Seward St. Elizabeth Medical Center | 2b | 23 (>80 yr) | Retrospective | Extended | – | 19 | – | – | |
| Raza et al, 2014 [70] | Roswell Park Cancer Institute | 2b | 99 | Retrospective | – | – | 20.7 | 36 | – | – |
| Yuh et al, 2014 [58] | City of Hope Cancer Center | 2b | 162 | Retrospective | Extended | – | 28 | 23 | – | – |
| Total | 19.3 | 23 |
IRCC = International Robotic Cystectomy Consortium; LND = lymph node dissection; RCT = randomized controlled trial; UNC = University of North Carolina.
The lymph node yield from all series was 19 (range: 3–55). Initial descriptions using a standard template of dissection achieved yields of 18 lymph nodes [59] . Number of lymph nodes recovered with an ELND ranged from 11 to 55. Abaza et al adopted a robotic template similar to the open technique, including external iliac, obturator, hypogastric, common iliac, and presacral up to the aortic bifurcation; the mean lymph node yield was 37.5 (SD: 13.2), demonstrating that lymph node counts could mirror those of open dissection if the same template was followed [88] . In a study of open completion LND after robot-assisted ELND in 11 men, Davis et al removed only an additional 4 lymph nodes with an open approach after 43 were removed with robot assistance [34] . Time of LND was rarely reported, although it ranged from 44 min in standard LND to 117 min in ELND[8] and [34]. The lymph node–positive rate was 22%. In series with >20 RARCs, lymph node–positive rates ranged from 6% to 42%. Reports of vascular injuries were rare, and lymphocele rates were 0–9%.
Table 2 summarizes the studies assessing the effects of patient characteristics and particular surgical aspects on lymph node yields in RARC series. Cumulative analysis from the International Robotic Cystectomy Consortium (IRCC) with respect to lymphadenectomy in 437 patients found a median of 17 lymph nodes removed, with a 20% node-positivity rate [23] . Patient age and sex did not affect the performance of lymphadenectomy. In a different series, increasing body mass index (BMI) did not appear to negatively affect lymph node yield, with >20 lymph nodes removed in normal, overweight, and obese patients [46] .
Table 2 Impact of patient characteristics and surgical aspects on lymph node yield in robot-assisted radical cystectomy series
| Reference | Institution | IDEAL stage |
Cases | Study design | Extension of LND | Operative time, min | Retrieved nodes, no. | pN+, % |
|---|---|---|---|---|---|---|---|---|
| Patient BMI | ||||||||
| Poch et al, 2012 [46] | Roswell Park Cancer Institute | 2b | 56 | Retrospective | – | – | 25 | 16 |
| BMI <25: 14 | 22 | 7 | ||||||
| BMI 25 to <30: 21 | 23 | 14 | ||||||
| BMI ≥30: 21 | 20 | 24 | ||||||
| Case volume | ||||||||
| Guru et al, 2008 [59] | Roswell Park Cancer Institute | 2a | 1–12 | Prospective | Extended | 46 | 33% >13 | 8 |
| 13–24 | 44 | 66% >13 | 33 | |||||
| 25–36 | 41 | 83% >13 | 25 | |||||
| 37–47 | 43 | 72% >13 | 64 | |||||
| 48–58 | 56 | 91% >13 | 18 | |||||
| Pruthi et al, 2008 [60] | University of North Carolina | 2b | 50 | Retrospective | Standard | – | 19 | – |
| 1–10 | 21 | |||||||
| 11–20 | 19 | |||||||
| 21–30 | 20 | |||||||
| 31–40 | 17 | |||||||
| 41–50 | 20 | |||||||
| Richards et al, 2011 [38] | Wake Forest University | 2b | 60 | Retrospective | Extended | – | 17 | 30 |
| 1–20 | 17 | |||||||
| 21–40 | 19.1 | |||||||
| 41–60 | 14.4 | |||||||
| Schumacher et al, 2011 [39] | Karolinska Institute | 2b | 45 | Retrospective | Standard 49% | – | 22.5 | – |
| Extended 31% | ||||||||
| 1–15 | Standard 40% | |||||||
| Extended 7% | ||||||||
| 16–30 | Standard 47% | |||||||
| Extended 53% | ||||||||
| 31–45 | Standard 60% | |||||||
| Extended 33% | ||||||||
| Prior RARP experience | ||||||||
| Hayn et al, 2010 [61] | IRCC | 2b | 496 | Retrospective | – | – | 17.8 | – |
| ≤50 previous RARP: 83 | 13.7 | |||||||
| 51–100 previous RARP: 187 | 19.8 | |||||||
| 101–150 previous RARP: 176 | 19.6 | |||||||
| >150 previous RARP: 50 | 11.8 * | |||||||
* Statistically significant.
BMI = body mass index; RCC = International Robotic Cystectomy Consortium; LND = lymph node dissection; RARP = robot-assisted radical prostatectomy.
It is interesting to note that in single-institution series, Richards et al [38] , Schumacher et al [39] , Guru et al [59] , and Pruthi et al [60] did not find higher lymph node yields with increasing sequential case numbers. However, in the IRCC, performance of lymphadenectomy was positively correlated with surgeon and institution volume but was reduced in patients with more advanced disease (pT4 stage), which may reflect operative avoidance of bulky nodal tissue.
Table 3 summarizes the occurrence of PSMs reported in the RARC series. The reported PSM rates were 5.6% (range: 0–26%). However, in series of >100 patients, margin rates ranged between 4% and 9%[48] and [53]. PSMs were reported in 1–1.5% of patients with pT2 disease and 0–25% of patients with pT3 and higher disease. PSM rates from the IRCC in 939 cases were 9% [53] .
Table 3 Positive surgical margins in robot-assisted radical cystectomy series
| Reference | Institution | IDEAL stage | Cases, no. | Study design | Pathologic stage, % | Overall PSM rate, % | PSM location | PSM rate, % | ||
|---|---|---|---|---|---|---|---|---|---|---|
| ≤pT2 | ≥pT3 | ≤pT2 | ≥pT3 | |||||||
| Menon et al, 2003 [4] | Henry Ford Hospital | 1 | 17 | Retrospective | – | – | 0 | – | 0 | 0 |
| Yohannes et al, 2003 [62] | Creighton University | 1 | 2 | Retrospective | 0 | 100 | 0 | – | 0 | 0 |
| Menon et al, 2004 [7] | Henry Ford Hospital | 1 | 3 female | Retrospective | 66 | 33 | 0 | – | 0 | 0 |
| Rhee et al, 2006 [84] | University of Virginia | 1 | 7 | Retrospective | 43 | 57 | 0 | – | 0 | 0 |
| Guru et al, 2007 [8] | Roswell Park Cancer Institute | 1 | 20 | Prospective | 40 | 60 | 15 | Prostate: 1 Ureter: 1 Vagina: 1 |
0 | 25 |
| Mottrie et al, 2007 [9] | O.L.V.–Clinic | 2a | 27 | Retrospective | 78 | 22 | 4 | Ureter: 1 | – | – |
| Pruthi et al, 2008 [71] | UNC | 2a | 20 | Retrospective | 70 | 20 | 0 | – | 0 | 0 |
| Hemal et al, 2008 [10] | All India Institute of Medical Sciences | 1 | 6 | Retrospective | 67 | 33 | 0 | – | 0 | 0 |
| Lowentritt et al, 2008 [11] | Tulane University | 2a | 4 | Retrospective | 25 | 75 | 0 | – | 0 | 0 |
| Murphy et al, 2008 [12] | Guy's Hospital | 2a | 23 | Retrospective | 74 | 17 | 0 | – | 0 | 0 |
| Park et al, 2008 [13] | Yonsei | 2a | 4 | Retrospective | 50 | 50 | 0 | – | 0 | 0 |
| Pruthi et al, 2008 [14] | UNC | 2a | 12 female | Retrospective | 58 | 25 | 0 | – | 0 | 0 |
| Pruthi et al, 2008 [15] | UNC | 2b | 50 | Retrospective | 66 | 14 | 0 | – | 0 | 0 |
| Pruthi et al, 2009 [19] | UNC | 2b | 50 10 female 40 male |
Retrospective | 66 50 70 |
18 30 15 |
0 | – | 0 | 0 |
| Wang et al, 2008 [72] | Cornell | 2b | 33 | Retrospective | 72 | 28 | 6 | Perivesical fat: 2 | 0 | 22 |
| Woods et al, 2008 [16] | Mayo Arizona Tulane University |
2b | 27 | Multi-institutional | – | – | 7 | – | 0 | – |
| Yuh et al, 2008 [17] | Roswell Park Cancer Institute | 2a | 54 | Retrospective | 44 | 56 | 13 | – | 0 | 23 |
| Gamboa et al, 2009 [18] | University of California, Irvine | 2a | 41 | Retrospective | – | – | 5 | – | 0 | – |
| Palou Redorta et al, 2009 [20] | Barcelona Autonomous University | 2a | 9 | Retrospective | 66 | 33 | 11 | – | – | – |
| Yuh et al, 2009 [21] | Roswell Park Cancer Institute | 2b | 73 | Retrospective | 45 | 55 | 10 | – | 0 | 18 |
| Guru et al, 2010 [22] | Roswell Park Cancer Institute | 2a | 20 | Prospective | 62 | 38 | 4 | – | 0 | 9 |
| Hayn et al, 2010 [61] | IRCC | 2b | 482 | Multi-institutional | 64 | 36 | 7 | – | – | – |
| Hellenthal et al, 2010 [68] | IRCC | 2b | 513 | Multi-institutional | 64 | 36 | 7 | – | 1.50 | 17 |
| Kang et al, 2010 [25] | Multicenter | 2b | 104 | Multi-institutional | 70 | 30 | 5 | – | – | – |
| Kasraeian et al, 2010 [26] | Montsouris Institute | 2a | 9 | Retrospective | 44 | 66 | 0 | – | – | – |
| Kauffman et al, 2011 [27] | Cornell | 2b | 85 | Retrospective | 64 | 36 | 6 | – | 0 | 16 |
| Kwon et al, 2010 [28] | Kyungpook National University | 2a | 17 | Prospective | 59 | 41 | 0 | – | 0 | 0 |
| Martin et al, 2010 [30] | Mayo Arizona Tulane University |
2b | 59 | Multi-institutional | 47 | 53 | – | – | – | – |
| Ng et al, 2010 [73] | Cornell | 2b | 83 | Retrospective | 61 | 39 | 7 | – | 0 | 19 |
| Nix et al, 2010 [74] | UNC | 3 | 21 | RCT | 67 | 14 | 0 | – | 0 | 0 |
| Pruthi et al, 2010 [31] | UNC | 2b | 100 | Retrospective | 67 | 13 | 0 | – | 0 | 0 |
| Richards et al, 2010 [75] | Wake Forest University | 2b | 35 | Retrospective | 60 | 40 | 3 | – | – | – |
| Akbulut et al, 2011 [32] | Ankara Ataturk Training and Research Hospital | 2a | 12 | Not reported | 58 | 42 | 0 | – | 0 | 0 |
| Canda et al, 2012 [33] | Ankara Ataturk Rraining and Research Hospital | 2a | 27 | Not reported | 56 | 44 | 4 | – | 0 | 4 |
| Davis et al, 2011 [34] | University of Texas M.D. Anderson Cancer Center | 2a | 11 | Retrospective | 92 | 8 | 0 | – | – | – |
| Jonsson et al, 2011 [35] | Karolinska Institute | 2b | 45 | Prospective | 78 | 22 | 2 | – | 0 | 10 |
| Khan et al, 2011 [36] | Guy's Hospital | 2a | 50 | Prospective | 72 | 28 | 2 | – | 0 | 7 |
| Manoharan et al, 2011 [37] | University of Miami | 2a | 14 | Retrospective | – | – | 0 | – | 0 | 0 |
| Martin et al, 2011 [76] | Mayo Arizona | 2b | 19 | Retrospective | 42 | 58 | – | – | – | – |
| Richards et al, 2011 [38] | Wake Forest University | 2b | 60 | Retrospective | 63 | 37 | 10 | – | – | – |
| Schumacher et al, 2011 [39] | Karolinska Institute | 2b | 45 | Retrospective | 78 | 22 | 2 | Ureter: 1 | 0 | 10 |
| Shah et al, 2011 [40] | Ohio State University | 2b | 30 | Retrospective | 65 | 35 | 7 | – | 0 | 22 |
| Cho et al, 2012 [42] | Hallym University College of Medicine | 2b | 35 | Retrospective | 86 | 14 | 3 | – | – | – |
| Goh et al, 2012 [43] | Keck School of Medicine, University of Southern California, Los Angeles | 2a | 15 | Prospective | 67 | 33 | 0 | – | – | – |
| Lau et al, 2012 [44] | City of Hope Cancer Center | 2b | 23 (aged >80 yr) | Retrospective | 61 | 39 | 13 | Ureter: 1 | – | – |
| Mmeje et al, 2013 [45] | Mayo Arizona | 2b | 50 | Multi-institutional | 34 | 66 | 2 | – | – | – |
| UNC | ||||||||||
| Poch et al, 2012 [46] | Roswell Park Cancer Institute | 2b | 56 | Retrospective | 55 | 45 | – | – | – | – |
| Richards et al, 2012 [77] | Wake Forest University | 2b | 20 (aged >75 yr) | Retrospective | 60 | 40 | 5 | – | – | – |
| Saar et al, 2013 [47] | Saarland University | 2b | 62 | Retrospective | 64 | 36 | 2 | – | – | – |
| Smith et al, 2012 [48] | Mayo Arizona | 2b | 227 | Multi-institutional | – | – | 2 | – | – | – |
| UNC, Tulane University | ||||||||||
| Styn et al, 2012 [78] | University of Michigan | 2b | 50 | Retrospective | 60 | 40 | 2 | – | – | – |
| Sung et al, 2012 [79] | Samsung Medical Center | 2b | 35 | Retrospective | 43 | 57 | – | – | – | – |
| Treiyer et al, 2012 [49] | Saarland University | 2b | 91 | Retrospective | 67 | 33 | 2 | Urethra: 1 Prostate: 1 | – | – |
| Tsui et al, 2012 [50] | Chang Gung Memorial | 2a | 8 | Retrospective | 75 | 25 | 0 | – | 0 | 0 |
| Yuh et al, 2012 [51] | City of Hope Cancer Center | 2b | 196 | Retrospective | 64 | 36 | 4 | – | – | – |
| Azzouni et al, 2013 [63] | Roswell Park Cancer Institute | 2b | 100 | Retrospective | 35 | 65 | 4 | – | – | – |
| Collins et al, 2013 [52] | Karolinska Institute | 2b | 113 | Prospective | 75 | 25 | 5 | Ureter: 1 | 1 | 18 |
| Johar et al, 2013 [53] | Multicenter | 2b | 939 | Retrospective | 49 | 51 | 9 | – | – | – |
| Maes et al, 2013 [80] | Metro Health Hospital | 2b | 14 | Retrospective | 43 | 57 | 21 | – | – | – |
| Marshall et al, 2013 [54] | IRCC | 2b | 765 | Multi-institutional | 59 | 41 | – | – | – | – |
| Musch et al, 2014 [81] | Klinikin Essen–Mitte | 2b | 100 | Prospective | 61 | 39 | 2 | – | – | – |
| Nazmy et al, 2014 [55] | City of Hope Cancer Center | 2b | 209 | Retrospective | 65 | 35 | 3 | – | – | – |
| Nepple et al, 2013 [82] | Washington University | 2b | 36 | Retrospective | 53 | 47 | 6 | – | 0 | 12 |
| Parekh et al, 2013 [83] | University of Texas Health Sciences Center at San Antonio | 3 | 20 | RCT | 50 | 50 | 5 | – | 0 | 10 |
| Tyritzis et al, 2013 [56] | Karolinska Institute | 2b | 70 | Retrospective | 86 | 14 | 1.5 | Ureter: 1 | 0 | 10 |
| Xylinas et al, 2013 [57] | Cornell | 2b | 175 | Retrospective | 65 | 35 | 5 | – | – | – |
| Phillips et al, 2014 [69] | Seward St. Elizabeth Medical Center | 2b | 23 (aged >80 yr) | Retrospective | 30 | 70 | 26 | – | – | – |
| Raza et al, 2014 [70] | Roswell Park Cancer Institute | 2b | 99 | Retrospective | 48 | 52 | 8 | – | – | |
| Yuh et al, 2014 [58] | City of Hope Cancer Center | 2b | 162 | Retrospective | 67 | 33 | 4 | – | – | – |
| Total | 60 | 40 | 5.6 | |||||||
IRCC = International Robotic Cystectomy Consortium; PSM = positive surgical margin; RCT = randomized controlled trial; UNC = University of North Carolina.
Table 4 summarizes the studies assessing the effects of patient characteristics and particular surgical aspects on PSM rates in RARC series. Notably, Richards et al [38] , Schumacher et al [39] , and the IRCC [68] did not demonstrate decreasing surgical margin rates with sequential case number. In a study of the role of previous robot-assisted radical prostatectomy (RARP) experience on RARC outcomes, there was a trend toward increased positive margins with increasing RARP volumes, but it did not reach statistical significance (p = 0.089) [61] . The authors chiefly attributed this situation to the performance of RARC on patients with higher risk (higher than T3) disease. One study reported that PSMs occurred only in the overweight or obese patients, although pT4 rates were much higher in those patients (26% vs 7%) [65] .
