Systematic review and meta-analysis of comparative studies reporting early outcomes after robot-assisted radical cystectomy versus open radical cystectomy

Systematic review and meta-analysis of comparative studies reporting early outcomes after robot-assisted radical cystectomy versus open radical cystectomy

Cancer Treatment Reviews 39 (2013) 551–560 Contents lists available at SciVerse ScienceDirect Cancer Treatment Reviews journal homepage: www.elsevie...

2MB Sizes 0 Downloads 90 Views

Cancer Treatment Reviews 39 (2013) 551–560

Contents lists available at SciVerse ScienceDirect

Cancer Treatment Reviews journal homepage: www.elsevierhealth.com/journals/ctrv

Hot Topic

Systematic review and meta-analysis of comparative studies reporting early outcomes after robot-assisted radical cystectomy versus open radical cystectomy Kaiwen Li a,b,1, Tianxin Lin a,b,1, Xinxiang Fan a,b,1, Kewei Xu a,b, Liangkuan Bi a,b, Yu Duan c, Yu Zhou d, Min Yu d, Jielin Li e, Jian Huang a,b,⇑ a

Department of Urology, Sun Yat-sen Memorial Hospital, Sun Yat-sen University, 107 W Yanjiang Road, Guangzhou 510020, China Key Laboratory of Malignant Tumor Gene Regulation and Target Therapy of Guangdong Higher Education Institutes, Sun Yat-sen University, 107 W Yanjiang Road, Guangzhou 510020, China c Clinical Medicine of Eight-year Program, Zhongshan Medical School, Sun Yat-sen University, 74 Zhongshan Second Road, Guangzhou 510000, China d Department of Hepatobilliary Surgery, Sun Yat-sen Memorial Hospital, Sun Yat-sen University, 107 W Yanjiang Road, Guangzhou 510020, China e Department of Urology, The Third Affiliated Hospital of Sun Yat-sen University, Shipai Gangding Road, Guangzhou 510000, China b

a r t i c l e

i n f o

Article history: Received 1 September 2012 Received in revised form 24 November 2012 Accepted 26 November 2012

Keywords: Urinary bladder neoplasms Cystectomy Meta-analysis Robotics

a b s t r a c t Background: Robot-assisted radical cystectomy (RARC) is increasingly being used in the management of bladder cancer. Studies comparing RARC and open radical cystectomy (ORC) have reported conflicting results. We conducted a systematic review and meta-analysis of the literature on the efficacy and advantages of RARC compared with ORC. Methods: An electronic database search of PubMed, Scopus, and the Cochrane Library was performed up to July 8, 2012. This systematic review and meta-analysis was performed based on all randomized controlled trials (RCTs) and observational comparative studies assessing the two techniques. Results: One RCT, eight studies with prospectively collected data, and four retrospective studies were identified, including 962 cases. Although RARC was associated with longer operative time (p < 0.001), patients in this group might benefit from less overall perioperative complications (p = 0.04), more lymph node yield (p = 0.009), less estimated blood loss (p < 0.001), a lower need for perioperative transfusion (p < 0.001), and shorter length of hospital stay (p < 0.001). Positive surgical margins did not differ significantly between techniques. Sensitivity analysis with prospective studies showed similar results to the original analysis, but no significant difference of lymph node yield and length of stay between two techniques. Conclusions: RARC is a mini-invasive alternative to ORC with less overall perioperative complications, more lymph node yields, less estimated blood loss, less need for a perioperative transfusion, and shorter length of stay. Ó 2012 Elsevier Ltd. All rights reserved.

Introduction The gold standard treatment for non-metastatic muscle-invasive and uncontrolled or high-risk superficial bladder cancer is open radical cystectomy (ORC) with pelvic lymph node dissection (PLND).1 Despite advances in oncologic efficacy, ORC still has a perioperative mortality ranging from 0% to 8% and a morbidity ranging from 30% to 65%.2–6 Since the first report of robot-assisted radical cystectomy (RARC) by Menon et al. in 2003,7 RARC has been adopted in many ⇑ Corresponding author at: Department of Urology, Sun Yat-sen Memorial Hospital, Sun Yat-sen University, 107 W Yanjiang Road, Guangzhou 510020, China. Tel.: +86 20 81332336; fax: +86 20 81332853. E-mail address: [email protected] (J. Huang). 1 These authors contributed equally to this work. 0305-7372/$ - see front matter Ó 2012 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.ctrv.2012.11.007

large medical centers, demonstrating its feasibility.3,8–10 However, most studies focus on longitudinal experience with RARC alone without comparison with the standard ORC.11–13 Recently, several high-quality studies comparing RARC with ORC were reported but with conflicting results.14–16 We performed a systematic review and meta-analysis of the available literature comparing RARC with ORC for bladder cancer, and we evaluated the effectiveness of a robot using perioperative and early oncologic outcomes.

Methods A prospective protocol of objectives, literature search strategies, inclusion and exclusion criteria, outcome measurements, and methods of statistical analysis was prepared a priori according

552

K. Li et al. / Cancer Treatment Reviews 39 (2013) 551–560

to the Preferred Reporting Items for Systematic Reviews and Meta-Analysis and Meta-Analysis of Observational Studies in Epidemiology recommendations for study reporting.17,18 Literature search strategy A literature search was performed in the electronic databases of PubMed, Scopus, and the Cochrane Library on July 8, 2012, restricted to the English language. The following terms and their combinations were searched in [Title/Abstract]: cystectomy, bladder resection, cystoprostatectomy, robotic, and da Vinci. The Related Articles function was also used to broaden the search. Additional studies were manually searched in the reference lists of all retrieved articles. When multiple reports describing the same population were published, the most recent or complete report was used. However, it was not applicable if the outcome measures were mutually exclusive or measured at different time periods. Inclusion and exclusion criteria, data extraction and outcomes of interest All available randomized controlled trials (RCTs) and observational comparative studies that compared RARC with ORC and had at least one of the quantitative outcomes were included. Comparative studies that selected patients with different clinical stages in two groups, as well as editorials, comments, letters to the editor, review articles, case reports, conference abstracts, and experimental animal studies were excluded. Two authors (Li k and Fan) independently extracted and summarized the data for the following parameters: authors, publication year, country, study design, matching criteria: age, gender, body mass index, American Society of Anesthesiologists (score), diversion type, clinical stage, Charlson index, neoadjuvant chemotherapy, previous abdominal/pelvic radiotherapy, previous pelvic/ abdominal surgery, and numbers of surgeon, and the outcomes of interest. Any disagreement was resolved by the adjudicating senior authors (Huang and Lin). The primary outcomes were overall perioperative (or early: within 30 d of the date of surgery)19 complication rates, positive surgical margins (PSMs; including urethral/ureteric and soft tissue PSMs) rates, and lymph node yield (LNY). If sufficient data were available, perioperative complications were subdivided into intraoperative complications and postoperative complications within 30 d of the date of surgery. Overall perioperative complications were classified according to the Clavien-Dindo grading system20 into major (grades 3–5) and minor (grades 1 and 2). Single types of early complications were also evaluated according to the classification by Ng et al.16 with some modification. PSM rates were subdivided into urethral/ureteric PSM rates and soft tissue PSM rates. The secondary outcomes were operative time, estimated blood loss (EBL), perioperative transfusion rates, and length of stay (LOS). If sufficient data were available, (1) operative time was a subgroup with three urinary diversion types: ileal conduit, orthotopic neobladder, and Indiana pouch; and (2) the units of red blood cells (RBCs) used for perioperative transfusion and intraoperative transfusion rates were studied.

Statistical analysis The meta-analyses were performed using Review Manager V.5.1.24 The weighted mean difference (WMD) and odds ratio (OR) were used to compare continuous and dichotomous variables, respectively. For studies that presented continuous data as median and range values, the means and standard deviations were calculated using statistical algorithms described by Hozo et al.25 A p < 0.05 was considered significant. Statistical heterogeneity between studies was assessed using the chi-square test with significance set at p < 0.10, and heterogeneity was quantified using the I2 statistic. A random-effect model was used for outcomes that displayed significant heterogeneity with I2 values >50%; otherwise, the fixed-effect model was used.22 Subgroup analyses were performed. Studies reporting LNY were grouped into subgroups according to the level (I, II, or III) of PLND described by Dorin et al.26 Sensitivity analyses were performed in RCTs and studies with prospectively collected data. Publication bias was assessed using the Egger tests27 and visual inspection of funnel plot. Results Thirteen studies6,14–16,28–36 including 962 cases (364 cases for RARC and 598 cases for ORC) fulfilled the predefined inclusion criteria and were included in the final analysis (Fig. 1). Two28,30 and another two16,31 publications had overlapping populations but with some different outcomes. The data of overlapping outcomes reported by Galich et al.28 or Wang et al.31 were excluded from meta-analysis because they reported lower number of cases compared with Sterrett et al.30 or Ng et al.,16 respectively. Characteristics of eligible studies and methodological quality Table 1 shows the characteristics of the included studies. There was only one RCT14 (evidence level: 2b). Eight studies declared prospective data collection15,16,28,30–34; four were retrospective studies.6,29,35,36 All of them had evidence level 3. All observational

Quality assessment Studies were rated for the level of evidence provided according to criteria by the Centre for Evidence-Based Medicine in Oxford, UK.21 The methodological quality of RCTs was assessed by the Cochrane risk of bias tool.22 The methodological quality of observational comparative studies was assessed by the modified Newcastle-Ottawa scale.23 A score of 0–9 was assigned to each study.

Fig. 1. Flow diagram of studies identified, included, and excluded. RARC = robotassisted radical cystectomy.

Table 1 Characteristics of the included studies. First author, year of publication

Level of evidence*

Country

Study design

No. of patients–

ORC (598)

Gender, RARC/ORC, No. of males

Level of PLND

Reconstruction method for urinary diversion

No. of neobladder reconstructions# RARC

ORC

Galich 200628 Sterrett 200630 Pruthi 200729 Wang 200831

3 3 3 3

USA USA USA USA

P P R P

13 19 20 33

24 33 24 21

70/70.5 69.9/65.8 62.3/68.2 70/66

10/18 14/24 20/24 29/13

I Unclear I II

Extracorporeal Unclear Extracorporeal Extracorporeal

5 NA 10 12

7 NA 5 5

Richards 20106

3

USA

R

35

35

65/66

30/25

II

Extracorporeal

3

4

3 3 2b 3

USA USA USA Korea

P P RCT R

19 83 21 35

14 104 20 104

74/68 70.9/67.2 67.4/69.3 62.2/65.9

12/14 65/73 14/17 31/85

Unclear II II II

Unclear Extracorporeal Extracorporeal Extracorporeal

NA 26 7 22

NA 29 6 19

Abaza 201233 Styn 201215

3 3

USA USA

P P

35 50

120 100

67.3/69.8 66.6/65.6

31/95 42/84

II I

Unclear Extracorporeal

NA NA

NA NA

Gondo 201234

3

Japan

P

11

15

68.9/69.7

9/13

II

Extracorporeal

4

6

Nepple 201236

3

USA

R

36

29

72/67

31/16

I

Extracorporeal

6

11

Martin 2010 Ng 201016 Nix 201014 Sung 201135

32

Match factors

Quality scores 

1, 2, 3, 5, 6, 11 1, 2, 3 2, 5, 6, 11 1, 3, 4, 5, 6, 9, 10, 11 1, 2, 3, 4, 5, 6, 8, 9, 10 1, 2, 3, 4 2, 3, 4, 5, 6, 7, 11 1, 2, 3, 4, 5, 6 1, 2, 3, 4, 6, 7, 8, 9, 10 1, 5, 6, 8 1, 2, 3, 4, 5, 6, 7, 8, 9, 10 1, 2, 3, 5, 7, 8, 9, 10, 11 3, 5, 6, 8, 11