Table 4 Predictors of positive surgical margins in robot-assisted radical cystectomy series
| Reference | Institution | IDEAL stage | Cases | Study design | Pathologic stage, % | Overall PSM rate, % | PSM rate, % | ||
|---|---|---|---|---|---|---|---|---|---|
| ≤pT2 | ≥pT3 | ≤pT2 | ≥pT3 | ||||||
| Case volume | |||||||||
| Guru et al, 2008 [59] | Roswell Park Cancer Institute | 2a | 1–12 | Prospective | 33 | 66 | 17 | – | – |
| 13–24 | 58 | 42 | 25 | ||||||
| 25–36 | 50 | 50 | 0 | ||||||
| 37–47 | 46 | 54 | 9 | ||||||
| 48–58 | 64 | 36 | 0 | ||||||
| Hayn et al, 2011 [64] | Roswell Park Cancer Institute | 2a | 1–50 | Prospective | 51 | 49 | 8 | – | – |
| 51–100 | 12 | ||||||||
| 101–164 | 6 | ||||||||
| Richards et al, 2011 [38] | Wake Forest University | 2b | 60 | Retrospective | 63 | 37 | 10 | – | – |
| 1–20 | 55 | 45 | 5 | ||||||
| 21–40 | 70 | 30 | 5 | ||||||
| 41–60 | 65 | 35 | 20 | ||||||
| Schumacher et al, 2011 [39] | Karolinska Institute | 2b | 45 | Retrospective | 78 | 22 | 2 | 0 | 10 |
| 1–15 | 87 | 13 | 0 | ||||||
| 16–30 | 67 | 33 | 7 | ||||||
| 31–45 | 80 | 20 | 0 | ||||||
| Azzouni et al, 2013 [63] | Roswell Park Cancer Institute | 2b | 100 | Retrospective | 35 | 65 | 4 | – | – |
| 1–25 | 36 | 64 | 4 | ||||||
| 26–50 | 40 | 60 | 4 | ||||||
| 51–75 | 44 | 56 | 4 | ||||||
| 76–100 | 20 | 80 | 4 | ||||||
| Previous RARP experience | |||||||||
| Hayn et al, 2010 [61] | IRCC | 2b | 482 | Retrospective | 64 | 36 | 7 | – | – |
| ≤50 previous RARP: 83 | 68 | 32 | 4 | ||||||
| 51–100 previous RARP: 173 | 76 | 24 | 5 | ||||||
| 101–150 previous RARP: 168 | 54 | 46 | 9.5 | ||||||
| >150 previous RARP: 48 | 42 | 58 | 12.5 | ||||||
| Patient BMI | |||||||||
| Butt et al, 2008 [65] | Roswell Park Cancer Institute | 2a | BMI <25: 14 | Retrospective | 64 | 36 | 0 | 0 | – |
| BMI 25–29: 18 | 28 | 72 | 28 | ||||||
| BMI ≥30: 17 | 42 | 58 | 6 | ||||||
| Poch et al, 2012 [46] | Roswell Park Cancer Institute | 2b | 56 | Retrospective | 55 | 45 | – | – | – |
| BMI <25: 14 | 50 | 50 | |||||||
| BMI 25 to <30: 21 | 57 | 43 | |||||||
| BMI ≥30: 21 | 52 | 48 | |||||||
| Intracorporeal vs extracorporeal diversion | |||||||||
| Kang et al, 2012 [85] | Korea University School of Medicine | 2a | 38 extracorporeal diversion | Retrospective | 76 | 24 | 2.5 | – | – |
| 4 intracorporeal diversion | 100 | 0 | 0 | ||||||
BMI = body mass index; IRCC = International Robotic Cystectomy Consortium; PSM = positive surgical margin; RARP = robot-assisted radical prostatectomy.
Table 5 summarizes the oncologic outcomes of current RARC publications. Neoadjuvant chemotherapy use was reported in 0–31% of patients. Adjuvant chemotherapy use was reported in 4–29% of patients.
Table 5 Survival outcomes in robot-assisted radical cystectomy series
| Reference | Institution | IDEAL stage | Cases, no. | Study design | Follow-up, mo | Neoadjuvant chemotherapy, % | Adjuvant chemotherapy, % | DFS estimates, % | CSS estimates, % | OS estimates, % | ||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 1 yr | 3 yr | 5 yr | 1 yr | 3 yr | 5 yr | 1 yr | 3 yr | 5 yr | ||||||||
| Pruthi et al, 2008 [15] | UNC | 2b | 50 | Retrospective | 13.2 | 0 | 22 | – | – | – | 94 (13 mo) | – | – | 90 (13 mo) | – | – |
| Murphy et al, 2008 [12] | Guy's Hospital | 2a | 23 | Retrospective | 17 | 29 | – | 91 (17 mo) | – | – | – | – | – | – | – | – |
| Josephson et al, 2010 [24] | City of Hope Cancer Center | 2b | 58 | Retrospective | 12 | 22 | – | – | 76 (2 yr) | – | – | 76 (2 yr) | – | – | 54 (2 yr) | – |
| Kang et al, 2010 [25] | Multicenter | 2b | 104 | Retrospective | 12 | – | – | 96 | – | – | – | – | – | – | – | – |
| Kauffman et al, 2011 [27] | Cornell | 2b | 85 | Retrospective | 18 | 20 | 12 | 79 | 73 (2 yr) | 88 | 85 (2 yr) | 83 | 79 (2 yr) | |||
| Martin et al, 2010 [30] | Mayo Arizona | 2b | 59 | Multi-institutional | 21 | 17 | – | 82 | 71 | – | – | – | – | 82 | 72 | – |
| Tulane University | ||||||||||||||||
| Pruthi et al, 2010 [31] | UNC | 2b | 100 | Retrospective | 21.2 | 5 | 18 | – | – | – | 94 (21 mo) | – | – | 91 (21 mo) | – | – |
| Canda et al, 2012 [33] | Ankara Ataturk Training and Research Hospital | 2a | 27 | Not reported | 6 | – | 4 | 85 (6 mo) | – | – | 89 (6 mo) | – | – | 72 (6 mo) | – | – |
| Mmeje et al, 2013 [45] | Mayo Arizona | 2b | 50 | Multi-institutional | 41.5 | 12 | 46 | – | 43 | 39 | – | – | – | – | 55 | 45 |
| UNC | ||||||||||||||||
| Treiyer et al, 2012 [49] | Saarland University | 2b | 91 | Retrospective | 15 | 0 | – | – | – | – | 94 (15 mo) | – | – | 93 (15 mo) | – | – |
| Collins et al, 2013 [52] | Karolinska Institute | 2b | 113 | Prospective | 25 | 31 | – | – | – | – | – | 81 | 67 | – | 80 | 66 |
| Khan et al, 2013 [66] | Guy's & St. Thomas Hospital | 1 | 14 | Prospective | 84 | 28 | 14 | 50 | 75 | 64 | ||||||
| Nepple et al, 2013 [82] | Washington University | 2b | 36 | Retrospective | 12 | 6 | – | – | 67 (2 yr) | – | – | 75 (2 yr) | – | – | 68 (2 yr) | – |
| Snow-Lisy et al, 2014 [67] | Cleveland Clinic | 2b | 17 | Retrospective | 67 | – | – | – | – | – | – | – | 69 | – | – | 39 |
| Tyritzis et al, 2013 [56] | Karolinska Institute | 2b | 70 | Retrospective | 30.3 | 24 | – | – | 81 (2 yr) | – | – | 89 (2 yr) | – | – | 89 (2 yr) | – |
| Xylinas et al, 2013 [57] | Cornell | 2b | 175 | Retrospective | 37 | 19 | – | 67 | 63 | – | 68 | 66 | – | – | – | |
| Raza et al, 2014 [70] | Roswell Park Cancer Institute | 2b | 99 | Retrospective | 73.9 | 6 | 29 | – | – | 53 | – | – | 68 | – | – | 42 |
| Yuh et al, 2014 [58] | City of Hope Cancer Center | 2b | 162 | Retrospective | 52 | 23 | – | – | 76 | 74 | – | 83 | 80 | – | 61 | 54 |
CSS = cancer-specific survival; DFS = disease-free survival; OS = overall survival; UNC = University of North Carolina.
Several studies further analyzed the use of adjuvant chemotherapy after RARC. General indications for selecting patients for adjuvant chemotherapy included pathologic stage pT3–4 or node-positive disease. Pruthi et al described the use of adjuvant chemotherapy in 18 of 100 RARC patients, with mean time to chemotherapy initiation at approximately 7 wk, which was faster than the authors’ historical time to chemotherapy in open cystectomy of 10 wk [31] . In a randomized trial of RARC (n = 21) compared with ORC (n = 20), 7 wk was also the mean time to initiation of chemotherapy after RARC [74] . In one analysis of patients with node-positive disease at the time of RARC, 46% received adjuvant chemotherapy [45] .
Survival represents the gold standard with respect to evaluating effectiveness and risks of treatment; however, RARC reports with 5-yr outcomes have become available only recently. Data remain limited for assessing long-term outcomes, patterns of recurrence, and means for predicting survival. The role of adjuvant treatments after RARC is also poorly defined.
Series detailing cancer control outcomes had a mean follow-up between 6 and 84 mo ( Table 5 ), although only 6 of 18 series (33%) reported a mean follow-up >36 mo. At 1, 2, 3, and 5 yr, DFS was 79–96%, 67–81%, 67–76%, and 53–74%, respectively; CSS was 88–94%, 75–89%, 68–83%, and 66–80%, respectively; and OS was 82–90%, 54–89%, 61–80%, and 39–66%, respectively. In the series with longest follow-up, Khan et al described only 14 patients with ≥5 yr of follow-up, showing DFS of 50%, CSS of 75%, and OS of 64% [66] .
Several series reported on adverse oncologic outcomes associated with increased pathologic stage or lymph node involvement[58] and [93]. In a series of 162 patients with urothelial carcinoma, Yuh et al found that 5-yr survival was worse with higher pathologic stage or lymph node positivity (p < 0.01). Patients with a lymph node density of 1–10% (defined as number of positive nodes divided by number of total nodes) had DFS, CSS, and OS of 34%, 49%, and 31%, respectively, whereas patients with lymph node density >10% had further reduced survival of 30%, 38%, and 20%, respectively. Predictors of DFS were lymph node density, pathologic stage, and age-adjusted Charlson comorbidity index, whereas the same measures plus receipt of transfusion were predictive for OS [58] . Similarly, in an analysis of 99 patients with follow-up >5 yr, pathologic stage and lymph node positivity were independent predictors of DFS, CSS, and OS, whereas positive margin status and Charlson comorbidity index predicted worse OS and CSS [70] .
In series with median follow-up of >36 mo, rates of local recurrence without distant disease ranged between 0% (n = 15) and 9% (n = 99)[57], [58], [66], [67], and [70]. No port-site recurrences occurred in these series. Xylinas et al examined 175 patients with a median follow-up of 37 mo, showing recurrence of disease in 29%. Of these patients, 8 (5%) had local recurrence alone, 11 had local and distant metastases, and 32 had distant metastases alone [57] .
In an analysis of patients with positive lymph nodes (n = 50) at the time of PLND, median time to recurrence was 10 mo after RARC [45] . Estimated OS at 36 and 60 mo was 55% and 45%, respectively, with recurrence-free survival at 36 and 60 mo of 43% and 39%, respectively. Similarly, Tyritzis et al reported recurrence-free survival of 34% and OS of 63% after 24 mo in node-positive patients [56] .
Table 6 presents the RARC series reporting on continence outcomes. Although functional outcomes are a major area of study in patients undergoing RARP, a lack of data remains for evaluation after RARC. Worldwide, the number of patients evaluated for continence after orthotopic bladder substitution is <200 from nine reports at the present time. There are also widespread differences in patient selection, methods of data collection, and outcome assessment.
Table 6 Urinary continence rates in robot-assisted radical cystectomy series
| Reference | Institution | IDEAL stage | Cases, no. | Study design | Nerve-sparing surgery, % | Intracorporeal diversion, % | Follow-up, mo | Method of data collection | Continence definition | Continence rate, % | ||
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 3 mo | 6 mo | 12 mo | ||||||||||
| Mottrie et al, 2007 [9] | O.L.V.–Clinic | 2a | 27 | Retrospective | 29 | 0 | 10.2 | – | – | 86 | – | – |
| Murphy et al, 2008 [12] | Guy's Hospital | 2a | 23 | Retrospective | 20 | 0 | 17 | – | – | – | – | 100 D |
| 75 N (17 mo) |
||||||||||||
| Palou Redorta et al, 2009 [20] | Barcelona Autonomous University | 2a | 9 | Retrospective | 100 | 0 | 7 | – | – | 100 D and N (7 mo) |
||
| Canda et al, 2012 [33] | Ankara Ataturk Training and Research Hospital | 2a | 27 | Not reported | 89 | 100 | 6 | – | D: 0–1 safety pads | – | 48 | – |
| N: dry with no protection | 11 | |||||||||||
| Jonsson et al, 2011 [35] | Karolinska Institute | 2b | 36 | Prospective | 55 | 100 | 25 | – | 0–1 pads | – | – | 83 D 66 N |
| Manoharan et al, 2011 [37] | University of Miami | 2a | 14 | Retrospective | – | 0 | – | – | – | 93 D | ||
| 71 N | ||||||||||||
| Torrey et al, 2012 [41] | City of Hope Cancer Center | 2b | 34 | Retrospective | 0 | 0 (all Indiana pouch) | 12.1 | Physician charting | – | – | – | 97 |
| Goh et al, 2012 [43] | Keck School of Medicine, University of Southern California, Los Angeles | 2a | 15 | Prospective | – | 100 | 3 | – | – | 75 | ||
| Tyritzis et al, 2013 [56] | Karolinska Institute | 2b | 70 | Retrospective | 58 BNS | 100 | 12 | Internally validated questionnaire | 0–1 pads | – | D: 77 men, |
D: 88 men, |
| 8 UNS | 40 women N: 54 men, 40 women |
67 women | ||||||||||
| N: 76 men, 76 women |
||||||||||||
BNS = bilateral nerve sparing; D = daytime; N = nocturnal; UNS = unilateral nerve sparing.
Follow-up for continence evaluation varied widely, from 6 to 25 mo. Nerve-sparing procedures were performed in 20–100% of patients. Only three of six series reported using a distinct definition forcontinence, which was generally no pad or one pad (safety) per day. One of the earliest RARC series reported an 86% continent rate (seven of eight men) after 3.5 mo [9] . More recent series published 6-mo continence rates of 48–100% for daytime continence and 11–100% for nighttime continence. At 12 mo after RARC, continence rates ranged from 83% to 100% in men and were 67% in women for daytime continence and 66–76% for nighttime continence.
Using strict definitions fordaytime continence(no or one security pad per day) andnighttime continence(goodindicates dry with no protection,fairindicates dry with one awakening), Canda et al examined 23 patients with intracorporeal Studer pouch. After excluding patients who died or were lost to follow-up, 11 of 15 men (73%) and 0 of 2 women were continent during the daytime. Three of these 17 patients (18%) had good nighttime continence, and 4 (24%) had fair continence [33] .
Only one series has described continence results in patients undergoing RARC and continent cutaneous diversion. Torrey et al examined 34 patients who had RARC and Indiana pouch continent cutaneous diversion and reported 97% continence at a mean follow-up of 20 mo for both daytime and nighttime. One patient continued to experience daytime and nighttime incontinence requiring the use of pads [41] .
Table 7 summarizes the series examining potency outcomes. Similar to continence outcomes, evaluation of erectile function after RARC is not well described. Early reports suggest that erections sufficient for penetration are achievable, although sample sizes were very small and lacked validated objective evaluations. Follow-up was again too short to form definitive conclusions, with only one study reporting outcomes up to 2 yr after RARC. As noted earlier, nerve-sparing procedures were performed in 20–100% of patients. The data recording used International Index of Erectile Function (IIEF) scores in five of seven series. However, only three series provided a clear definition ofpotency[12], [35], and [56].
Table 7 Erectile function in robot-assisted radical cystectomy series
| Reference | Institution | IDEAL stage | Cases, no. | Nerve-sparing surgery, % | Study design | Follow-up, mo | Method of data collection | Potency definition | Potency rate at follow-up |
|---|---|---|---|---|---|---|---|---|---|
| Mottrie et al, 2007 [9] | O.L.V.–Clinic | 2a | 27 | 29 | Retrospective | 10.2 | – | – | 86% |
| Murphy et al, 2008 [12] | Guy's Hospital | 2a | 23 | 20 | Retrospective | 17 | IIEF | IIEF >21 with or without PDE5-I | 75% |
| Palou Redorta et al, 2009 [20] | Barcelona Autonomous University | 2a | 9 | 100 | Retrospective | 7 | – | – | 100% |
| Akbulut et al, 2011 [32] | Ankara Ataturk Training and Research Hospital | 2a | 12 | 82 bilateral 9 unilateral |
Not reported | 7.1 | IIEF | None provided | A single patient with IIEF >18 |
| Canda et al, 2012 [33] | Ankara Ataturk Training and Research Hospital | 2a | 27 | 89 | Not reported | 6 | IIEF | None provided | A single patient with IIEF >18 |
| Jonsson et al, 2011 [35] | Karolinska Institute | 2b | 36 | 55 | Prospective | 25 | IIEF | Adequate for penetration with or without PDE5-I | 41% at 12 mo 75% of patients having nerve sparing |
| Tyritzis et al, 2013 [56] | Karolinska Institute | 2b | 70 | 58 bilateral 8 unilateral |
Retrospective | 12 | IIEF | Adequate for penetration with or without PDE5-I | 63% at 12 mo |
IIEF = International Index of Erectile Function; PDE5-I = phosphodiesterase type 5 inhibitor.