⁄⁄⁄⁄⁄⁄⁄ ⁄⁄⁄⁄⁄⁄⁄⁄ ⁄⁄⁄⁄⁄⁄⁄⁄ ⁄⁄⁄⁄⁄⁄⁄⁄⁄ ⁄⁄⁄⁄⁄⁄⁄⁄⁄ ⁄⁄⁄⁄⁄⁄⁄ ⁄⁄⁄⁄⁄⁄⁄⁄⁄ RCT ⁄⁄⁄⁄⁄⁄⁄ ⁄⁄⁄⁄⁄⁄⁄⁄⁄ ⁄⁄⁄⁄⁄⁄⁄⁄ ⁄⁄⁄⁄⁄⁄⁄ ⁄⁄⁄⁄⁄⁄⁄⁄⁄

K. Li et al. / Cancer Treatment Reviews 39 (2013) 551–560

RARC (364)

Age, RARC/ORC, mean or median

RARC = robot-assisted radical cystectomy; ORC = open radical cystectomy; PLND = pelvic lymph node dissection; P = prospectively collected data; R = Purely retrospective; RCT = randomized controlled trial. Matching factors: 1 = age; 2 = gender; 3 = body mass index; 4 = American Society of Anesthesiology score; 5 = diversion type; 6 = clinical stage; 7 = Charlson index; 8 = neoadjuvant chemotherapy; 9 = previous abdominal/pelvic radiotherapy; 10 = previous pelvic/abdominal surgery; 11 = numbers of surgeon. * Level of evidence: according to criteria by the Centre for Evidence-Based Medicine.22   One asterisk represented one score of cohort studies using the Newcastle-Ottawa Scale.23 – Total numbers of patients were calculated excluding the data reported by Galich et al.27 and Wang et al.31 # Total numbers of patients were calculated in 10 studies excluding the data reported by Wang et al.31

553

554

K. Li et al. / Cancer Treatment Reviews 39 (2013) 551–560

comparative studies had a score of P6 and were considered high quality. Ten adopted the extracorporeal urinary diversion method.6,14–16,28,29,31,34–36 Four,15,28,29,36 seven,6,14,16,31,33–35 and 30,32 two studies declared PLND as level I, II, and unclear. PLND was declared using the same template in both groups in all studies. Compared with other studies, there was extremely higher rate of neobladders reconstructed by Sung et al.35 in the RARC than the ORC group. Its data were not combined in the following analyses: neobladders rate, complications, operation time, estimated blood loss, transfusion rates and length of stay. Similar neobladders were reconstructed between RARC and ORC groups (61 of 206 vs 68 of 227 in seven studies,6,14,16,28,29,34,36 p = 0.51). Primary outcomes Perioperative complications Pooling data of five studies that assessed overall perioperative complications in 474 patients showed significantly lower rate of overall perioperative complications in the RARC versus ORC groups (OR: 0.68; 95% CI, 0.46–0.98; p = 0.04) (Table 2 and Fig. 2).

Intraoperative and postoperative complication rates showed no significant difference between the two groups (p = 0.36 and p = 0.15, respectively). Overall perioperative complications were further divided into major and minor complications in three studies. There was no significant difference in minor complications between the two groups (p = 0.92) but significantly lower major complication rates in the RARC versus ORC groups (p = 0.002) (Table 2). Single types of overall perioperative complications were studied in six studies (Table 3). There was a significantly lower rate of myocardial infarction and pulmonary diseases (including respiratory failure and pneumonia) in the RARC versus the ORC group (p = 0.04 and p = 0.009, respectively). There was no significant difference in the other complications showed in Table 3 and the complications studied only by Ng et al.16 Positive surgical margins Pooling data of seven studies that assessed PSMs in 579 patients showed no significant difference between the RARC and ORC groups (OR: 0.85; 95% CI, 0.48–1.49; p = 0.57) (Table 2 and Fig. 3).

Table 2 Primary and secondary outcomes comparing robot-assisted radical cystectomy with open radical cystectomy. Outcomes studied

No. of studies

Reference of studies

(a) Primary outcomes Overall Intraoperative Postoperative Major Minor PSM Urethral/ureteric PSM Soft tissue PSM Lymph node yield

5 2 3 3 3 7 5 7 9

6,14–16,34

(b) Secondary outcomes Operation time, min Ileal conduit, min Orthopedic neobladder, min Pouch, min EBL,10 ml Perioperative transfusion RBCs used, units Intraoperative transfusion Length of hospital stay, d

9 2 2 1 9 3 1 3 9

6,14–16,29,30,32,34,36

31,34 28,31,34 6,16,34 6,16,34 6,15,16,28,29,34,36 6,15,28,31,36 6,15,16,28,29,34,36 6,14–16,29,33–36

31,35 31,35 31 6,14–16,29,30,32,34,36 6,15,30 16 15,34,36 6,14–16,29,30,32,34,36

No. of patients, RARC/ORC

OR/WMD 

200/274 44/36 57/60 129/154 129/154 248/331 167/209 248/331 325/549

0.68 2.94 0.53 0.36 1.02 0.85 1.38 0.85 2.25 

294/374 68/125 68/125 33/21 294/374 104/168 83/104 97/144 294/374

74.80  8.22  87.17  27.50  41.45  0.15 2.23  0.11 1.37 

95% CI

0.46, 0.29, 0.23, 0.19, 0.62, 0.48, 0.38, 0.48, 0.57,

p value*

Study heterogeneity

v2

df

I2,%

p value*

0.98 29.74 1.26 0.68 1.69 1.49 5.08 1.49 3.94

0.04 0.36 0.15 0.002 0.92 0.57 0.63 0.57 0.009

4.18 0.12 0.65 2.19 0.52 3.53 1.69 3.52 16.81

4 1 2 2 2 6 2 6 8

4 0 0 9 0 0 0 0 52

0.38 0.73 0.72 0.33 0.52 0.74 0.43 0.74 0.03

35.77, 113.84 36.89, 20.45 68.03, 106.32 3.77, 51.23 59.84, 23.07 0.07, 0.31 3.05, 1.41 0.04, 0.28 2.31, 0.43

<0.001 0.57 <0.001 0.02 <0.001 <0.001 <0.001 <0.001 <0.001

101.05 0.08 0.13 NA 50.96 5.57 NA 0.29 13.83

8 1 1 NA 8 2 NA 2 8

92 0 0 NA 84 64 NA 0 42

<0.001 0.78 0.72 NA <0.001 0.06 NA 0.86 0.09

RARC = robot-assisted radical cystectomy; ORC = open radical cystectomy; OR = odds ratio; WMD = weighted mean difference; CI = confidence interval; PSM = positive surgical margin; major = Clavien-Dindo grades (1–2); minor = Clavien-Dindo grades (3–5); NA = data not available, EBL = estimated blood loss; RBCs = red blood cells. WMD. * Statistically significant results are shown in bold.  

Fig. 2. Forest plot and meta-analysis of overall perioperative complication rates. RARC = robot-assisted radical cystectomy; ORC = open radical cystectomy; CI = confidence interval; M–H = Mantel–Haenszel test.

555

K. Li et al. / Cancer Treatment Reviews 39 (2013) 551–560 Table 3 Single types of perioperative complications. Complication types

Wound dehiscence Fistula/leak Ileus Renal failure Infectious diseases Myocardial infarction Pulmonary diseases Thromboembolism Miscellaneous Perioperative death

No. of studies

Reference of studies

3 3 4 3 3 3 3 3 3 6

14,16,28 14,16,28 14–16,28 14,16,28 14,16,28 14,16,28 14,16,28 14,16,28 14,16,28 6,14–16,28,30,31,34

No. of patients, RARC/ORC

128/137 128/137 178/237 128/137 128/137 128/137 128/137 128/137 128/137 219/307

OR

0.10 0.29 0.79 0.43 1.18 0.16 0.06 0.80 0.25 0.74

95% CI

0.04, 0.06, 0.47, 0.11, 0.55, 0.03, 0.01, 0.27, 0.04, 0.21,

1.36 1.36 1.33 1.66 2.53 0.94 0.49 2.35 1.50 2.59

p value*

0.10 0.12 0.37 0.22 0.68 0.04 0.009 0.69 0.13 0.64

Study heterogeneity

v2

df

I2,%

p value*

0.03 1.58 0.37 0.38 0.49 0.15 0.47 1.49 0.09 1.91

1 2 3 1 2 1 1 2 1 4

0 0 0 0 0 0 0 0 0 0

0.86 0.45 0.95 0.54 0.78 0.92 0.49 0.47 0.77 0.75

OR = odds ratio; RARC = robot-assisted radical cystectomy; ORC = open radical cystectomy; CI = confidence interval; fistula/leak = urinary fistula/leak or enterocutaneous fistula; ileus = ileus and small bowel obstruction; infectious diseases = sepsis, abscess of abdomen or drainage, and urinary tract infection; pulmonary = respiratory failure and pneumonia; thromboembolic = deep venous thromboembolism and pulmonary embolism; miscellaneous = dehydration and meatal diseases. * Statistically significant results are shown in bold.

Fig. 3. Forest plot and meta-analysis of positive surgical margin rates. RARC = robot-assisted radical cystectomy; ORC = open radical cystectomy; CI = confidence interval; M– H = Mantel–Haenszel test.

Urethral/ureteric and soft tissue PSMs were available for five studies and seven studies, respectively, and they showed no significant differences between the two groups (p = 0.63 and p = 0.57, respectively) (Table 2). Lymph node yield Pooling data of nine studies that counted LNY in 874 patients showed significantly more LNY in RARC than the ORC group (WMD: 2.25; 95% CI, 0.57–3.94; p = 0.009) (Table 2 and Fig. 4). Secondary outcomes Operative time Pooling data of nine studies including 668 patients showed significantly longer operative time in the RARC than the ORC group (p < 0.001) (Table 2 and Fig. 5). When calculating the operative time of radical cystectomy using different urinary diversion methods, operative time of ileal conduit, neobladder, and pouch were available for two studies, two studies, and one study, respectively, and showed no significant differences except in the ileal conduit method between the two groups (p = 0.57, p < 0.001, and p = 0.02; respectively) (Table 2).