In some early, small series, Mottrie et al [9] and Murphy et al [12] reported sufficient erections in six of seven and three of four men, respectively. Similar to well-described literature on RARP, phosphodiesterase type 5 inhibitors (PDE5-Is) were frequently administered to patients for penile rehabilitation after RARC; however, no comparative data in this setting have demonstrated a benefit.
Several series with intracorporeal neobladder have evaluated erectile function postoperatively, with varying results. In the experience of the Karolinska Institute, 41 of 62 men (66%) underwent nerve-sparing RARC. Of these 41 men, 26 (63%) were potent with or without the use of PDE5-Is after 12 mo [56] . In contrast, Canda et al found IIEF scores >18 in only 1 of 11 preoperatively potent men, although follow-up was shorter (6 mo) [33] .
Table 8 summarizes comparative studies evaluating lymph node yield after ORC, LRC, and RARC. In two randomized studies of ORC compared with RARC, lymph node yields were not statistically different[74] and [83]. Cumulative analyses showed no significant difference in lymph node yield between RARC and ORC (OR: 2.94; 95% CI, −0.28 to 6.15;p = 0.07) ( Fig. 2 ).
Table 8 Comparative studies evaluating lymph node yield after open, laparoscopic, and robot-assisted radical cystectomy
| Comparison | Level of evidence | Reference | Cases, no. | Study design | Extension of LND | Retrieved nodes, no. | pN+, no. (%) | Metastatic nodes, no., median |
|---|---|---|---|---|---|---|---|---|
| ORC vs RARC | 2 | |||||||
| Nix et al, 2010 [74] | 21 RARC | RCT | Standard | 19 | 4 (19) | – | ||
| 20 ORC | 18 | 7 (35) | ||||||
| Parekh et al, 2013 [83] | 20 RARC | RCT | Standard | 17.2 ± 13 | 4 (20) | – | ||
| 20 ORC | 24.2 ± 16.4 | 4 (20) | ||||||
| 3 | ||||||||
| Pruthi et al, 2008 [71] | 20 RARC | Gender matched | Standard | 19 | 2 (10) | – | ||
| 24 ORC | Retrospective | 16 | 5 (21) | |||||
| Wang et al, 2008 [72] | 33 RARC | Nonmatched | Standard | 17 | 19 | – | ||
| 21 ORC | 20 | 34 * | ||||||
| Ng et al, 2010 [73] | 83 RARC | Nonmatched | Standard | 17.9 ± 10.4 | 13 (16) | – | ||
| 104 ORC | 15.7 ± 13.2 | 24 (23) * | ||||||
| Richards et al, 2010 [75] | 35 RARC | Nonmatched | Extended | 16 | 10 (29) | – | ||
| 35 ORC | 15 | 10 (29) | ||||||
| Martin et al, 2011 [76] | 19 RARC | Nonmatched | – | 16 | – | – | ||
| 14 ORC | 13 | |||||||
| Gondo et al, 2012 [92] | 11 RARC | Nonmatched | Extended | 20.7 ± 8.2 | 9 | – | ||
| 15 ORC | 13.8 ± 6.6 * | 13 | ||||||
| Khan et al, 2012 [87] | 48 RARC | Prospective | Extended | 16 | 5 | – | ||
| 52 ORC | 11 | 15 | ||||||
| Richards et al, 2012 [77] | 20 RARC | Nonmatched | Extended | 17 | 7 (35) | – | ||
| 20 ORC (>75 yr) | 15 | 3 (15) | ||||||
| Styn et al, 2012 [78] | 50 RARC | 1:2 by age, sex, clinical stage, diversion | – | 14.3 ± 9.1 | 6 (12) | – | ||
| 100 ORC | 15.2 ± 9.5 | 19 (19) | ||||||
| Sung et al, 2012 [79] | 35 RARC | Nonmatched | Standard | 19.1 ± 8.2 | 9 (26) | – | ||
| 104 ORC | 12.9 ± 9.0 * | 27 (26) | ||||||
| Knox et al, 2013 [86] | 58 RARC | Nonmatched | Extended | 21 | 1 | – | ||
| 84 ORC | 17 | 3 | ||||||
| Maes et al, 2013 [80] | 14 RARC | Nonmatched | Extended | 11.9 | 1 (7) | – | ||
| 14 ORC | 9.5 | 5 (35) | ||||||
| Musch et al, 2014 [81] | 100 RARC | Nonmatched | – | 27.5 ± 11.0 | 20 (20) | – | ||
| 42 ORC | 19.6 ± 8.8 * | 9 (21) | ||||||
| Nepple et al, 2013 [82] | 36 RARC | Nonmatched | Standard | 17 | 8 (22) | – | ||
| 29 ORC | 14 | 7 (24) | ||||||
| 4 | Abaza et al, 2012 [88] | 35 RARC | Nonmatched | Extended | 37.5 ± 13.2 | 12 (34) | 1.5 | |
| 120 ORC | 36.9 ± 14.8 | 36 (30) | 2 | |||||
| LRC vs RARC | ||||||||
| 3 | Khan et al, 2012 [87] | 48 RARC | Prospective | Extended | 16 | 5 | – | |
| 58 LRC | 10 | 10 | ||||||
| 4 | Abraham et al, 2007 [89] | 14 RARC | Nonmatched | 10 extended | 22.3 | 2 (10) | ||
| 20 LRC | 16 extended | 16.5 | 2 (12.5) | |||||
* Statistically significant.
LND = lymph node dissection; LRC = laparoscopic radical cystectomy: ORC = open radical cystectomy; RARC = robot-assisted radical cystectomy; RCT = randomized controlled trial.
Fig. 2 Comparison of lymph node yields following robot-assisted or open radical cystectomy. CI = confidence interval; ORC = open radical cystectomy; RARC = robot-assisted radical cystectomy; SD = standard deviation.
Table 9 summarizes PSM rates in RARC, ORC, and LRC. In two randomized trials comparing RARC and ORC, Nix et al and Parekh et al did not show any increase in positive margins with RARC[74] and [83]. Cumulative analyses showed no significant difference in rates of surgical margins between RARC and ORC (5% and 7%, respectively; OR: 0.71; 95% CI, 0.46–1.1;p = 0.13) ( Fig. 3 ). In two comparative nonrandomized studies between RARC and LRC, no significant differences in PSM rates were detected (p = 0.86)[87] and [89].
Table 9 Comparative studies evaluating positive surgical margins after open, laparoscopic, and robot-assisted radical cystectomy
| Comparison | Level of evidence | Reference | Cases, no. | Pathologic stage, % | Overall PSM, no. (%) | PSM in pT2 cancer | |
|---|---|---|---|---|---|---|---|
| pT2 | pT3 | ||||||
| ORC vs RARC | 2b | ||||||
| Nix et al, 2010 [74] | 21 RARC | 67 | 14 | 0 | 0 | ||
| 20 ORC | 40 | 25 | 0 | 0 | |||
| Parekh et al, 2013 [83] | 20 RARC | 50 | 50 | 1 (5) | 0 | ||
| 20 ORC | 65 | 35 | 1 (5) | 0 | |||
| 3 | |||||||
| Rhee et al, 2006 [84] | 7 RARC | 86 | 14 | 0 | 0 | ||
| 23 ORC | 43 | 57 | 0 | 0 | |||
| Galich et al, 2006 [90] | 13 RARC | 54 | 46 | 0 | – | ||
| 24 ORC | 37 | 63 | 3 (12) | ||||
| Pruthi et al, 2007 [71] | 20 RARC | 78 | 22 | 0 | 0 | ||
| 24 ORC | 63 | 37 | 0 | ||||
| Wang et al, 2008 [72] | 33 RARC | 72 | 28 | 2 (6) | – | ||
| 21 ORC | 43 | 57 | 3 (14) | ||||
| Ng et al, 2010 [73] | 83 RARC | 61 | 39 | 6 (7) | 0 | ||
| 104 ORC | 58 | 42 | 9 (9) | 0 | |||
| Richards et al, 2010 [75] | 35 RARC | 60 | 40 | 1 (3) | – | ||
| 35 ORC | 57 | 43 | 3 (9) | ||||
| Martin et al, 2011 [76] | 19 RARC | 42 | 58 | – | – | ||
| 14 ORC | 93 | 7 | |||||
| Gondo et al, 2012 [92] | 11 RARC | 91 | 9 | 1 (9) | – | ||
| 15 ORC | 53 | 47 | 2 (13) | ||||
| Khan et al, 2012 [87] | 48 RARC | 75 | 25 | 0 | – | ||
| 52 ORC | 50 | 50 | 6 (10) | ||||
| Richards et al, 2012 [77] | 20 RARC | 60 | 40 | 1 (5) | – | ||
| 20 ORC (>75 yr) | 50 | 50 | 2 (10) | ||||
| Styn et al, 2012 [78] | 50 RARC | 60 | 40 | 1 (2) | – | ||
| 100 ORC | 72 | 28 | 1 (1) | ||||
| Sung et al, 2012 [79] | 35 RARC | 43 | 57 | – | – | ||
| 104 ORC | 38 | 62 | |||||
| Kader et al, 2013 [91] | 100 RARC | 58 | 42 | 12 (12) | – | ||
| 100 ORC | 53 | 47 | 11 (11) | ||||
| Knox et al, 2013 [86] | 58 RARC | 66 | 34 | 4 (7) | – | ||
| 84 ORC | 43 | 57 | 7 (8) | ||||
| Maes et al, 2013 [80] | 14 RARC | 43 | 57 | 3 (21) | – | ||
| 14 ORC | 57 | 43 | 2 (14) | ||||
| Musch et al, 2013 [81] | 100 RARC | 61 | 39 | 2 (2) | – | ||
| 42 ORC | 57 | 43 | 1 (2) | ||||
| Nepple et al, 2013 [82] | 36 RARC | 53 | 47 | 2 (6) | 0 | ||
| 29 ORC | 58 | 42 | 2 (7) | 0 | |||
| 4 | Abaza et al, 2012 [88] | 35 RARC | 60 | 23 | 2 (6) | 0 | |
| 120 ORC | 45 | 42 | 8 (7) | 0 | |||
| LRC vs RARC | |||||||
| 3 | Khan et al, 2012 [87] | 48 RARC | 75 | 25 | 0 | – | |
| 58 LRC | 57 | 43 | 2 (4) | ||||
| 4 | Abraham et al, 2007 [89] | 14 RARC | – | – | 1 (7) | 0 | |
| 20 LRC | 0 | – | |||||
LRC = laparoscopic radical cystectomy; ORC = open radical cystectomy; PSM = positive surgical margin; RARC = robot-assisted radical cystectomy.
Fig. 3 Comparison of positive surgical margin rates following robot-assisted or open radical cystectomy. CI = confidence interval; M-H = Mantel-Haenszel; ORC = open radical cystectomy; RARC = robot-assisted radical cystectomy.
Table 10 summarizes series that emphasized early oncologic comparisons for RARC, LRC, and ORC, though interpretation should be cautious with small series of shorter follow-up and potential bias of patient selection. A nonrandomized comparison of ORC (n = 52) with RARC (n = 48) with a follow-up of 38 mo showed disease-specific survival of 69% in the ORC group compared with 79% in the RARC group [87] . A series by Nepple et al showed similar estimates in DFS, CSS, and OS, although patients were not matched [82] .
Table 10 Comparative studies evaluating recurrence-free, cancer-specific, and overall survival estimates after open, laparoscopic, and robot-assisted radical cystectomy
| Comparison | Level of evidence | Reference | Cases, no. | Study design | Follow–up, mo | Neoadjuvant chemotherapy, % | DFS estimates, % | CSS estimates, % | OS estimates, % |
|---|---|---|---|---|---|---|---|---|---|
| ORC vs RARC | 3 | ||||||||
| Khan et al, 2012 [87] | 48 RARC | Prospective | 38 | – | – | 79 | – | ||
| 52 ORC | 69 | ||||||||
| Nepple et al, 2013 [82] | 36 RARC | Nonmatched | 12 | 6 | 67 (2 yr) | 75 (2 yr) | 68 (2 yr) | ||
| 29 ORC | 14 | 58 (2 yr) | 63 (2 yr) | 63 (2 yr) | |||||
| LRC vs RARC | 3 | ||||||||
| Khan et al, 2012 [87] | 48 RARC | Prospective | 38 | – | – | 79 | – | ||
| 58 LRC | 93 |
CSS = cancer-specific survival; DFS = disease-free survival; LRC = laparoscopic radical cystectomy; ORC = open radical cystectomy; OS = overall survival; RARC = robot-assisted radical cystectomy.
Our systematic review sought to identify and report the current state of the literature for RARC with regard to pathologic, oncologic, and functional outcomes. Various oncologic parameters, including pathologic findings and postoperative survival rates, were examined. With regard to nodal dissection, robotic ELND achieves a similar nodal yield to open ELND when performed by experienced surgeons. Nearly all RARC series reported nodal yields >15. With regard to margin rates, most series reported PSM rates of <10%, with rates of approximately 1% in pT2 disease. Although the IRCC (n = 513) reported a very high positive margin rate of 39% in pT4 patients [68] , other authors have reported rates similar to those noted in ORC series.
Although these immediate pathologic variables may act as surrogates for quality of resection, long-term survival outcomes must be analogous to those of ORC for RARC to be a viable surgical option. Currently, oncologic data are immature, and adequate comparative studies of RARC and ORC are nonexistent. In a few analyses measuring CSS and OS at 5 yr postoperatively, results appear similar to those reported in ORC; however, larger numbers and longer follow-up are needed for adequate comparison. At present, data reporting functional analysis of continence and potency recovery after RARC are inadequate to compare RARC reliably with ORC.
PLND, in conjunction with radical cystectomy, provides a staging benefit as well as a possible advantage for survival in retrospective studies. Stein et al examined 1054 patients treated with radical cystectomy and PLND with a 24% node-positive rate; these patients experienced 5- and 10-yr recurrence-free survival of 35% and 34%, respectively [2] . Although prospective validation is necessary, Leissner et al suggested that ELND improved outcomes in both low-volume node-positive and node-negative patients with greater number of lymph nodes removed [93] . The true survival benefit of ELND must be proven in a prospective fashion to overcome the Will Rogers phenomenon of apparent improved survival that results from stage migration with more thorough dissection.
Early critical concerns of RARC involved whether LND could be performed robotically with the same quality as during ORC. This review suggests that thorough robotic ELND dissection at the time of RARC is possible following a similar template as is performed during ORC. Half of current RARC series describe an extended template dissection, with the average number of lymph nodes removed between 11 and 55. In a small study of open completion LND after robotic LND, only four additional lymph nodes were recovered [34] . Although few series described the time necessary to perform a complete robotic LND, some authors described operative times approaching 2 h for the node dissection alone, suggesting that robotic LND may lengthen operative time. Further study is necessary to determine whether the LND segment of RARC is significantly longer compared with open LND. Complications specific to LND—particularly vascular injuries—were rare, as were lymphoceles, with an incidence <10%. However, complication rates may often be underreported, as reporting guidelines lack standardization.
Assessment of RARC lymph node yields as related to patient characteristics (eg, BMI) or surgeon characteristics (eg, prior RARP experience) has not shown a specific association. Similar to the ORC literature, Bochner et al reported that only extent of LND was associated with lymph node yield when examining variables such as receipt of neoadjuvant chemotherapy, pathologic stage, surgeon, and pathologist [94] . Although performance of LND was associated with higher surgeon volume, analysis of several learning curve evaluations did not find increases in lymph node yield with increasing case number. This result may seem counterintuitive, but it may be that these experienced robotic surgeons were able to translate surgical technique from RARP and PLND and thus reduce the number of cases needed to reach stable lymph node yields. Instead, a reduction in LND time could occur with experience, although it has not been specifically examined. In the IRCC database, patients with pT4 disease had lower nodal yields, possibly related to more difficult dissection or to RARC being performed for palliative intent.
PSM at cystectomy is a measure of disease burden and a predictor of outcome. In a previous study of 1589 patients who underwent radical cystectomy at Memorial Sloan Kettering Cancer Center, the positive margin rate was 4.2%. Risk factors for PSMs were female sex, higher pathologic stage, vascular invasion, mixed histology, and lymph node involvement. Patients with PSMs had a 5-yr CSS of only 32% [95] . In another large multi-institutional analysis of 4400 ORC patients, the incidence of PSMs was 6.3% [96] . A potential challenge of RARC and limitation of current robotic technology is in treating bulkier tumors because of the lack of tactile feedback.