Three studies reported perioperative transfusion in 272 patients, and the pooled data showed a significantly lower rate in the RARC than the ORC group (p < 0.001) (Table 2 and Fig. 7). When the units of RBC used were counted, just one study showed a significant difference favoring the RARC group (p < 0.001). Pooled data of three studies that evaluated intraoperative transfusion showed a significant difference favoring the RARC group (p < 0.001) (Table 2). However, no data of postoperative transfusion could be obtained in all studies. Length of stay Pooling data of nine studies including 668 patients that evaluated LOS showed significantly shorter in the RARC versus ORC groups (p < 0.001) (Table 2 and Fig. 8). Subgroup analysis Three and six studies reported PLND to level I and level II, respectively. No study declared PLND to level III. There was significantly more LNY of PLND to level II in RARC than ORC (p = 0.04). The data of three studies with PLND up to level I showed no significant difference of LNY between two groups (p = 0.17) (Fig. 4). Sensitivity analysis and publication bias

Estimated blood loss and perioperative transfusion Pooling data of nine studies including 668 patients that evaluated EBL showed significantly lower blood loss in the RARC than the ORC group (p < 0.001) (Table 2 and Fig. 6).

One RCT14 and eight studies15,16,28,30–34 with prospectively collected data were included in the sensitivity analysis. The degree of between-study heterogeneity decreased slightly for LNY and

556

K. Li et al. / Cancer Treatment Reviews 39 (2013) 551–560

Fig. 4. Forest plot and meta-analysis of lymph node yield. RARC = robot-assisted radical cystectomy; ORC = open radical cystectomy; CI = confidence interval; IV = inverse variance.

Fig. 5. Forest plot and meta-analysis of operation time. RARC = robot-assisted radical cystectomy; ORC = open radical cystectomy; CI = confidence interval; IV = inverse variance.

operative time but not for overall perioperative complications, EBL, and overall perioperative transfusion and LOS. Between-study heterogeneity remained statistically significant for operative time, EBL. However, between-study heterogeneity for LOS was significantly different compared with the original analysis (p = 0.03 vs p = 0.09) (Table 4). Fig. 9 shows a funnel plot of the studies that reported overall perioperative complication rates. All studies lie inside the 95% CIs, with an even distribution around the vertical (Egger’s test: p = 0.846), indicating no obvious publication bias.

Discussion This was the first systematic review and meta-analysis of comparative studies of RARC versus ORC for bladder cancer. One RCT, eight retrospective studies with prospectively collected data, and 4 purely retrospective studies including 962 patients were included. Pooled data indicated significantly lower overall perioperative complications, more LNY, longer operative time, less EBL, lower perioperative transfusion, lower intraoperative transfusion, and shorter LOS in the RARC than the ORC group.

K. Li et al. / Cancer Treatment Reviews 39 (2013) 551–560

557

Fig. 6. Forest plot and meta-analysis of estimated blood loss. RARC = robot-assisted radical cystectomy; ORC = open radical cystectomy; CI = confidence interval; IV = inverse variance.

Fig. 7. Forest plot and meta-analysis of perioperative transfusion rates. RARC = robot-assisted radical cystectomy; ORC = open radical cystectomy; CI = confidence interval; M–H = Mantel–Haenszel test.

Fig. 8. Forest plot and meta-analysis of length of stay. RARC = robot-assisted radical cystectomy; ORC = open radical cystectomy; CI = confidence interval; IV = inverse variance.

No significant differences in PSMs between the two groups were found. Perioperative complications are one of the most important early end points in the evaluation of any surgical procedure. Overall perioperative complication rates showed significantly lower in the RARC group in this review, indicating that RARC might be safer and more effective than ORC. However, perioperative complication rates and mortality rates were 50.5% and 1.6% for the RARC group and 60.9% and 0.9% for the ORC group, which are seen in other studies regarding cystectomy2–5 without much improvement. Although RARC requires a 45° Trendelenberg position7 and might cause some adverse effects, such as intraoperative and postoperative pulmonary disorders, less pulmonary complications were

found in the RARC versus ORC group. A possible reason for this observation is reduced pain from robotic cystectomy, which could promote earlier pulmonary rehabilitation.16 Fewer major complications were also found in patients in the RARC group, indicating fewer additional treatments were needed. For cystectomy, Ng et al.16 reported that robotic cystectomy, less EBL, less American Society of Anesthesiologists (ASA) score were associated with lower rates of overall perioperative complications, while robotic cystectomy, less EBL, less ASA score, less Charlson index were associated with major complications. The ASA scores in the five studies6,14–16,34 regarding to overall perioperative complications and the Charlson index or ASA scores in the three studies6,16,34 regarding to major complications were matched in patients

558

K. Li et al. / Cancer Treatment Reviews 39 (2013) 551–560

Table 4 Sensitivity analysis of one randomized controlled trial and nine studies with prospectively collected data. Outcomes studied

No. studies

Reference of studies

(a) Primary outcomes Overall perioperative complications Intraoperative Postoperative Major Minor PSM Urethral/ureteric PSM Soft tissue PSM Lymph node yield

4 2 3 2 2 4 3 5 5

14–16,34

(b) Secondary outcomes Operation time, min Ileal conduit, min Orthotopic neobladder, min Pouch, min EBL, 10 ml Perioperative transfusion RBCs used, U Intraoperative transfusion Length of hospital stay, d

6 1 1 1 6 2 1 2 6

14–16,30,32,34

31,34 28,31,34 16,34 16,34 15,16,28,34 15,28,31 14,15,28,31,33 14–16,33,34

31 31 31 14–16,30,32,34 15,30 16 15,34 14–16,30,32,34

No. patients, RARC/ORC

OR/WMD 

165/239 44/36 57/60 94/119 94/119 157/243 96/145 152/285 109/357

0.66 2.94 0.53 0.26 1.03 0.93 3.13 0.88 1.13 

203/286 33/21 33/21 33/21 203/286 69/133 83/104 61/115 203/286

56.69  3.80  88.50  27.50  36.10  0.21 2.23  0.08 1.73 

95% CI

p value*

Study heterogeneity

v2

df

I2,%

p value*

0.44, 0.99 0.29, 29.74 0.23, 1.26 0.11, 0.59 0.58, 1.85 0.48, 1.81 0.51, 19.36 0.42, 1.85 1.08, 3.34