The present systematic review demonstrates that PSMs are uncommon in RARC series and appropriately rare for pT2 disease. No significant difference was found when comparing the surgical margin rate between RARC and ORC. The high variability of positive margins across studies, between 0% and 26%, suggests significant heterogeneity in cancer characteristics, patient selection, and surgical technique and experience, among other variables. From the systematic review, the weighted average of positive margins in RARC series was 5.6%, which is comparable to the large open series cited earlier. In the aforementioned analysis of 4400 ORC patients, margin-positive rates by stage were 2.3% for pT2, 7.6% for pT3, and 24% for pT4 disease [96] . The effects of the learning curve as institutions adopted this new technology and patient selection toward earlier stage disease likely affected reported margin rates and should be considered when interpreting outcomes. Nonetheless, higher reported rates of positive margins in pT4 disease in some RARC series suggest that caution be taken for higher stage disease, with particular attention paid to the risk of margin involvement.
Several RARC series did not show decreasing margin rates with sequential case volume. A few reasons could explain this observation: (1) The positive margin numbers may be too low to detect a subgroup difference; (2) the learning curve for reducing margins at RARC could be extremely high, with a number not yet reached in smaller learning curve assessments; or (3) over time, more experienced surgeons may be more willing to take on bulky or higher stage tumors. This final hypothesis is supported by a multivariate analysis adjusting for pathologic stage that shows that differences in stage of disease accounted for an increase in margin rates with more experienced robotic surgeons [61] .
Chemotherapy use alongside surgery in the treatment of MIBC can be implemented either before or after cystectomy. While neoadjuvant chemotherapy has been shown to confer an OS advantage of 5% in randomized trials [97] , the benefit of adjuvant chemotherapy is less proven. In a recent meta-analysis of nine randomized controlled trials comprising 945 patients that investigated the use of adjuvant chemotherapy, benefits to both OS and DFS were appreciated. Patients receiving adjuvant chemotherapy after cystectomy had 23% relative risk reduction in the risk of death (OS:p = 0.049) and 34% relative decrease in the risk of disease recurrence (DFS:p = 0.014) [98] . For the current systematic review, neoadjuvant chemotherapy use was 0–31%, and adjuvant chemotherapy was delivered to 4–29% of patients. Adjuvant therapy was chiefly administered in patients with advanced-stage pT3 or higher or with positive lymph nodes. Although time to initiation of adjuvant chemotherapy was shorter by 3 wk in the analysis by Pruthi et al, further validation is required [31] .
Long-term freedom from disease recurrence and bladder cancer–related death is the primary measure of treatment efficacy with radical cystectomy. Particularly with assessments of survival, gathering data for comparison with the open standard is challenging secondary to the necessity of controlling for cancer characteristics, additional therapies, and the length of follow-up required to detect significant differences. Shorter-interval examinations of survival may not amply capture events such as local recurrence, distant recurrence, or secondary therapies. Only two series in this systematic review compared survival for RARC and ORC. These studies were not randomized and included sequential series of retrospective groups (LOE 4)[82] and [87]. In a series by Nepple et al, 36 patients who underwent RARC were compared with 29 patients who underwent ORC with a median follow-up of only 12 mo. Estimated 2-yr DFS (67% vs 58%), CSS (75% vs 63%), and OS (68% vs 63%) after RARC and ORC were similar for the two techniques, respectively [82] .
Because of limitations of present studies, comparisons must be made to large historical retrospective open series. A long-term analysis of survival in 1100 chemotherapy-naive cystectomy patients by Hautmann et al demonstrated 10-yr CSS and OS rates of 67% and 44%, respectively [99] . For this systematic review, 5-yr estimates for DFS, CSS, and OS were 53–74%, 66–80%, and 39–66%, respectively. Analogous to stratified outcomes in ORC, survival outcomes were worse in RARC series with increasing pathologic stage and with lymph node metastases. Local control of disease appears to be adequate such that the majority of recurrences after RARC are distant or outside the pelvis. A potential concern for port-site metastases with RARC remains of particular interest. Although no specific published series address this concern and most larger RARC oncologic series did not report any incidents, a few case reports suggest that this concern requires further study.
Since the original description of neurovascular bundle preservation during radical prostatectomy by Walsh et al [104] , techniques to improve functional outcomes through meticulous nerve sparing have been translated to radical cystectomy. Turner et al determined that nerve sparing improved urinary continence after orthotopic urinary diversion [100] , and nerve sparing has been shown to assist with recovery of erectile function objectively based on IIEF [101] . Long-term functional evaluations of ileal neobladder continent diversions have demonstrated daytime continence rates of 92% and nighttime continence rates of 80% [102] .
To date, very limited data are available regarding functional outcomes of continence or potency after RARC. These analyses have chiefly been limited to only a few centers that exhibit significant heterogeneity. The 12-mo reported continence rates were 88–100% in men and 67% in women for daytime continence and between 66% and 75% for nighttime continence. Potency recovery exhibited even greater variation, with sufficient erection rates between 9% and 81%. Functional outcomes are likely influenced by patient factors and selection, comorbidity, prior treatments, surgeon experience, and technique (eg, the use of cautery vs clips). In addition, methodology of reporting, definitions ofcontinence, measurement tools, rehabilitation programs, and inconsistencies in follow-up can affect the actual measurement of continence and potency. Specific functional concerns of RARC related to patient selection are that many patients may be older or have poor baseline erectile function. Moreover, technical concerns for a possible PSM, which portends a dismal outcome, may affect the performance of nerve sparing. The lack of conclusive data regarding functional recovery after RARC is a necessary area for future study.
There is no evidence to date that the results from a recent systematic review on RARP finding slight advantages to continence and potency recovery compared with open radical prostatectomy or laparoscopic radical prostatectomy extrapolate to RARC [103] . Precise definitions ofcontinenceandpotencyare necessary so that future data acquisition can be carried out in a standardized, stringent, and uniform fashion for both ORC and RARC.
From a methodological perspective, the most relevant limitations of this systematic review are the quality of the available studies, the small number of patients in and the retrospective nature of most series, the shorter-term follow-up of these studies, and the lack of standardized definitions. The papers included in the present review included only two small randomized controlled trials; the remaining series are LOE 3 or 4. Comparisons made in these single-institution studies inevitably carry the risk of selection bias. Even in randomized controlled studies, there were unlikely to have been equally experienced open and robotic surgeons operating on comparable patients. Heterogeneity in lymph node templates, sampling methods, specimen handling, and pathologic review may affect lymph node yields. Most cumulative outcomes were weighted by the results of experienced surgeons, which may make conclusions difficult to generalize. The inability to account for surgeon factors or specific technique modifications is another limitation. Most series failed to provide specific information concerning relevant aspects of the reconstructive portions of the operation.
Sufficient lymph node yields are achievable through robotic PLND if an extended template is followed. PSM rates appear similar with RARC and ORC. Conclusive long-term survival outcomes for RARC are limited, although oncologic outcomes of ≤5 yr are similar to those reported for ORC. Initial functional outcomes appear favorable; however, additional research on continence and potency after RARC is needed.
Author contributions: Bertram Yuh had full access to all the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.
Study concept and design: Yuh, Wilson, Bochner, Montorsi, Chan, Thalmann, Palou, Stenzl, Guru, Catto, Novara, Wiklund.
Acquisition of data: Yuh, Wilson, Bochner, Montorsi, Chan, Thalmann, Palou, Stenzl, Guru, Catto, Novara, Wiklund.
Analysis and interpretation of data: Yuh, Novara.
Drafting of the manuscript: Yuh, Wilson, Bochner, Montorsi, Chan, Thalmann, Palou, Stenzl, Guru, Catto, Novara, Wiklund.
Critical revision of the manuscript for important intellectual content: Yuh, Wilson, Bochner, Montorsi, Chan, Thalmann, Palou, Stenzl, Guru, Catto, Novara, Wiklund.
Statistical analysis: Yuh, Novara.
Obtaining funding: Wilson.
Administrative, technical, or material support: Yuh, Wilson, Bochner, Montorsi, Chan, Thalmann, Palou, Stenzl, Guru, Catto, Novara, Wiklund.
Supervision: Yuh, Wilson, Bochner, Montorsi, Chan, Thalmann, Palou, Stenzl, Guru, Catto, Novara, Wiklund.
Other(specify): None.
Financial disclosures: Bertram Yuh certifies that all conflicts of interest, including specific financial interests and relationships and affiliations relevant to the subject matter or materials discussed in the manuscript (eg, employment/affiliation, grants or funding, consultancies, honoraria, stock ownership or options, expert testimony, royalties, or patents filed, received, or pending), are the following: Dr. Wiklund has received proctoring assignments and a research grant from Intuitive Surgical, Sunnyvale, CA, USA; Dr. Wilson has been a consultant and a speaker for Intuitive Surgical, Sunnyvale, CA, USA.
Funding/Support and role of the sponsor: None.
Radical cystectomy and pelvic lymph node dissection (PLND) is the gold standard treatment for muscle-invasive bladder cancer (MIBC) and high-risk non–muscle-invasive disease [1] . Patients undergoing this operation can experience 66% recurrence-free survival at 10 yr after surgery [2] . The addition of neoadjuvant platinum-based chemotherapy has been shown to improve overall survival (OS) rates by approximately 5% [3] . Robot-assisted radical cystectomy (RARC) was initially described by Menon et al in 2003 [4] . Over time, many international centers have adopted RARC.
Oncologic outcomes from large population-based cohorts of RARC with lengthy follow-up are lacking. Early on in RARC history, surrogates for oncologic control were reported using positive surgical margin (PSM) rates and lymph node yields. More recently, 5-yr survival figures have become available. The majority of these outcomes, however, capture institutions early in their learning curves and incorporate patients potentially selected for the robotic technique, thus avoiding more advanced-stage or technically difficult cases. Data on functional consequences of RARC are even more limited; therefore, the quality of nerve sparing and its effect on potency recovery and continence are inadequately understood.
Because of the expanding evidence available in the field of RARC, and in preparation for the Pasadena international consensus meeting on best practice in RARC and urinary reconstruction, we performed a systematic literature review of perioperative, functional, and oncologic outcomes of RARC in comparison with open radical cystectomy (ORC) and laparoscopic radical cystectomy (LRC).
We report on the systematic review and cumulative analysis of oncologic and functional outcomes of RARC. We systematically examined lymph node yields, PSMs, cancer-specific survival (CSS), disease-free survival (DFS), and OS. In addition, functional outcomes after RARC, including urinary continence and erectile function, were systematically examined.
A systematic literature search was initially performed in September 2013 using the Medline, Scopus, and Web of Science databases. The searches included only a free-text protocol using the termsrobot-assisted radical cystectomyorda Vinci radical cystectomyorrobot* radical cystectomyin all the fields of the records for Medline and Scopus searches and in theTitleandTopicfields for the Web of Science search. No limits were applied. A full update of the searches was performed on April 28, 2014.
Two authors (G.N. and B.Y.) separately reviewed the records to select RARC case series and studies that compared RARC with ORC and RARC with LRC. Discrepancies were resolved by open discussion. Other significant studies cited in the reference lists of the selected papers were evaluated, as were studies published after the systematic search.
All noncomparative studies reporting the following data on RARC were collected: intraoperative and perioperative data (operative time, blood loss, transfusion rate, in-hospital stay, readmission, complication rates), functional data (urinary continence, erectile function), and oncologic data (PSMs, lymph node yield, DFS, CSS, OS). The present review included only studies reporting on functional and oncologic data.
Studies reporting on partial cystectomy, prostate-sparing cystectomy, salvage cystectomy, cystectomy for urachal cancer or benign disease, single-case reports, pure laparoscopic (or mixed) series, or laparoendoscopic single-site or natural orifice transluminal endoscopic surgery for radical cystectomy; experimental studies on animal models; congress abstracts; review papers; editorials; population-based studies; and book chapters were not included in the review. All data retrieved from the selected studies were recorded in an electronic database.
All papers were categorized according to the 2011 levels of evidence (LOEs) for therapy studies: LOE 1, systematic review of randomized trials or n-of-1 trials; LOE 2, randomized trial or observational study with dramatic effect; LOE 3, nonrandomized controlled cohort/follow-up study; LOE 4, case series, case–control study, or historically controlled study; or LOE 5, mechanism-based reasoning [5] . Papers were categorized according to the IDEAL recommendations [6] .
Cumulative analysis was conducted using Review Manager v5.2 software designed for composing Cochrane Reviews (Cochrane Collaboration, Oxford, UK). Statistical heterogeneity was tested using the chi-square test. Apvalue <0.10 was used to indicate heterogeneity. Where there was a lack of heterogeneity, fixed-effects models were used for the cumulative analysis. Random-effects models were used in case of heterogeneity. For continuous outcomes, the results were expressed as weighted mean differences and standard deviations (SDs); for dichotomous variables, results were given as odds ratios (ORs) and 95% confidence intervals (CIs). Because of limitations in the Review Manager v5.2 software, meta-analysis of continuous variables was possible only when rough data were presented as mean and SD. Authors of the papers were contacted to provide missing data, whenever necessary. For all statistical analyses, two-sidedp < 0.05 was considered statistically significant.
Figure 1 shows a flowchart of this systematic review of the literature.
In total, 65 surgical series[4], [7], [8], [9], [10], [11], [12], [13], [14], [15], [16], [17], [18], [19], [20], [21], [22], [23], [24], [25], [26], [27], [28], [29], [30], [31], [32], [33], [34], [35], [36], [37], [38], [39], [40], [41], [42], [43], [44], [45], [46], [47], [48], [49], [50], [51], [52], [53], [54], [55], [56], [57], [58], [59], [60], [61], [62], [63], [64], [65], [66], [67], [68], [69], and [70]and 22 comparative studies[71], [72], [73], [74], [75], [76], [77], [78], [79], [80], [81], [82], [83], [84], [85], [86], [87], [88], [89], [90], [91], and [92]reported on pathologic, oncologic (n = 18), or functional (n = 9) outcomes of RARC.
Most surgical series were retrospective, single-center studies (LOE 4). Exceptions included prospective studies[8], [22], [28], [35], [36], [43], [52], [59], [64], [66], [81], and [87]and some multi-institutional retrospective collaboration studies[16], [23], [25], [30], [45], [48], [54], [61], and [68]. Only two of the comparative studies were randomized[74] and [83](LOE 2b); all other comparative studies were nonrandomized, whether prospective or retrospective (LOE 4).
Table 1 summarizes the number of lymph nodes recovered in published RARC series. The majority of studies (86%) reported extent of lymph node dissection (LND), with more centers performing extended LND (ELND) in recent series. Standard LND typically involved the removal of obturator, internal iliac, external iliac, and some portion of the common iliac lymph nodes bilaterally. ELND templates typically brought the proximal extent up to the aortic bifurcation or inferior mesenteric artery. Approximately half of the analyzed studies reported following an extended template of dissection.
Table 1 Lymph node yields in robot-assisted radical cystectomy series
| Reference | Institution | IDEAL stage | Cases, no. | Study design | Extension of LND | Operative time, min | Retrieved nodes, no. | pN+, % | Metastatic nodes, median, no. | Complications due to LND |
|---|---|---|---|---|---|---|---|---|---|---|
| Menon et al, 2003 [4] | Henry Ford Hospital | 1 | 17 | Retrospective | Standard | – | – | 6 | – | – |
| Menon et al, 2004 [7] | Henry Ford Hospital | 1 | 3 female | Retrospective | Standard | – | 12 | 0 | – | – |
| Guru et al, 2007 [8] | Roswell Park Cancer Institute | 1 | 20 | Prospective | Standard | 44 | 13 | 15 | 1 | – |
| Mottrie et al, 2007 [9] | O.L.V.–Clinic | 2a | 27 | Retrospective | Extended | – | 23 | 9 | – | – |
| Pruthi et al, 2008 [71] | UNC | 2a | 20 | Retrospective | Standard | – | 19 | 10 | – | – |
| Hemal et al, 2008 [10] | All India Institute of Medical Sciences | 1 | 6 | Retrospective | Standard | – | 12 | 17 | – | – |
| Lowentritt et al, 2008 [11] | Tulane University | 2a | 4 | Retrospective | Standard | – | 12 | 25 | – | – |
| Murphy et al, 2008 [12] | Guy's Hospital | 2a | 23 | Retrospective | Standard | – | 16 | 9 | – | – |
| Park et al, 2008 [13] | Yonsei | 2a | 4 | Retrospective | Standard | – | 17 | 0 | – | – |
| Pruthi et al, 2008 [14] | UNC | 2a | 12 female | Retrospective | Standard, then extended | – | 19 | 17 | – | – |
| Pruthi et al, 2008 [15] | UNC | 2b | 50 | Retrospective | Standard | – | 19 | 20 | – | – |
| Wang et al, 2008 [72] | Cornell | 2b | 33 | Retrospective | Standard | – | 17 | 19 | – | – |
| Woods et al, 2008 [16] | Mayo Arizona Tulane University |
2b | 27 | Multi-institutional | Extended | – | 12.3 | 33 | 3.1 | 0 |
| Yuh et al, 2008 [17] | Roswell Park Cancer Institute | 2a | 54 | Retrospective | Extended | – | 17 | – | – | – |
| Gamboa et al, 2009 [18] | University of California, Irvine |
2a | 41 | Retrospective | Standard | – | 23 | 14 | 4 | – |
| Pruthi et al, 2009 [19] | UNC | 2b | 50 | Retrospective | Standard, then extended | – | 19 | 16 | – | – |
| 10 female | 19 | |||||||||
| 40 male | 18 | |||||||||
| Palou Redorta et al, 2009 [20] | Barcelona Autonomous University | 2a | 9 | Retrospective | Extended | 60 | 10 | 0 | – | – |
| Yuh et al, 2009 [21] | Roswell Park Cancer Institute | 2b | 73 | Retrospective | Extended | – | 19 | – | – | – |
| Guru et al, 2010 [22] | Roswell Park Cancer Institute | 2a | 26 | Prospective | Extended | – | 21 | 29 | 1 | Internal iliac artery injury: 1 |
| Hellenthal et al, 2011 [23] | IRCC | 2b | 437 | Multi-institutional | – | – | 17 | 20 | – | – |
| Josephson et al, 2010 [24] | City of Hope Cancer Center | 2b | 58 | Retrospective | Extended | – | 27 | 24 | – | – |
| Kang et al, 2010 [25] | Multicenter | 2b | 71 standard LND 33 extended LND |
Retrospective | Standard Extended |
– | 15.7 24.7 |
10 | – | – |
| Kasraeian et al, 2010 [26] | Montsouris Institute | 2a | 9 | Retrospective | Extended | – | 11 | 22 | – | 0 |
| Kauffman et al, 2011 [27] | Cornell | 2b | 85 | Retrospective | Extended | – | 19 | 15 | – | – |
| Kwon et al, 2010 [28] | Kyungpook National University | 2a | 17 | Prospective | Standard | – | 6 | 6 | 1 | 0 |
| Lavery et al, 2011 [29] | Ohio State University | 2a | 15 | Retrospective | Extended | 107 | 41.8 | 20 | – | 0 |
| Martin et al, 2010 [30] | Mayo Arizona Tulane University |
2b | 59 | Multi-institutional | Extended | – | – | 34 | – | – |
| Ng et al, 2010 [73] | Cornell | 2b | 83 | Retrospective | Standard | – | 16 | 16 | – | – |
| Nix et al, 2010 [74] | UNC | 3 | 21 | RCT | Standard | – | 19 | 19 | – | – |
| Pruthi et al, 2010 [31] | UNC | 2b | 100 | Retrospective | Standard, then extended | – | 19 | 20 | – | – |
| Richards et al, 2010 [75] | Wake Forest University | 2b | 35 | Retrospective | Extended | – | 16 | 29 | – | – |
| Akbulut et al, 2011 [32] | Ankara Ataturk Training and Research Hospital | 2a | 12 | Not reported | Extended | – | 21.3 | 42 | – | 8 |
| Canda et al, 2012 [33] | Ankara Ataturk Training and Research Hospital | 2a | 27 | Not reported | Extended | – | 24.8 | 22 | – | – |
| Davis et al, 2011 [34] | University of Texas M.D. Anderson Cancer Center | 2a | 11 | Retrospective | Extended | 117 | 43 | 9 | 1 | – |
| Jonsson et al, 2011 [35] | Karolinska Institute | 2b | 45 36 neobladder 9 ileal conduit |
Prospective | Standard Extended |
– | 19 19 27 |
20 17 33 |
– | – |
| Khan et al, 2011 [36] | Guy's Hospital | 2a | 50 | Prospective | – | – | 17 | – | – | – |
| Manoharan et al, 2011 [37] | University of Miami | 2a | 14 | Retrospective | Standard | – | 12 | – | – | – |
| Martin et al, 2011 [76] | Mayo Arizona | 2b | 19 | Retrospective | – | – | 16 | – | – | – |
| Richards et al, 2011 [38] | Wake Forest University | 2b | 60 | Retrospective | Extended | – | 17 | 30 | – | Lymphocele: 1 |
| Schumacher et al, 2011 [39] | Karolinska Institute | 2b | 45 | Retrospective | Standard 49%, extended 31% | – | 22.5 | – | 1.5 | Lymphocele: 2 |
| Shah et al, 2011 [40] | Ohio State University | 2b | 30 | Retrospective | Extended | – | – | 30 | – | – |
| Torrey et al, 2011 [41] | City of Hope Cancer Center | 2b | 34 | Retrospective | Extended | – | 28.9 | – | – | – |
| Cho et al, 2012 [42] | Hallym University College of Medicine | 2b | 35 | Retrospective | Standard | – | – | 6 | – | – |
| Goh et al, 2012 [43] | Keck School of Medicine, University of Southern California, Los Angeles | 2a | 15 | Prospective | Superextended | – | 55 | 26 | – | – |
| Lau et al, 2012 [44] | City of Hope Cancer Center | 2b | 23 (aged >80 yr) | Retrospective | Extended | – | 20.4 | 22 | – | – |
| Mmeje et al, 2013 [45] | Mayo Arizona UNC |
2b | 50 | Multi-institutional | Extended | – | 18 | 100 | 3 | – |
| Poch et al, 2012 [46] | Roswell Park Cancer Institute | 2b | 56 | Retrospective | – | – | 25 | 16 | – | – |
| Richards et al, 2012 [77] | Wake Forest University | 2b | 20 (aged >75 yr) | Retrospective | Extended | – | 17 | 35 | – | – |
| Saar et al, 2013 [47] | Saarland University | 2b | 62 | Retrospective | – | – | 14.2 | 21 | – | – |
| Smith et al, 2012 [48] | Mayo Arizona UNC, Tulane University |
2b | 227 | Multi-institutional | – | – | 18 | 20 | – | – |
| Styn et al, 2012 [78] | University of Michigan | 2b | 50 | Retrospective | – | – | 14.3 | 12 | – | – |
| Sung et al, 2012 [79] | Samsung Medical Center | 2b | 35 | Retrospective | Standard | – | 19.1 | 26 | – | Lymphocele: 1 |
| Treiyer et al, 2012 [49] | Saarland University | 2b | 91 | Retrospective | Standard | – | 14.5 | 14 | – | – |
| Tsui et al, 2012 [50] | Chang Gung Memorial Taiwan | 2a | 8 | Retrospective | Standard | – | 3 | 12.50 | – | – |
| Yuh et al, 2012 [51] | City of Hope Cancer Center | 2b | 196 | Retrospective | Extended | – | 28 | 22 | – | Lymphocele: 3 |
| Collins et al, 2013 [52] | Karolinska Institute | 2b | 113 | Prospective | Extended 56%, standard 34%, limited 5%, none 5% | – | 21 | 20 | – | Lymphocele: 5 |
| Johar et al, 2013 [53] | IRCC | 2b | 939 | Multi-institutional | – | – | 18.1 | 26 | – | – |
| Maes et al, 2013 [80] | Metro Health Hospital | 2b | 14 | Retrospective | Extended | – | 11.9 | 7 | – | – |
| Marshall et al, 2013 [54] | IRCC | 2b | 765 | Multi-institutional | Extended 58%, standard 40%, no LND 2% | – | 18 | 27 | – | – |
| Musch et al, 2014 [81] | Klinikin Essen–Mitte | 2b | 100 | Prospective | – | – | 26.5 | 20 | – | Lymphocele: 4 |
| Nazmy et al, 2014 [55] | City of Hope Cancer Center | 2b | 209 | Retrospective | Extended | – | – | 22 | – | Lymphocele: 3 |
| Nepple et al, 2013 [82] | Washington University | 2b | 36 | Retrospective | Standard | – | 17 | 22 | – | – |
| Parekh et al, 2013 [83] | University of Texas Health Sciences Center at San Antonio | 3 | 20 | RCT | Standard | – | 11 | 20 | – | – |
| Tyritzis et al, 2013 [56] | Karolinska Institute | 2b | 70 | Retrospective | Standard 43% Extended 48% |
– | 21 | 14 | – | Lymphocele: 6 Lymphedema: 1 |
| Xylinas et al, 2013 [57] | Cornell | 2b | 175 | Retrospective | Standard | – | 19 | 17 | – | Lymphocele: 2 |
| Phillips et al, 2014 [69] | Seward St. Elizabeth Medical Center | 2b | 23 (>80 yr) | Retrospective | Extended | – | 19 | – | – | |
| Raza et al, 2014 [70] | Roswell Park Cancer Institute | 2b | 99 | Retrospective | – | – | 20.7 | 36 | – | – |
| Yuh et al, 2014 [58] | City of Hope Cancer Center | 2b | 162 | Retrospective | Extended | – | 28 | 23 | – | – |
| Total | 19.3 | 23 |
IRCC = International Robotic Cystectomy Consortium; LND = lymph node dissection; RCT = randomized controlled trial; UNC = University of North Carolina.
The lymph node yield from all series was 19 (range: 3–55). Initial descriptions using a standard template of dissection achieved yields of 18 lymph nodes [59] . Number of lymph nodes recovered with an ELND ranged from 11 to 55. Abaza et al adopted a robotic template similar to the open technique, including external iliac, obturator, hypogastric, common iliac, and presacral up to the aortic bifurcation; the mean lymph node yield was 37.5 (SD: 13.2), demonstrating that lymph node counts could mirror those of open dissection if the same template was followed [88] . In a study of open completion LND after robot-assisted ELND in 11 men, Davis et al removed only an additional 4 lymph nodes with an open approach after 43 were removed with robot assistance [34] . Time of LND was rarely reported, although it ranged from 44 min in standard LND to 117 min in ELND[8] and [34]. The lymph node–positive rate was 22%. In series with >20 RARCs, lymph node–positive rates ranged from 6% to 42%. Reports of vascular injuries were rare, and lymphocele rates were 0–9%.
Table 2 summarizes the studies assessing the effects of patient characteristics and particular surgical aspects on lymph node yields in RARC series. Cumulative analysis from the International Robotic Cystectomy Consortium (IRCC) with respect to lymphadenectomy in 437 patients found a median of 17 lymph nodes removed, with a 20% node-positivity rate [23] . Patient age and sex did not affect the performance of lymphadenectomy. In a different series, increasing body mass index (BMI) did not appear to negatively affect lymph node yield, with >20 lymph nodes removed in normal, overweight, and obese patients [46] .
Table 2 Impact of patient characteristics and surgical aspects on lymph node yield in robot-assisted radical cystectomy series
| Reference | Institution | IDEAL stage |
Cases | Study design | Extension of LND | Operative time, min | Retrieved nodes, no. | pN+, % |
|---|---|---|---|---|---|---|---|---|
| Patient BMI | ||||||||
| Poch et al, 2012 [46] | Roswell Park Cancer Institute | 2b | 56 | Retrospective | – | – | 25 | 16 |
| BMI <25: 14 | 22 | 7 | ||||||
| BMI 25 to <30: 21 | 23 | 14 | ||||||
| BMI ≥30: 21 | 20 | 24 | ||||||
| Case volume | ||||||||
| Guru et al, 2008 [59] | Roswell Park Cancer Institute | 2a | 1–12 | Prospective | Extended | 46 | 33% >13 | 8 |
| 13–24 | 44 | 66% >13 | 33 | |||||
| 25–36 | 41 | 83% >13 | 25 | |||||
| 37–47 | 43 | 72% >13 | 64 | |||||
| 48–58 | 56 | 91% >13 | 18 | |||||
| Pruthi et al, 2008 [60] | University of North Carolina | 2b | 50 | Retrospective | Standard | – | 19 | – |
| 1–10 | 21 | |||||||
| 11–20 | 19 | |||||||
| 21–30 | 20 | |||||||
| 31–40 | 17 | |||||||
| 41–50 | 20 | |||||||
| Richards et al, 2011 [38] | Wake Forest University | 2b | 60 | Retrospective | Extended | – | 17 | 30 |
| 1–20 | 17 | |||||||
| 21–40 | 19.1 | |||||||
| 41–60 | 14.4 | |||||||
| Schumacher et al, 2011 [39] | Karolinska Institute | 2b | 45 | Retrospective | Standard 49% | – | 22.5 | – |
| Extended 31% | ||||||||
| 1–15 | Standard 40% | |||||||
| Extended 7% | ||||||||
| 16–30 | Standard 47% | |||||||
| Extended 53% | ||||||||
| 31–45 | Standard 60% | |||||||
| Extended 33% | ||||||||
| Prior RARP experience | ||||||||
| Hayn et al, 2010 [61] | IRCC | 2b | 496 | Retrospective | – | – | 17.8 | – |
| ≤50 previous RARP: 83 | 13.7 | |||||||
| 51–100 previous RARP: 187 | 19.8 | |||||||
| 101–150 previous RARP: 176 | 19.6 | |||||||
| >150 previous RARP: 50 | 11.8 * | |||||||
* Statistically significant.
BMI = body mass index; RCC = International Robotic Cystectomy Consortium; LND = lymph node dissection; RARP = robot-assisted radical prostatectomy.
It is interesting to note that in single-institution series, Richards et al [38] , Schumacher et al [39] , Guru et al [59] , and Pruthi et al [60] did not find higher lymph node yields with increasing sequential case numbers. However, in the IRCC, performance of lymphadenectomy was positively correlated with surgeon and institution volume but was reduced in patients with more advanced disease (pT4 stage), which may reflect operative avoidance of bulky nodal tissue.
Table 3 summarizes the occurrence of PSMs reported in the RARC series. The reported PSM rates were 5.6% (range: 0–26%). However, in series of >100 patients, margin rates ranged between 4% and 9%[48] and [53]. PSMs were reported in 1–1.5% of patients with pT2 disease and 0–25% of patients with pT3 and higher disease. PSM rates from the IRCC in 939 cases were 9% [53] .