0.04 0.36 0.15 0.001 0.91 0.84 NA 0.73 0.32

4.07 0.12 0.65 0.09 0.51 2.36 NA 2.43 7.06

3 1 2 1 1 3 NA 3 4

26 0 0 0 0 0 NA 0 43

0.25 0.73 0.72 0.77 0.47 0.50 NA 0.34 0.13

11.17, 102.21 46.01, 38.41 68.07, 108.93 3.77, 51.23 61.69, 10.52 0.08, 0.52 3.05, 1.41 0.01, 0.44 3.71, 0.24

0.01 0.86 <0.001 0.02 0.006 <0.001 <0.001 0.004 0.08

45.49 NA NA NA 41.94 3.30 NA 0.03 12.41

5 NA NA NA 5 1 NA 1 5

89 NA NA NA 88 70 NA 0 60

<0.001 NA NA NA <0.001 0.07 NA 0.86 0.03

RARC = robot-assisted radical cystectomy; ORC = open radical cystectomy; OR = odds ratio; WMD = weighted mean difference; CI = confidence interval; PSM = positive surgical margin; EBL = estimated blood loss; major = Clavien-Dindo grades (1–2); minor = Clavien-Dindo grades (3–5);   WMD. * Statistically significant results were shown in bold.

Fig. 9. Funnel plots illustrating meta-analysis of overall perioperative complication rates. SE = standard error; OR = odds ratio.

between RARC and ORC groups. In the meanwhile, EBL in the present meta-analysis was less in the RARC group, which might reduce complications too. Surgical margin status at cystectomy has been cited as critical to both disease-specific and overall survival. In 2004, Herr37 and coworkers reviewed 1091 cystectomy cases at four institutions and proposed standards for radical cystectomy and pelvic lymph node dissection for bladder cancer. This report suggests that an acceptable positive margin rate is fewer than 10% of all cases and fewer than 15% for bulky (pT3–pT4) tumors. We found 8.5% and 10.3% of surgical margins were positive in RARC and ORC groups, respectively, both of which were also comparable with other ORCs reported.38 When assessing surgical margin status, pathologic stages are essential. Two studies16,34 reported all 19 patients with

overall PSMs had pT3/T4 disease. Four studies6,31,33,34 reported all 21 patients with soft tissue PSMs had pT3/T4 disease. Another two studies6,28 reported two patients with urethral/ureteric PSMs presenting carcinoma in situ in the open group. No PSMs were reported in patients having 6pT2 disease. However, we cannot conclude that if the RARC can yield equivalent surgical margin rates compared with ORC in patients with extravesical disease in the present meta-analysis. Indeed, a study39 by the International Robotic Cystectomy Consortium showed that the surgical margin status for RARC with extravesical disease (=pT3) was 16.6%, whereas the status for the ORC series reported by a recent large multicenter study with 4410 patients was 12.4%. Thus, RARC seems unable to yield equivalent surgical margin status in patients with extravesical disease. Further well-designed comparative studies should take such issue into consideration. The adequacy of PLND and the number of nodes retrieved has also been the subject of criticism of minimally invasive cystectomy.40 It is expected that better three-dimensional visualization will facilitate more surgical precision in the extirpation of local disease in RARC.40 We found more lymph nodes were yielded in RARC in the original analysis. However, in the only one RCT and the following sensitivity analysis, there was no significant difference between the two techniques. We might be more likely to make a conclusion according to the results of the one RCT and the sensitivity analysis. Dorin et al. reported that the dissection of well-defined anatomic areas is more important than the number of nodes removed.26 We found in level II PLND that pooled data showed more LNY in the RARC group. Because no studies declared different PLND templates in the two groups, the impact of different areas between the two groups was not the main focus of this meta-analysis. Only Khan et al. reported the number of positive nodes was 5% versus 15% in RARC versus ORC.35 However, variation within patients in lymph node numbers may also affect the number of nodes retrieved. It might also contribute to the difference of LNY between the two groups. In their early experience, surgeons may be reluctant to construct orthotopic reservoirs. However, except for the study reported by Sung et al.35, we found similar rates of orthotopic

K. Li et al. / Cancer Treatment Reviews 39 (2013) 551–560

neobladders reconstruction between two groups, and most were performed extracorporeally (Table 1). If extracorporeal diversions are made, the da Vinci system must be undocked after cystectomy and PLND. However, because it takes only a few minutes to undock the da Vinci system, the undocking is not a valid reason for operative time during RARC with extracorporeal urinary diversion. Previous experience may be associated with operative time before the initiation of robotic surgery.41 Although the previous experience in 15–75 robot-assisted and 300–400 open radical cystectomies or prostatectomies was reported14,15,28,33,34 and most declared surgeons were fellowship trained, the initial volume reported might not be enough to reduce the operative time of RARC to be comparable with ORC. RARC is less likely to lose blood because of the pneumoperitoneum mostly. Significant less EBL and a lower need for transfusion in the perioperative period was found in this review. Less EBL may be associated with decreased perioperative mortality and morbidity as well42, which is one of the main advantages of RARC that should be valued. In general, the LOS will be longer for the patients with neobladder reconstruction than with ileal conduit. Not regarding to the LOS reported by Sung et al.35, there was significantly lower in LOS in the RARC versus ORC group. We may make a conclusion that RARC was better than ORC regarding to LOS, because neobladder reconstruction rates were similar between these two groups. Given in differences in postoperative patient management preferences and hospital discharge criteria, whether these potential benefits can be proved in future well-designed RCTs is not known. In the meanwhile, further evaluation including differences between urinary diversion type with regard to LOS in ideally required. Some studies tried to present a learning curve and surgeon volume. Wang et al. reported significantly lower operative time comparing the initial 16 with final 17 RARCs.31 Styn et al. drew a learning curve indicating apparently decreased EBL and operative time in RARC.15 Abaza et al. reported 30 min less in their final 20 than their initial 15 RARCs.33 Pruthi et al. reported 60 min less in their final 10 RARCs compared with overall RARCs.29 Sung et al. showed that total operative time decreased dramatically as the number of surgeries performed increased. The operative time for the final five cases was not different between the two groups.35 With further experience with RARC, lower complication rates and better results may be noted. Nevertheless, further study is needed for verification, especially at a multi-institutional level. There was only one RCT.14 It randomized only 41 patients and demonstrated similar LNY, no PSMs, longer operative time, lower EBL, and similar LOS. To assess any impact of study design on the effect estimates, we performed a sensitivity analysis including only one RCT and eight studies with prospectively collected data. Results were similar to the original analysis, except for LNY and LOS. Although a meta-analysis of RCTs would be ideal, the limited number of RCTs prevented us from reaching any definitive conclusions based on sensitivity analysis alone. However, it is difficult to conduct RCTs comparing these surgical techniques; this situation highlights the importance of conducting meta-analyses. Between-study heterogeneity was significant for continuous, but not for dichotomous variables. Included studies adopted different matching criteria, levels of PLND, and measurement of outcomes. These differences might contribute to between-study heterogeneity. Pooling of data using the random-effects model might reduce the effect of heterogeneity but does not abolish it. The present meta-analysis has some limitations that must be considered. The main limitation is that it relied on a minority of the eligible studies. There was just one RCT with inappropriate randomization and 12 observational studies. The sample size of some studies was small that the statistical power to detect a difference in the outcomes was limited. Although two studies43,44 that

559

selected patients with different clinical stages in two groups were excluded, other unmeasurable selection bias might limit the power of this meta-analysis. In addition, because most studies came from high-volume institutions or centers of excellence, the results may be difficult to transfer to community-based practice. Also, costeffectiveness analysis was lack in this meta-analysis. Finally, the follow-up period was generally short, so long-term outcomes of RARC compared with ORC, especially cancer control, urinary continence, and sexual potency, remained to be proved. We applied multiple strategies, strict criteria to include and evaluate the methodological quality of the studies, and subgroup and sensitivity analysis to minimize the heterogeneity. Conclusions This meta-analysis indicates that RARC may be associated with lower overall perioperative complications, more LNY, longer operation time, reduced EBL, a lower need for transfusion, and shorter LOS. The two surgical techniques appear to be equivalent in terms of PSMs. Despite our rigorous methodology, the inherent limitations of included studies prevented us from reaching definitive conclusions. Future large-volume, well-designed RCTs with extensive follow-up are awaited to confirm and update the findings of this analysis. Author contributions J. Huang 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: Huang, K. Li. Acquisition of data: K. Li, Lin, J. Li. Analysis and interpretation of data: Lin, Fan, Zhou. Drafting of the manuscript: K. Li, Fan. Critical revision of the manuscript for important intellectual content: Huang, Lin. Statistical analysis: K. Li, Duan, Yu. Obtaining funding: Huang, Lin. Administrative, technical, or material support: Xu, Bi. Supervision and final approval: Huang. Other (specify): None. Conflict of interest statement All authors certify that all conflicts of interest, including specific financial interests and relationships and affiliations relevant to the subject matter or materials discussed in the manuscript (e.g., employment/affiliation, grants or funding, consultancies, honoraria, stock ownership or options, expert testimony, royalties, or patents filed, received, or pending), are the following: None. Funding source This work was supported by The National Natural Science Foundation of China (81071688, 30972983, 81172431, 91029742), Guangdong province Natural Scientific Foundation (07117366, 6104605), the Yat-sen Scholarship for Young Scientists (to T. Lin), the Clinical Key Project of Public Health Ministry (to J. Huang), the Program for New Century Excellent Talents in University (NCET-10-0852, to T. Lin). Acknowledgments This work was supported by The National Natural Science Foundation of China (81071688, 30972983, 81172431, 91029742),