Table 3 Positive surgical margins in robot-assisted radical cystectomy series
| Reference | Institution | IDEAL stage | Cases, no. | Study design | Pathologic stage, % | Overall PSM rate, % | PSM location | PSM rate, % | ||
|---|---|---|---|---|---|---|---|---|---|---|
| ≤pT2 | ≥pT3 | ≤pT2 | ≥pT3 | |||||||
| Menon et al, 2003 [4] | Henry Ford Hospital | 1 | 17 | Retrospective | – | – | 0 | – | 0 | 0 |
| Yohannes et al, 2003 [62] | Creighton University | 1 | 2 | Retrospective | 0 | 100 | 0 | – | 0 | 0 |
| Menon et al, 2004 [7] | Henry Ford Hospital | 1 | 3 female | Retrospective | 66 | 33 | 0 | – | 0 | 0 |
| Rhee et al, 2006 [84] | University of Virginia | 1 | 7 | Retrospective | 43 | 57 | 0 | – | 0 | 0 |
| Guru et al, 2007 [8] | Roswell Park Cancer Institute | 1 | 20 | Prospective | 40 | 60 | 15 | Prostate: 1 Ureter: 1 Vagina: 1 |
0 | 25 |
| Mottrie et al, 2007 [9] | O.L.V.–Clinic | 2a | 27 | Retrospective | 78 | 22 | 4 | Ureter: 1 | – | – |
| Pruthi et al, 2008 [71] | UNC | 2a | 20 | Retrospective | 70 | 20 | 0 | – | 0 | 0 |
| Hemal et al, 2008 [10] | All India Institute of Medical Sciences | 1 | 6 | Retrospective | 67 | 33 | 0 | – | 0 | 0 |
| Lowentritt et al, 2008 [11] | Tulane University | 2a | 4 | Retrospective | 25 | 75 | 0 | – | 0 | 0 |
| Murphy et al, 2008 [12] | Guy's Hospital | 2a | 23 | Retrospective | 74 | 17 | 0 | – | 0 | 0 |
| Park et al, 2008 [13] | Yonsei | 2a | 4 | Retrospective | 50 | 50 | 0 | – | 0 | 0 |
| Pruthi et al, 2008 [14] | UNC | 2a | 12 female | Retrospective | 58 | 25 | 0 | – | 0 | 0 |
| Pruthi et al, 2008 [15] | UNC | 2b | 50 | Retrospective | 66 | 14 | 0 | – | 0 | 0 |
| Pruthi et al, 2009 [19] | UNC | 2b | 50 10 female 40 male |
Retrospective | 66 50 70 |
18 30 15 |
0 | – | 0 | 0 |
| Wang et al, 2008 [72] | Cornell | 2b | 33 | Retrospective | 72 | 28 | 6 | Perivesical fat: 2 | 0 | 22 |
| Woods et al, 2008 [16] | Mayo Arizona Tulane University |
2b | 27 | Multi-institutional | – | – | 7 | – | 0 | – |
| Yuh et al, 2008 [17] | Roswell Park Cancer Institute | 2a | 54 | Retrospective | 44 | 56 | 13 | – | 0 | 23 |
| Gamboa et al, 2009 [18] | University of California, Irvine | 2a | 41 | Retrospective | – | – | 5 | – | 0 | – |
| Palou Redorta et al, 2009 [20] | Barcelona Autonomous University | 2a | 9 | Retrospective | 66 | 33 | 11 | – | – | – |
| Yuh et al, 2009 [21] | Roswell Park Cancer Institute | 2b | 73 | Retrospective | 45 | 55 | 10 | – | 0 | 18 |
| Guru et al, 2010 [22] | Roswell Park Cancer Institute | 2a | 20 | Prospective | 62 | 38 | 4 | – | 0 | 9 |
| Hayn et al, 2010 [61] | IRCC | 2b | 482 | Multi-institutional | 64 | 36 | 7 | – | – | – |
| Hellenthal et al, 2010 [68] | IRCC | 2b | 513 | Multi-institutional | 64 | 36 | 7 | – | 1.50 | 17 |
| Kang et al, 2010 [25] | Multicenter | 2b | 104 | Multi-institutional | 70 | 30 | 5 | – | – | – |
| Kasraeian et al, 2010 [26] | Montsouris Institute | 2a | 9 | Retrospective | 44 | 66 | 0 | – | – | – |
| Kauffman et al, 2011 [27] | Cornell | 2b | 85 | Retrospective | 64 | 36 | 6 | – | 0 | 16 |
| Kwon et al, 2010 [28] | Kyungpook National University | 2a | 17 | Prospective | 59 | 41 | 0 | – | 0 | 0 |
| Martin et al, 2010 [30] | Mayo Arizona Tulane University |
2b | 59 | Multi-institutional | 47 | 53 | – | – | – | – |
| Ng et al, 2010 [73] | Cornell | 2b | 83 | Retrospective | 61 | 39 | 7 | – | 0 | 19 |
| Nix et al, 2010 [74] | UNC | 3 | 21 | RCT | 67 | 14 | 0 | – | 0 | 0 |
| Pruthi et al, 2010 [31] | UNC | 2b | 100 | Retrospective | 67 | 13 | 0 | – | 0 | 0 |
| Richards et al, 2010 [75] | Wake Forest University | 2b | 35 | Retrospective | 60 | 40 | 3 | – | – | – |
| Akbulut et al, 2011 [32] | Ankara Ataturk Training and Research Hospital | 2a | 12 | Not reported | 58 | 42 | 0 | – | 0 | 0 |
| Canda et al, 2012 [33] | Ankara Ataturk Rraining and Research Hospital | 2a | 27 | Not reported | 56 | 44 | 4 | – | 0 | 4 |
| Davis et al, 2011 [34] | University of Texas M.D. Anderson Cancer Center | 2a | 11 | Retrospective | 92 | 8 | 0 | – | – | – |
| Jonsson et al, 2011 [35] | Karolinska Institute | 2b | 45 | Prospective | 78 | 22 | 2 | – | 0 | 10 |
| Khan et al, 2011 [36] | Guy's Hospital | 2a | 50 | Prospective | 72 | 28 | 2 | – | 0 | 7 |
| Manoharan et al, 2011 [37] | University of Miami | 2a | 14 | Retrospective | – | – | 0 | – | 0 | 0 |
| Martin et al, 2011 [76] | Mayo Arizona | 2b | 19 | Retrospective | 42 | 58 | – | – | – | – |
| Richards et al, 2011 [38] | Wake Forest University | 2b | 60 | Retrospective | 63 | 37 | 10 | – | – | – |
| Schumacher et al, 2011 [39] | Karolinska Institute | 2b | 45 | Retrospective | 78 | 22 | 2 | Ureter: 1 | 0 | 10 |
| Shah et al, 2011 [40] | Ohio State University | 2b | 30 | Retrospective | 65 | 35 | 7 | – | 0 | 22 |
| Cho et al, 2012 [42] | Hallym University College of Medicine | 2b | 35 | Retrospective | 86 | 14 | 3 | – | – | – |
| Goh et al, 2012 [43] | Keck School of Medicine, University of Southern California, Los Angeles | 2a | 15 | Prospective | 67 | 33 | 0 | – | – | – |
| Lau et al, 2012 [44] | City of Hope Cancer Center | 2b | 23 (aged >80 yr) | Retrospective | 61 | 39 | 13 | Ureter: 1 | – | – |
| Mmeje et al, 2013 [45] | Mayo Arizona | 2b | 50 | Multi-institutional | 34 | 66 | 2 | – | – | – |
| UNC | ||||||||||
| Poch et al, 2012 [46] | Roswell Park Cancer Institute | 2b | 56 | Retrospective | 55 | 45 | – | – | – | – |
| Richards et al, 2012 [77] | Wake Forest University | 2b | 20 (aged >75 yr) | Retrospective | 60 | 40 | 5 | – | – | – |
| Saar et al, 2013 [47] | Saarland University | 2b | 62 | Retrospective | 64 | 36 | 2 | – | – | – |
| Smith et al, 2012 [48] | Mayo Arizona | 2b | 227 | Multi-institutional | – | – | 2 | – | – | – |
| UNC, Tulane University | ||||||||||
| Styn et al, 2012 [78] | University of Michigan | 2b | 50 | Retrospective | 60 | 40 | 2 | – | – | – |
| Sung et al, 2012 [79] | Samsung Medical Center | 2b | 35 | Retrospective | 43 | 57 | – | – | – | – |
| Treiyer et al, 2012 [49] | Saarland University | 2b | 91 | Retrospective | 67 | 33 | 2 | Urethra: 1 Prostate: 1 | – | – |
| Tsui et al, 2012 [50] | Chang Gung Memorial | 2a | 8 | Retrospective | 75 | 25 | 0 | – | 0 | 0 |
| Yuh et al, 2012 [51] | City of Hope Cancer Center | 2b | 196 | Retrospective | 64 | 36 | 4 | – | – | – |
| Azzouni et al, 2013 [63] | Roswell Park Cancer Institute | 2b | 100 | Retrospective | 35 | 65 | 4 | – | – | – |
| Collins et al, 2013 [52] | Karolinska Institute | 2b | 113 | Prospective | 75 | 25 | 5 | Ureter: 1 | 1 | 18 |
| Johar et al, 2013 [53] | Multicenter | 2b | 939 | Retrospective | 49 | 51 | 9 | – | – | – |
| Maes et al, 2013 [80] | Metro Health Hospital | 2b | 14 | Retrospective | 43 | 57 | 21 | – | – | – |
| Marshall et al, 2013 [54] | IRCC | 2b | 765 | Multi-institutional | 59 | 41 | – | – | – | – |
| Musch et al, 2014 [81] | Klinikin Essen–Mitte | 2b | 100 | Prospective | 61 | 39 | 2 | – | – | – |
| Nazmy et al, 2014 [55] | City of Hope Cancer Center | 2b | 209 | Retrospective | 65 | 35 | 3 | – | – | – |
| Nepple et al, 2013 [82] | Washington University | 2b | 36 | Retrospective | 53 | 47 | 6 | – | 0 | 12 |
| Parekh et al, 2013 [83] | University of Texas Health Sciences Center at San Antonio | 3 | 20 | RCT | 50 | 50 | 5 | – | 0 | 10 |
| Tyritzis et al, 2013 [56] | Karolinska Institute | 2b | 70 | Retrospective | 86 | 14 | 1.5 | Ureter: 1 | 0 | 10 |
| Xylinas et al, 2013 [57] | Cornell | 2b | 175 | Retrospective | 65 | 35 | 5 | – | – | – |
| Phillips et al, 2014 [69] | Seward St. Elizabeth Medical Center | 2b | 23 (aged >80 yr) | Retrospective | 30 | 70 | 26 | – | – | – |
| Raza et al, 2014 [70] | Roswell Park Cancer Institute | 2b | 99 | Retrospective | 48 | 52 | 8 | – | – | |
| Yuh et al, 2014 [58] | City of Hope Cancer Center | 2b | 162 | Retrospective | 67 | 33 | 4 | – | – | – |
| Total | 60 | 40 | 5.6 | |||||||
IRCC = International Robotic Cystectomy Consortium; PSM = positive surgical margin; RCT = randomized controlled trial; UNC = University of North Carolina.
Table 4 summarizes the studies assessing the effects of patient characteristics and particular surgical aspects on PSM rates in RARC series. Notably, Richards et al [38] , Schumacher et al [39] , and the IRCC [68] did not demonstrate decreasing surgical margin rates with sequential case number. In a study of the role of previous robot-assisted radical prostatectomy (RARP) experience on RARC outcomes, there was a trend toward increased positive margins with increasing RARP volumes, but it did not reach statistical significance (p = 0.089) [61] . The authors chiefly attributed this situation to the performance of RARC on patients with higher risk (higher than T3) disease. One study reported that PSMs occurred only in the overweight or obese patients, although pT4 rates were much higher in those patients (26% vs 7%) [65] .
Table 4 Predictors of positive surgical margins in robot-assisted radical cystectomy series
| Reference | Institution | IDEAL stage | Cases | Study design | Pathologic stage, % | Overall PSM rate, % | PSM rate, % | ||
|---|---|---|---|---|---|---|---|---|---|
| ≤pT2 | ≥pT3 | ≤pT2 | ≥pT3 | ||||||
| Case volume | |||||||||
| Guru et al, 2008 [59] | Roswell Park Cancer Institute | 2a | 1–12 | Prospective | 33 | 66 | 17 | – | – |
| 13–24 | 58 | 42 | 25 | ||||||
| 25–36 | 50 | 50 | 0 | ||||||
| 37–47 | 46 | 54 | 9 | ||||||
| 48–58 | 64 | 36 | 0 | ||||||
| Hayn et al, 2011 [64] | Roswell Park Cancer Institute | 2a | 1–50 | Prospective | 51 | 49 | 8 | – | – |
| 51–100 | 12 | ||||||||
| 101–164 | 6 | ||||||||
| Richards et al, 2011 [38] | Wake Forest University | 2b | 60 | Retrospective | 63 | 37 | 10 | – | – |
| 1–20 | 55 | 45 | 5 | ||||||
| 21–40 | 70 | 30 | 5 | ||||||
| 41–60 | 65 | 35 | 20 | ||||||
| Schumacher et al, 2011 [39] | Karolinska Institute | 2b | 45 | Retrospective | 78 | 22 | 2 | 0 | 10 |
| 1–15 | 87 | 13 | 0 | ||||||
| 16–30 | 67 | 33 | 7 | ||||||
| 31–45 | 80 | 20 | 0 | ||||||
| Azzouni et al, 2013 [63] | Roswell Park Cancer Institute | 2b | 100 | Retrospective | 35 | 65 | 4 | – | – |
| 1–25 | 36 | 64 | 4 | ||||||
| 26–50 | 40 | 60 | 4 | ||||||
| 51–75 | 44 | 56 | 4 | ||||||
| 76–100 | 20 | 80 | 4 | ||||||
| Previous RARP experience | |||||||||
| Hayn et al, 2010 [61] | IRCC | 2b | 482 | Retrospective | 64 | 36 | 7 | – | – |
| ≤50 previous RARP: 83 | 68 | 32 | 4 | ||||||
| 51–100 previous RARP: 173 | 76 | 24 | 5 | ||||||
| 101–150 previous RARP: 168 | 54 | 46 | 9.5 | ||||||
| >150 previous RARP: 48 | 42 | 58 | 12.5 | ||||||
| Patient BMI | |||||||||
| Butt et al, 2008 [65] | Roswell Park Cancer Institute | 2a | BMI <25: 14 | Retrospective | 64 | 36 | 0 | 0 | – |
| BMI 25–29: 18 | 28 | 72 | 28 | ||||||
| BMI ≥30: 17 | 42 | 58 | 6 | ||||||
| Poch et al, 2012 [46] | Roswell Park Cancer Institute | 2b | 56 | Retrospective | 55 | 45 | – | – | – |
| BMI <25: 14 | 50 | 50 | |||||||
| BMI 25 to <30: 21 | 57 | 43 | |||||||
| BMI ≥30: 21 | 52 | 48 | |||||||
| Intracorporeal vs extracorporeal diversion | |||||||||
| Kang et al, 2012 [85] | Korea University School of Medicine | 2a | 38 extracorporeal diversion | Retrospective | 76 | 24 | 2.5 | – | – |
| 4 intracorporeal diversion | 100 | 0 | 0 | ||||||
BMI = body mass index; IRCC = International Robotic Cystectomy Consortium; PSM = positive surgical margin; RARP = robot-assisted radical prostatectomy.
Table 5 summarizes the oncologic outcomes of current RARC publications. Neoadjuvant chemotherapy use was reported in 0–31% of patients. Adjuvant chemotherapy use was reported in 4–29% of patients.
Table 5 Survival outcomes in robot-assisted radical cystectomy series
| Reference | Institution | IDEAL stage | Cases, no. | Study design | Follow-up, mo | Neoadjuvant chemotherapy, % | Adjuvant chemotherapy, % | DFS estimates, % | CSS estimates, % | OS estimates, % | ||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 1 yr | 3 yr | 5 yr | 1 yr | 3 yr | 5 yr | 1 yr | 3 yr | 5 yr | ||||||||
| Pruthi et al, 2008 [15] | UNC | 2b | 50 | Retrospective | 13.2 | 0 | 22 | – | – | – | 94 (13 mo) | – | – | 90 (13 mo) | – | – |
| Murphy et al, 2008 [12] | Guy's Hospital | 2a | 23 | Retrospective | 17 | 29 | – | 91 (17 mo) | – | – | – | – | – | – | – | – |
| Josephson et al, 2010 [24] | City of Hope Cancer Center | 2b | 58 | Retrospective | 12 | 22 | – | – | 76 (2 yr) | – | – | 76 (2 yr) | – | – | 54 (2 yr) | – |
| Kang et al, 2010 [25] | Multicenter | 2b | 104 | Retrospective | 12 | – | – | 96 | – | – | – | – | – | – | – | – |
| Kauffman et al, 2011 [27] | Cornell | 2b | 85 | Retrospective | 18 | 20 | 12 | 79 | 73 (2 yr) | 88 | 85 (2 yr) | 83 | 79 (2 yr) | |||
| Martin et al, 2010 [30] | Mayo Arizona | 2b | 59 | Multi-institutional | 21 | 17 | – | 82 | 71 | – | – | – | – | 82 | 72 | – |
| Tulane University | ||||||||||||||||
| Pruthi et al, 2010 [31] | UNC | 2b | 100 | Retrospective | 21.2 | 5 | 18 | – | – | – | 94 (21 mo) | – | – | 91 (21 mo) | – | – |
| Canda et al, 2012 [33] | Ankara Ataturk Training and Research Hospital | 2a | 27 | Not reported | 6 | – | 4 | 85 (6 mo) | – | – | 89 (6 mo) | – | – | 72 (6 mo) | – | – |
| Mmeje et al, 2013 [45] | Mayo Arizona | 2b | 50 | Multi-institutional | 41.5 | 12 | 46 | – | 43 | 39 | – | – | – | – | 55 | 45 |
| UNC | ||||||||||||||||
| Treiyer et al, 2012 [49] | Saarland University | 2b | 91 | Retrospective | 15 | 0 | – | – | – | – | 94 (15 mo) | – | – | 93 (15 mo) | – | – |
| Collins et al, 2013 [52] | Karolinska Institute | 2b | 113 | Prospective | 25 | 31 | – | – | – | – | – | 81 | 67 | – | 80 | 66 |
| Khan et al, 2013 [66] | Guy's & St. Thomas Hospital | 1 | 14 | Prospective | 84 | 28 | 14 | 50 | 75 | 64 | ||||||
| Nepple et al, 2013 [82] | Washington University | 2b | 36 | Retrospective | 12 | 6 | – | – | 67 (2 yr) | – | – | 75 (2 yr) | – | – | 68 (2 yr) | – |
| Snow-Lisy et al, 2014 [67] | Cleveland Clinic | 2b | 17 | Retrospective | 67 | – | – | – | – | – | – | – | 69 | – | – | 39 |
| Tyritzis et al, 2013 [56] | Karolinska Institute | 2b | 70 | Retrospective | 30.3 | 24 | – | – | 81 (2 yr) | – | – | 89 (2 yr) | – | – | 89 (2 yr) | – |
| Xylinas et al, 2013 [57] | Cornell | 2b | 175 | Retrospective | 37 | 19 | – | 67 | 63 | – | 68 | 66 | – | – | – | |
| Raza et al, 2014 [70] | Roswell Park Cancer Institute | 2b | 99 | Retrospective | 73.9 | 6 | 29 | – | – | 53 | – | – | 68 | – | – | 42 |
| Yuh et al, 2014 [58] | City of Hope Cancer Center | 2b | 162 | Retrospective | 52 | 23 | – | – | 76 | 74 | – | 83 | 80 | – | 61 | 54 |
CSS = cancer-specific survival; DFS = disease-free survival; OS = overall survival; UNC = University of North Carolina.
Several studies further analyzed the use of adjuvant chemotherapy after RARC. General indications for selecting patients for adjuvant chemotherapy included pathologic stage pT3–4 or node-positive disease. Pruthi et al described the use of adjuvant chemotherapy in 18 of 100 RARC patients, with mean time to chemotherapy initiation at approximately 7 wk, which was faster than the authors’ historical time to chemotherapy in open cystectomy of 10 wk [31] . In a randomized trial of RARC (n = 21) compared with ORC (n = 20), 7 wk was also the mean time to initiation of chemotherapy after RARC [74] . In one analysis of patients with node-positive disease at the time of RARC, 46% received adjuvant chemotherapy [45] .
Survival represents the gold standard with respect to evaluating effectiveness and risks of treatment; however, RARC reports with 5-yr outcomes have become available only recently. Data remain limited for assessing long-term outcomes, patterns of recurrence, and means for predicting survival. The role of adjuvant treatments after RARC is also poorly defined.
Series detailing cancer control outcomes had a mean follow-up between 6 and 84 mo ( Table 5 ), although only 6 of 18 series (33%) reported a mean follow-up >36 mo. At 1, 2, 3, and 5 yr, DFS was 79–96%, 67–81%, 67–76%, and 53–74%, respectively; CSS was 88–94%, 75–89%, 68–83%, and 66–80%, respectively; and OS was 82–90%, 54–89%, 61–80%, and 39–66%, respectively. In the series with longest follow-up, Khan et al described only 14 patients with ≥5 yr of follow-up, showing DFS of 50%, CSS of 75%, and OS of 64% [66] .
Several series reported on adverse oncologic outcomes associated with increased pathologic stage or lymph node involvement[58] and [93]. In a series of 162 patients with urothelial carcinoma, Yuh et al found that 5-yr survival was worse with higher pathologic stage or lymph node positivity (p < 0.01). Patients with a lymph node density of 1–10% (defined as number of positive nodes divided by number of total nodes) had DFS, CSS, and OS of 34%, 49%, and 31%, respectively, whereas patients with lymph node density >10% had further reduced survival of 30%, 38%, and 20%, respectively. Predictors of DFS were lymph node density, pathologic stage, and age-adjusted Charlson comorbidity index, whereas the same measures plus receipt of transfusion were predictive for OS [58] . Similarly, in an analysis of 99 patients with follow-up >5 yr, pathologic stage and lymph node positivity were independent predictors of DFS, CSS, and OS, whereas positive margin status and Charlson comorbidity index predicted worse OS and CSS [70] .
In series with median follow-up of >36 mo, rates of local recurrence without distant disease ranged between 0% (n = 15) and 9% (n = 99)[57], [58], [66], [67], and [70]. No port-site recurrences occurred in these series. Xylinas et al examined 175 patients with a median follow-up of 37 mo, showing recurrence of disease in 29%. Of these patients, 8 (5%) had local recurrence alone, 11 had local and distant metastases, and 32 had distant metastases alone [57] .
In an analysis of patients with positive lymph nodes (n = 50) at the time of PLND, median time to recurrence was 10 mo after RARC [45] . Estimated OS at 36 and 60 mo was 55% and 45%, respectively, with recurrence-free survival at 36 and 60 mo of 43% and 39%, respectively. Similarly, Tyritzis et al reported recurrence-free survival of 34% and OS of 63% after 24 mo in node-positive patients [56] .
Table 6 presents the RARC series reporting on continence outcomes. Although functional outcomes are a major area of study in patients undergoing RARP, a lack of data remains for evaluation after RARC. Worldwide, the number of patients evaluated for continence after orthotopic bladder substitution is <200 from nine reports at the present time. There are also widespread differences in patient selection, methods of data collection, and outcome assessment.
Table 6 Urinary continence rates in robot-assisted radical cystectomy series
| Reference | Institution | IDEAL stage | Cases, no. | Study design | Nerve-sparing surgery, % | Intracorporeal diversion, % | Follow-up, mo | Method of data collection | Continence definition | Continence rate, % | ||
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 3 mo | 6 mo | 12 mo | ||||||||||
| Mottrie et al, 2007 [9] | O.L.V.–Clinic | 2a | 27 | Retrospective | 29 | 0 | 10.2 | – | – | 86 | – | – |
| Murphy et al, 2008 [12] | Guy's Hospital | 2a | 23 | Retrospective | 20 | 0 | 17 | – | – | – | – | 100 D |
| 75 N (17 mo) |
||||||||||||
| Palou Redorta et al, 2009 [20] | Barcelona Autonomous University | 2a | 9 | Retrospective | 100 | 0 | 7 | – | – | 100 D and N (7 mo) |
||
| Canda et al, 2012 [33] | Ankara Ataturk Training and Research Hospital | 2a | 27 | Not reported | 89 | 100 | 6 | – | D: 0–1 safety pads | – | 48 | – |
| N: dry with no protection | 11 | |||||||||||
| Jonsson et al, 2011 [35] | Karolinska Institute | 2b | 36 | Prospective | 55 | 100 | 25 | – | 0–1 pads | – | – | 83 D 66 N |
| Manoharan et al, 2011 [37] | University of Miami | 2a | 14 | Retrospective | – | 0 | – | – | – | 93 D | ||
| 71 N | ||||||||||||
| Torrey et al, 2012 [41] | City of Hope Cancer Center | 2b | 34 | Retrospective | 0 | 0 (all Indiana pouch) | 12.1 | Physician charting | – | – | – | 97 |
| Goh et al, 2012 [43] | Keck School of Medicine, University of Southern California, Los Angeles | 2a | 15 | Prospective | – | 100 | 3 | – | – | 75 | ||
| Tyritzis et al, 2013 [56] | Karolinska Institute | 2b | 70 | Retrospective | 58 BNS | 100 | 12 | Internally validated questionnaire | 0–1 pads | – | D: 77 men, |
D: 88 men, |
| 8 UNS | 40 women N: 54 men, 40 women |
67 women | ||||||||||
| N: 76 men, 76 women |
||||||||||||
BNS = bilateral nerve sparing; D = daytime; N = nocturnal; UNS = unilateral nerve sparing.
Follow-up for continence evaluation varied widely, from 6 to 25 mo. Nerve-sparing procedures were performed in 20–100% of patients. Only three of six series reported using a distinct definition forcontinence, which was generally no pad or one pad (safety) per day. One of the earliest RARC series reported an 86% continent rate (seven of eight men) after 3.5 mo [9] . More recent series published 6-mo continence rates of 48–100% for daytime continence and 11–100% for nighttime continence. At 12 mo after RARC, continence rates ranged from 83% to 100% in men and were 67% in women for daytime continence and 66–76% for nighttime continence.
Using strict definitions fordaytime continence(no or one security pad per day) andnighttime continence(goodindicates dry with no protection,fairindicates dry with one awakening), Canda et al examined 23 patients with intracorporeal Studer pouch. After excluding patients who died or were lost to follow-up, 11 of 15 men (73%) and 0 of 2 women were continent during the daytime. Three of these 17 patients (18%) had good nighttime continence, and 4 (24%) had fair continence [33] .
Only one series has described continence results in patients undergoing RARC and continent cutaneous diversion. Torrey et al examined 34 patients who had RARC and Indiana pouch continent cutaneous diversion and reported 97% continence at a mean follow-up of 20 mo for both daytime and nighttime. One patient continued to experience daytime and nighttime incontinence requiring the use of pads [41] .
Table 7 summarizes the series examining potency outcomes. Similar to continence outcomes, evaluation of erectile function after RARC is not well described. Early reports suggest that erections sufficient for penetration are achievable, although sample sizes were very small and lacked validated objective evaluations. Follow-up was again too short to form definitive conclusions, with only one study reporting outcomes up to 2 yr after RARC. As noted earlier, nerve-sparing procedures were performed in 20–100% of patients. The data recording used International Index of Erectile Function (IIEF) scores in five of seven series. However, only three series provided a clear definition ofpotency[12], [35], and [56].
Table 7 Erectile function in robot-assisted radical cystectomy series
| Reference | Institution | IDEAL stage | Cases, no. | Nerve-sparing surgery, % | Study design | Follow-up, mo | Method of data collection | Potency definition | Potency rate at follow-up |
|---|---|---|---|---|---|---|---|---|---|
| Mottrie et al, 2007 [9] | O.L.V.–Clinic | 2a | 27 | 29 | Retrospective | 10.2 | – | – | 86% |
| Murphy et al, 2008 [12] | Guy's Hospital | 2a | 23 | 20 | Retrospective | 17 | IIEF | IIEF >21 with or without PDE5-I | 75% |
| Palou Redorta et al, 2009 [20] | Barcelona Autonomous University | 2a | 9 | 100 | Retrospective | 7 | – | – | 100% |
| Akbulut et al, 2011 [32] | Ankara Ataturk Training and Research Hospital | 2a | 12 | 82 bilateral 9 unilateral |
Not reported | 7.1 | IIEF | None provided | A single patient with IIEF >18 |
| Canda et al, 2012 [33] | Ankara Ataturk Training and Research Hospital | 2a | 27 | 89 | Not reported | 6 | IIEF | None provided | A single patient with IIEF >18 |
| Jonsson et al, 2011 [35] | Karolinska Institute | 2b | 36 | 55 | Prospective | 25 | IIEF | Adequate for penetration with or without PDE5-I | 41% at 12 mo 75% of patients having nerve sparing |
| Tyritzis et al, 2013 [56] | Karolinska Institute | 2b | 70 | 58 bilateral 8 unilateral |
Retrospective | 12 | IIEF | Adequate for penetration with or without PDE5-I | 63% at 12 mo |
IIEF = International Index of Erectile Function; PDE5-I = phosphodiesterase type 5 inhibitor.
In some early, small series, Mottrie et al [9] and Murphy et al [12] reported sufficient erections in six of seven and three of four men, respectively. Similar to well-described literature on RARP, phosphodiesterase type 5 inhibitors (PDE5-Is) were frequently administered to patients for penile rehabilitation after RARC; however, no comparative data in this setting have demonstrated a benefit.
Several series with intracorporeal neobladder have evaluated erectile function postoperatively, with varying results. In the experience of the Karolinska Institute, 41 of 62 men (66%) underwent nerve-sparing RARC. Of these 41 men, 26 (63%) were potent with or without the use of PDE5-Is after 12 mo [56] . In contrast, Canda et al found IIEF scores >18 in only 1 of 11 preoperatively potent men, although follow-up was shorter (6 mo) [33] .
Table 8 summarizes comparative studies evaluating lymph node yield after ORC, LRC, and RARC. In two randomized studies of ORC compared with RARC, lymph node yields were not statistically different[74] and [83]. Cumulative analyses showed no significant difference in lymph node yield between RARC and ORC (OR: 2.94; 95% CI, −0.28 to 6.15;p = 0.07) ( Fig. 2 ).
Table 8 Comparative studies evaluating lymph node yield after open, laparoscopic, and robot-assisted radical cystectomy
| Comparison | Level of evidence | Reference | Cases, no. | Study design | Extension of LND | Retrieved nodes, no. | pN+, no. (%) | Metastatic nodes, no., median |
|---|---|---|---|---|---|---|---|---|
| ORC vs RARC | 2 | |||||||
| Nix et al, 2010 [74] | 21 RARC | RCT | Standard | 19 | 4 (19) | – | ||
| 20 ORC | 18 | 7 (35) | ||||||
| Parekh et al, 2013 [83] | 20 RARC | RCT | Standard | 17.2 ± 13 | 4 (20) | – | ||
| 20 ORC | 24.2 ± 16.4 | 4 (20) | ||||||
| 3 | ||||||||
| Pruthi et al, 2008 [71] | 20 RARC | Gender matched | Standard | 19 | 2 (10) | – | ||
| 24 ORC | Retrospective | 16 | 5 (21) | |||||
| Wang et al, 2008 [72] | 33 RARC | Nonmatched | Standard | 17 | 19 | – | ||
| 21 ORC | 20 | 34 * | ||||||
| Ng et al, 2010 [73] | 83 RARC | Nonmatched | Standard | 17.9 ± 10.4 | 13 (16) | – | ||
| 104 ORC | 15.7 ± 13.2 | 24 (23) * | ||||||
| Richards et al, 2010 [75] | 35 RARC | Nonmatched | Extended | 16 | 10 (29) | – | ||
| 35 ORC | 15 | 10 (29) | ||||||
| Martin et al, 2011 [76] | 19 RARC | Nonmatched | – | 16 | – | – | ||
| 14 ORC | 13 | |||||||
| Gondo et al, 2012 [92] | 11 RARC | Nonmatched | Extended | 20.7 ± 8.2 | 9 | – | ||
| 15 ORC | 13.8 ± 6.6 * | 13 | ||||||
| Khan et al, 2012 [87] | 48 RARC | Prospective | Extended | 16 | 5 | – | ||
| 52 ORC | 11 | 15 | ||||||
| Richards et al, 2012 [77] | 20 RARC | Nonmatched | Extended | 17 | 7 (35) | – | ||
| 20 ORC (>75 yr) | 15 | 3 (15) | ||||||
| Styn et al, 2012 [78] | 50 RARC | 1:2 by age, sex, clinical stage, diversion | – | 14.3 ± 9.1 | 6 (12) | – | ||
| 100 ORC | 15.2 ± 9.5 | 19 (19) | ||||||
| Sung et al, 2012 [79] | 35 RARC | Nonmatched | Standard | 19.1 ± 8.2 | 9 (26) | – | ||
| 104 ORC | 12.9 ± 9.0 * | 27 (26) | ||||||
| Knox et al, 2013 [86] | 58 RARC | Nonmatched | Extended | 21 | 1 | – | ||
| 84 ORC | 17 | 3 | ||||||
| Maes et al, 2013 [80] | 14 RARC | Nonmatched | Extended | 11.9 | 1 (7) | – | ||
| 14 ORC | 9.5 | 5 (35) | ||||||
| Musch et al, 2014 [81] | 100 RARC | Nonmatched | – | 27.5 ± 11.0 | 20 (20) | – | ||
| 42 ORC | 19.6 ± 8.8 * | 9 (21) | ||||||
| Nepple et al, 2013 [82] | 36 RARC | Nonmatched | Standard | 17 | 8 (22) | – | ||
| 29 ORC | 14 | 7 (24) | ||||||
| 4 | Abaza et al, 2012 [88] | 35 RARC | Nonmatched | Extended | 37.5 ± 13.2 | 12 (34) | 1.5 | |
| 120 ORC | 36.9 ± 14.8 | 36 (30) | 2 | |||||
| LRC vs RARC | ||||||||
| 3 | Khan et al, 2012 [87] | 48 RARC | Prospective | Extended | 16 | 5 | – | |
| 58 LRC | 10 | 10 | ||||||
| 4 | Abraham et al, 2007 [89] | 14 RARC | Nonmatched | 10 extended | 22.3 | 2 (10) | ||
| 20 LRC | 16 extended | 16.5 | 2 (12.5) | |||||
* Statistically significant.
LND = lymph node dissection; LRC = laparoscopic radical cystectomy: ORC = open radical cystectomy; RARC = robot-assisted radical cystectomy; RCT = randomized controlled trial.
Fig. 2 Comparison of lymph node yields following robot-assisted or open radical cystectomy. CI = confidence interval; ORC = open radical cystectomy; RARC = robot-assisted radical cystectomy; SD = standard deviation.
Table 9 summarizes PSM rates in RARC, ORC, and LRC. In two randomized trials comparing RARC and ORC, Nix et al and Parekh et al did not show any increase in positive margins with RARC[74] and [83]. Cumulative analyses showed no significant difference in rates of surgical margins between RARC and ORC (5% and 7%, respectively; OR: 0.71; 95% CI, 0.46–1.1;p = 0.13) ( Fig. 3 ). In two comparative nonrandomized studies between RARC and LRC, no significant differences in PSM rates were detected (p = 0.86)[87] and [89].
Table 9 Comparative studies evaluating positive surgical margins after open, laparoscopic, and robot-assisted radical cystectomy
| Comparison | Level of evidence | Reference | Cases, no. | Pathologic stage, % | Overall PSM, no. (%) | PSM in pT2 cancer | |
|---|---|---|---|---|---|---|---|
| pT2 | pT3 | ||||||
| ORC vs RARC | 2b | ||||||
| Nix et al, 2010 [74] | 21 RARC | 67 | 14 | 0 | 0 | ||
| 20 ORC | 40 | 25 | 0 | 0 | |||
| Parekh et al, 2013 [83] | 20 RARC | 50 | 50 | 1 (5) | 0 | ||
| 20 ORC | 65 | 35 | 1 (5) | 0 | |||
| 3 | |||||||
| Rhee et al, 2006 [84] | 7 RARC | 86 | 14 | 0 | 0 | ||
| 23 ORC | 43 | 57 | 0 | 0 | |||
| Galich et al, 2006 [90] | 13 RARC | 54 | 46 | 0 | – | ||
| 24 ORC | 37 | 63 | 3 (12) | ||||
| Pruthi et al, 2007 [71] | 20 RARC | 78 | 22 | 0 | 0 | ||
| 24 ORC | 63 | 37 | 0 | ||||
| Wang et al, 2008 [72] | 33 RARC | 72 | 28 | 2 (6) | – | ||
| 21 ORC | 43 | 57 | 3 (14) | ||||
| Ng et al, 2010 [73] | 83 RARC | 61 | 39 | 6 (7) | 0 | ||
| 104 ORC | 58 | 42 | 9 (9) | 0 | |||
| Richards et al, 2010 [75] | 35 RARC | 60 | 40 | 1 (3) | – | ||
| 35 ORC | 57 | 43 | 3 (9) | ||||
| Martin et al, 2011 [76] | 19 RARC | 42 | 58 | – | – | ||
| 14 ORC | 93 | 7 | |||||
| Gondo et al, 2012 [92] | 11 RARC | 91 | 9 | 1 (9) | – | ||
| 15 ORC | 53 | 47 | 2 (13) | ||||
| Khan et al, 2012 [87] | 48 RARC | 75 | 25 | 0 | – | ||
| 52 ORC | 50 | 50 | 6 (10) | ||||
| Richards et al, 2012 [77] | 20 RARC | 60 | 40 | 1 (5) | – | ||
| 20 ORC (>75 yr) | 50 | 50 | 2 (10) | ||||
| Styn et al, 2012 [78] | 50 RARC | 60 | 40 | 1 (2) | – | ||
| 100 ORC | 72 | 28 | 1 (1) | ||||
| Sung et al, 2012 [79] | 35 RARC | 43 | 57 | – | – | ||
| 104 ORC | 38 | 62 | |||||
| Kader et al, 2013 [91] | 100 RARC | 58 | 42 | 12 (12) | – | ||
| 100 ORC | 53 | 47 | 11 (11) | ||||
| Knox et al, 2013 [86] | 58 RARC | 66 | 34 | 4 (7) | – | ||
| 84 ORC | 43 | 57 | 7 (8) | ||||
| Maes et al, 2013 [80] | 14 RARC | 43 | 57 | 3 (21) | – | ||
| 14 ORC | 57 | 43 | 2 (14) | ||||
| Musch et al, 2013 [81] | 100 RARC | 61 | 39 | 2 (2) | – | ||
| 42 ORC | 57 | 43 | 1 (2) | ||||
| Nepple et al, 2013 [82] | 36 RARC | 53 | 47 | 2 (6) | 0 | ||
| 29 ORC | 58 | 42 | 2 (7) | 0 | |||
| 4 | Abaza et al, 2012 [88] | 35 RARC | 60 | 23 | 2 (6) | 0 | |
| 120 ORC | 45 | 42 | 8 (7) | 0 | |||
| LRC vs RARC | |||||||
| 3 | Khan et al, 2012 [87] | 48 RARC | 75 | 25 | 0 | – | |
| 58 LRC | 57 | 43 | 2 (4) | ||||
| 4 | Abraham et al, 2007 [89] | 14 RARC | – | – | 1 (7) | 0 | |
| 20 LRC | 0 | – | |||||
LRC = laparoscopic radical cystectomy; ORC = open radical cystectomy; PSM = positive surgical margin; RARC = robot-assisted radical cystectomy.
Fig. 3 Comparison of positive surgical margin rates following robot-assisted or open radical cystectomy. CI = confidence interval; M-H = Mantel-Haenszel; ORC = open radical cystectomy; RARC = robot-assisted radical cystectomy.
Table 10 summarizes series that emphasized early oncologic comparisons for RARC, LRC, and ORC, though interpretation should be cautious with small series of shorter follow-up and potential bias of patient selection. A nonrandomized comparison of ORC (n = 52) with RARC (n = 48) with a follow-up of 38 mo showed disease-specific survival of 69% in the ORC group compared with 79% in the RARC group [87] . A series by Nepple et al showed similar estimates in DFS, CSS, and OS, although patients were not matched [82] .
Table 10 Comparative studies evaluating recurrence-free, cancer-specific, and overall survival estimates after open, laparoscopic, and robot-assisted radical cystectomy
| Comparison | Level of evidence | Reference | Cases, no. | Study design | Follow–up, mo | Neoadjuvant chemotherapy, % | DFS estimates, % | CSS estimates, % | OS estimates, % |
|---|---|---|---|---|---|---|---|---|---|
| ORC vs RARC | 3 | ||||||||
| Khan et al, 2012 [87] | 48 RARC | Prospective | 38 | – | – | 79 | – | ||
| 52 ORC | 69 | ||||||||
| Nepple et al, 2013 [82] | 36 RARC | Nonmatched | 12 | 6 | 67 (2 yr) | 75 (2 yr) | 68 (2 yr) | ||
| 29 ORC | 14 | 58 (2 yr) | 63 (2 yr) | 63 (2 yr) | |||||
| LRC vs RARC | 3 | ||||||||
| Khan et al, 2012 [87] | 48 RARC | Prospective | 38 | – | – | 79 | – | ||
| 58 LRC | 93 |
CSS = cancer-specific survival; DFS = disease-free survival; LRC = laparoscopic radical cystectomy; ORC = open radical cystectomy; OS = overall survival; RARC = robot-assisted radical cystectomy.
Our systematic review sought to identify and report the current state of the literature for RARC with regard to pathologic, oncologic, and functional outcomes. Various oncologic parameters, including pathologic findings and postoperative survival rates, were examined. With regard to nodal dissection, robotic ELND achieves a similar nodal yield to open ELND when performed by experienced surgeons. Nearly all RARC series reported nodal yields >15. With regard to margin rates, most series reported PSM rates of <10%, with rates of approximately 1% in pT2 disease. Although the IRCC (n = 513) reported a very high positive margin rate of 39% in pT4 patients [68] , other authors have reported rates similar to those noted in ORC series.
Although these immediate pathologic variables may act as surrogates for quality of resection, long-term survival outcomes must be analogous to those of ORC for RARC to be a viable surgical option. Currently, oncologic data are immature, and adequate comparative studies of RARC and ORC are nonexistent. In a few analyses measuring CSS and OS at 5 yr postoperatively, results appear similar to those reported in ORC; however, larger numbers and longer follow-up are needed for adequate comparison. At present, data reporting functional analysis of continence and potency recovery after RARC are inadequate to compare RARC reliably with ORC.
PLND, in conjunction with radical cystectomy, provides a staging benefit as well as a possible advantage for survival in retrospective studies. Stein et al examined 1054 patients treated with radical cystectomy and PLND with a 24% node-positive rate; these patients experienced 5- and 10-yr recurrence-free survival of 35% and 34%, respectively [2] . Although prospective validation is necessary, Leissner et al suggested that ELND improved outcomes in both low-volume node-positive and node-negative patients with greater number of lymph nodes removed [93] . The true survival benefit of ELND must be proven in a prospective fashion to overcome the Will Rogers phenomenon of apparent improved survival that results from stage migration with more thorough dissection.
Early critical concerns of RARC involved whether LND could be performed robotically with the same quality as during ORC. This review suggests that thorough robotic ELND dissection at the time of RARC is possible following a similar template as is performed during ORC. Half of current RARC series describe an extended template dissection, with the average number of lymph nodes removed between 11 and 55. In a small study of open completion LND after robotic LND, only four additional lymph nodes were recovered [34] . Although few series described the time necessary to perform a complete robotic LND, some authors described operative times approaching 2 h for the node dissection alone, suggesting that robotic LND may lengthen operative time. Further study is necessary to determine whether the LND segment of RARC is significantly longer compared with open LND. Complications specific to LND—particularly vascular injuries—were rare, as were lymphoceles, with an incidence <10%. However, complication rates may often be underreported, as reporting guidelines lack standardization.
Assessment of RARC lymph node yields as related to patient characteristics (eg, BMI) or surgeon characteristics (eg, prior RARP experience) has not shown a specific association. Similar to the ORC literature, Bochner et al reported that only extent of LND was associated with lymph node yield when examining variables such as receipt of neoadjuvant chemotherapy, pathologic stage, surgeon, and pathologist [94] . Although performance of LND was associated with higher surgeon volume, analysis of several learning curve evaluations did not find increases in lymph node yield with increasing case number. This result may seem counterintuitive, but it may be that these experienced robotic surgeons were able to translate surgical technique from RARP and PLND and thus reduce the number of cases needed to reach stable lymph node yields. Instead, a reduction in LND time could occur with experience, although it has not been specifically examined. In the IRCC database, patients with pT4 disease had lower nodal yields, possibly related to more difficult dissection or to RARC being performed for palliative intent.
PSM at cystectomy is a measure of disease burden and a predictor of outcome. In a previous study of 1589 patients who underwent radical cystectomy at Memorial Sloan Kettering Cancer Center, the positive margin rate was 4.2%. Risk factors for PSMs were female sex, higher pathologic stage, vascular invasion, mixed histology, and lymph node involvement. Patients with PSMs had a 5-yr CSS of only 32% [95] . In another large multi-institutional analysis of 4400 ORC patients, the incidence of PSMs was 6.3% [96] . A potential challenge of RARC and limitation of current robotic technology is in treating bulkier tumors because of the lack of tactile feedback.
The present systematic review demonstrates that PSMs are uncommon in RARC series and appropriately rare for pT2 disease. No significant difference was found when comparing the surgical margin rate between RARC and ORC. The high variability of positive margins across studies, between 0% and 26%, suggests significant heterogeneity in cancer characteristics, patient selection, and surgical technique and experience, among other variables. From the systematic review, the weighted average of positive margins in RARC series was 5.6%, which is comparable to the large open series cited earlier. In the aforementioned analysis of 4400 ORC patients, margin-positive rates by stage were 2.3% for pT2, 7.6% for pT3, and 24% for pT4 disease [96] . The effects of the learning curve as institutions adopted this new technology and patient selection toward earlier stage disease likely affected reported margin rates and should be considered when interpreting outcomes. Nonetheless, higher reported rates of positive margins in pT4 disease in some RARC series suggest that caution be taken for higher stage disease, with particular attention paid to the risk of margin involvement.
Several RARC series did not show decreasing margin rates with sequential case volume. A few reasons could explain this observation: (1) The positive margin numbers may be too low to detect a subgroup difference; (2) the learning curve for reducing margins at RARC could be extremely high, with a number not yet reached in smaller learning curve assessments; or (3) over time, more experienced surgeons may be more willing to take on bulky or higher stage tumors. This final hypothesis is supported by a multivariate analysis adjusting for pathologic stage that shows that differences in stage of disease accounted for an increase in margin rates with more experienced robotic surgeons [61] .
Chemotherapy use alongside surgery in the treatment of MIBC can be implemented either before or after cystectomy. While neoadjuvant chemotherapy has been shown to confer an OS advantage of 5% in randomized trials [97] , the benefit of adjuvant chemotherapy is less proven. In a recent meta-analysis of nine randomized controlled trials comprising 945 patients that investigated the use of adjuvant chemotherapy, benefits to both OS and DFS were appreciated. Patients receiving adjuvant chemotherapy after cystectomy had 23% relative risk reduction in the risk of death (OS:p = 0.049) and 34% relative decrease in the risk of disease recurrence (DFS:p = 0.014) [98] . For the current systematic review, neoadjuvant chemotherapy use was 0–31%, and adjuvant chemotherapy was delivered to 4–29% of patients. Adjuvant therapy was chiefly administered in patients with advanced-stage pT3 or higher or with positive lymph nodes. Although time to initiation of adjuvant chemotherapy was shorter by 3 wk in the analysis by Pruthi et al, further validation is required [31] .
Long-term freedom from disease recurrence and bladder cancer–related death is the primary measure of treatment efficacy with radical cystectomy. Particularly with assessments of survival, gathering data for comparison with the open standard is challenging secondary to the necessity of controlling for cancer characteristics, additional therapies, and the length of follow-up required to detect significant differences. Shorter-interval examinations of survival may not amply capture events such as local recurrence, distant recurrence, or secondary therapies. Only two series in this systematic review compared survival for RARC and ORC. These studies were not randomized and included sequential series of retrospective groups (LOE 4)[82] and [87]. In a series by Nepple et al, 36 patients who underwent RARC were compared with 29 patients who underwent ORC with a median follow-up of only 12 mo. Estimated 2-yr DFS (67% vs 58%), CSS (75% vs 63%), and OS (68% vs 63%) after RARC and ORC were similar for the two techniques, respectively [82] .
Because of limitations of present studies, comparisons must be made to large historical retrospective open series. A long-term analysis of survival in 1100 chemotherapy-naive cystectomy patients by Hautmann et al demonstrated 10-yr CSS and OS rates of 67% and 44%, respectively [99] . For this systematic review, 5-yr estimates for DFS, CSS, and OS were 53–74%, 66–80%, and 39–66%, respectively. Analogous to stratified outcomes in ORC, survival outcomes were worse in RARC series with increasing pathologic stage and with lymph node metastases. Local control of disease appears to be adequate such that the majority of recurrences after RARC are distant or outside the pelvis. A potential concern for port-site metastases with RARC remains of particular interest. Although no specific published series address this concern and most larger RARC oncologic series did not report any incidents, a few case reports suggest that this concern requires further study.
Since the original description of neurovascular bundle preservation during radical prostatectomy by Walsh et al [104] , techniques to improve functional outcomes through meticulous nerve sparing have been translated to radical cystectomy. Turner et al determined that nerve sparing improved urinary continence after orthotopic urinary diversion [100] , and nerve sparing has been shown to assist with recovery of erectile function objectively based on IIEF [101] . Long-term functional evaluations of ileal neobladder continent diversions have demonstrated daytime continence rates of 92% and nighttime continence rates of 80% [102] .
To date, very limited data are available regarding functional outcomes of continence or potency after RARC. These analyses have chiefly been limited to only a few centers that exhibit significant heterogeneity. The 12-mo reported continence rates were 88–100% in men and 67% in women for daytime continence and between 66% and 75% for nighttime continence. Potency recovery exhibited even greater variation, with sufficient erection rates between 9% and 81%. Functional outcomes are likely influenced by patient factors and selection, comorbidity, prior treatments, surgeon experience, and technique (eg, the use of cautery vs clips). In addition, methodology of reporting, definitions ofcontinence, measurement tools, rehabilitation programs, and inconsistencies in follow-up can affect the actual measurement of continence and potency. Specific functional concerns of RARC related to patient selection are that many patients may be older or have poor baseline erectile function. Moreover, technical concerns for a possible PSM, which portends a dismal outcome, may affect the performance of nerve sparing. The lack of conclusive data regarding functional recovery after RARC is a necessary area for future study.
There is no evidence to date that the results from a recent systematic review on RARP finding slight advantages to continence and potency recovery compared with open radical prostatectomy or laparoscopic radical prostatectomy extrapolate to RARC [103] . Precise definitions ofcontinenceandpotencyare necessary so that future data acquisition can be carried out in a standardized, stringent, and uniform fashion for both ORC and RARC.
From a methodological perspective, the most relevant limitations of this systematic review are the quality of the available studies, the small number of patients in and the retrospective nature of most series, the shorter-term follow-up of these studies, and the lack of standardized definitions. The papers included in the present review included only two small randomized controlled trials; the remaining series are LOE 3 or 4. Comparisons made in these single-institution studies inevitably carry the risk of selection bias. Even in randomized controlled studies, there were unlikely to have been equally experienced open and robotic surgeons operating on comparable patients. Heterogeneity in lymph node templates, sampling methods, specimen handling, and pathologic review may affect lymph node yields. Most cumulative outcomes were weighted by the results of experienced surgeons, which may make conclusions difficult to generalize. The inability to account for surgeon factors or specific technique modifications is another limitation. Most series failed to provide specific information concerning relevant aspects of the reconstructive portions of the operation.
Sufficient lymph node yields are achievable through robotic PLND if an extended template is followed. PSM rates appear similar with RARC and ORC. Conclusive long-term survival outcomes for RARC are limited, although oncologic outcomes of ≤5 yr are similar to those reported for ORC. Initial functional outcomes appear favorable; however, additional research on continence and potency after RARC is needed.
Author contributions: Bertram Yuh had full access to all the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.
Study concept and design: Yuh, Wilson, Bochner, Montorsi, Chan, Thalmann, Palou, Stenzl, Guru, Catto, Novara, Wiklund.
Acquisition of data: Yuh, Wilson, Bochner, Montorsi, Chan, Thalmann, Palou, Stenzl, Guru, Catto, Novara, Wiklund.
Analysis and interpretation of data: Yuh, Novara.
Drafting of the manuscript: Yuh, Wilson, Bochner, Montorsi, Chan, Thalmann, Palou, Stenzl, Guru, Catto, Novara, Wiklund.
Critical revision of the manuscript for important intellectual content: Yuh, Wilson, Bochner, Montorsi, Chan, Thalmann, Palou, Stenzl, Guru, Catto, Novara, Wiklund.
Statistical analysis: Yuh, Novara.
Obtaining funding: Wilson.
Administrative, technical, or material support: Yuh, Wilson, Bochner, Montorsi, Chan, Thalmann, Palou, Stenzl, Guru, Catto, Novara, Wiklund.
Supervision: Yuh, Wilson, Bochner, Montorsi, Chan, Thalmann, Palou, Stenzl, Guru, Catto, Novara, Wiklund.
Other(specify): None.
Financial disclosures: Bertram Yuh certifies that all conflicts of interest, including specific financial interests and relationships and affiliations relevant to the subject matter or materials discussed in the manuscript (eg, employment/affiliation, grants or funding, consultancies, honoraria, stock ownership or options, expert testimony, royalties, or patents filed, received, or pending), are the following: Dr. Wiklund has received proctoring assignments and a research grant from Intuitive Surgical, Sunnyvale, CA, USA; Dr. Wilson has been a consultant and a speaker for Intuitive Surgical, Sunnyvale, CA, USA.
Funding/Support and role of the sponsor: None.
Robotic surgery has extended exponentially in recent years. In the field of urology is where it has found one of its greatest exponents, mainly in the radical prostatectomy. Its role in other surgeries is still under analysis. Most series for robotic radical cystectomy are short, with short follow-up and many use open surgery for urinary diversion. This makes it difficult to determine the actual role of robotic surgery in radical cystectomy. In this work the authors carry out a systematic review of published series focusing on oncological and functional results. Despite the lack of long follow-up and the limitations inherent to analyzed studies, it can be seen that the quality of lymphadenectomy is maintained and that the rate of positive margins is similar to open surgery. However, more evidence regarding the long-term oncological control, long-term functional results and a standardization of the technique, which allows the systematic implementation of intracorporeal urinary diversion even are necessary in order to define the definitive role of robotic surgery applied to radical cystectomy.