560

K. Li et al. / Cancer Treatment Reviews 39 (2013) 551–560

Guangdong province Natural Scientific Foundation (07117366, 6104605), the Yat-sen Scholarship for Young Scientists (to T. Lin), the Clinical Key Project of Public Health Ministry (to J. Huang), the Program for New Century Excellent Talents in University (NCET-10-0852, to T. Lin). The funders had no role in the study design, in the collection, analysis and interpretation of data; in the writing of the manuscript; and in the decision to submit the manuscript for publication. No others made contributions to this work. Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.ctrv.2012.11.007. References 1. Stenzl A, Cowan NC, De Santis M, et al. Treatment of muscle-invasive and metastatic bladder cancer: update of the EAU guidelines. Eur Urol 2011;59: 1009–18. 2. McCabe JE, Jibawi A, Javle P. Defining the minimum hospital case-load to achieve optimum outcomes in radical cystectomy. BJU Int 2005;96:806–10. 3. Challacombe BJ, Bochner BH, Dasgupta P, et al. The role of laparoscopic and robotic cystectomy in the management of muscle-invasive bladder cancer with special emphasis on cancer control and complications. Eur Urol 2011;60: 767–75. 4. Davis JW, Castle EP, Pruthi RS, Ornstein DK, Guru KA. Robot-assisted radical cystectomy: an expert panel review of the current status and future direction. Urol Oncol 2010;28:480–6. 5. Elting LS, Pettaway C, Bekele BN, et al. Correlation between annual volume of cystectomy, professional staffing, and outcomes: a statewide, population-based study. Cancer 2005;104:975–84. 6. Richards KA, Hemal AK, Kader AK, Pettus JA. Robot assisted laparoscopic pelvic lymphadenectomy at the time of radical cystectomy rivals that of open surgery: single institution report. Urology 2010;76:1400–4. 7. Menon M, Hemal AK, Tewari A, et al. Nerve-sparing robot-assisted radical cystoprostatectomy and urinary diversion. BJU Int 2003;92:232–6. 8. Pruthi RS, Nielsen ME, Nix J, Smith A, Schultz H, Wallen EM. Robotic radical cystectomy for bladder cancer: surgical and pathological outcomes in 100 consecutive cases. J Urol 2010;183:510–4. 9. Kauffman EC, Ng CK, Lee MM, et al. Critical analysis of complications after robotic-assisted radical cystectomy with identification of preoperative and operative risk factors. BJU Int 2010;105:520–7. 10. Cha EK, Wiklund NP, Scherr DS. Recent advances in robot-assisted radical cystectomy. Curr Opin Urol 2011;21:65–70. 11. Murphy DG, Challacombe BJ, Elhage O, et al. Robotic-assisted laparoscopic radical cystectomy with extracorporeal urinary diversion: initial experience. Eur Urol 2008;54:570–80. 12. Kang SG, Kang SH, Lee YG, et al. Robot-assisted radical cystectomy and pelvic lymph node dissection: a multi-institutional study from Korea. J Endourol 2010;24:1435–40. 13. Mottrie A, Carpentier P, Schatteman P, et al. Robot-assisted laparoscopic radical cystectomy: initial experience on 27 consecutive patients. J Robot Surg 2007;1:197–201. 14. Nix J, Smith A, Kurpad R, Nielsen ME, Wallen EM, Pruthi RS. Prospective randomized controlled trial of robotic versus open radical cystectomy for bladder cancer: perioperative and pathologic results. Eur Urol 2010;57:196–201. 15. Styn NR, Montgomery JS, Wood DP, et al. Matched comparison of roboticassisted and open radical cystectomy. Urology 2012;79:1303–8. 16. Ng CK, Kauffman EC, Lee MM, et al. A Comparison of postoperative complications in open versus robotic cystectomy. Eur Urol 2010;57:274–82. 17. Moher D, Liberati A, Tetzlaff J, Altman DGThe PRISMA Group. Preferred reporting items for systematic reviews and meta-analyses: the PRISMA statement. PLoS Med 2009;6:e1000097. 18. Stroup DF, Berlin JA, Morton SC, et al. Meta-analysis of observational studies in epidemiology: a proposal for reporting. Meta-analysis Of Observational Studies in Epidemiology (MOOSE) group. JAMA 2000;283:2008–12.

19. Huang J, Lin T, Liu H, et al. Laparoscopic radical cystectomy with orthotopic ileal neobladder for bladder cancer: oncologic results of 171 cases with a median 3-year follow-up. Eur Urol 2010;58:442–9. 20. Dindo D, Demartines N, Clavien PA. Classification of surgical complications: a new proposal with evaluation in a cohort of 6336 patients and results of a survey. Ann Surg 2004;240:205–13. 21. Phillips B, Ball C, Sackett D, et al. Levels of evidence and grades of recommendation. Oxford Centre for Evidence-based Medicine Web site. . [accessed 26.04.12. 22. Higgins JPT, Green S. Cochrane handbook for systematic reviews of interventions Version 5.1.0. The Cochrane Collaboration; 2011 [updated March 2011]. 23. Stang A. Critical evaluation of the Newcastle-Ottawa scale for the assessment of the quality of nonrandomized studies in meta-analyses. Eur J Epidemiol 2010;25:603–5. 24. Review Manager (RevMan) [Computer program]. Version 5.1. Copenhagen: The Nordic Cochrane Centre. The Cochrane Collaboration 2011. 25. Hozo SP, Djulbegovic B, Hozo I. Estimating the mean and variance from the median, range, and the size of a sample. BMC Med Res Methodol 2005;5:13. 26. Dorin RP, Daneshmand S, Eisenberg MS, et al. Lymph node dissection technique is more important than lymph node count in identifying nodal metastases in radical cystectomy patients: a comparative mapping study. Eur Urol 2011;60:946–52. 27. Egger M, Davey Smith G, Schneider M, et al. Bias in meta-analysis detected by a simple, graphical test. BMJ 1997;315:629–34. 28. Galich A, Sterrett S, Nazemi T, Pohlman G, Smith L, Balaji KC. Comparative analysis of early perioperative outcomes following radical cystectomy by either the robotic or open method. JSLS 2006;10:145–50. 29. Pruthi RS, Wallen EM. Robotic assisted laparoscopic radical cystoprostatectomy: operative and pathological outcomes. J Urol 2007;178:814–8. 30. Sterrett S, Mammen T, Nazemi T, et al. Major urological oncological surgeries can be performed using minimally invasive robotic or laparoscopic methods with similar early perioperative outcomes compared to conventional open methods. World J Urol 2007;25:193–8. 31. Wang GJ, Barocas DA, Raman JD, Scherr DS. Robotic vs open radical cystectomy: prospective comparison of perioperative outcomes and pathological measures of early oncological efficacy. BJU Int 2008;101:89–93. 32. Martin AD, Nunez RN, Castle EP. Robot-assisted radical cystectomy versus open radical cystectomy: a complete cost analysis. Urology 2011;77:621–5. 33. Abaza R, Dangle PP, Gong MC, Bahnson RR, Pohar KS. Quality of lymphadenectomy is equivalent with robotic and open cystectomy using an extended template. J Urol 2012;187:1200–4. 34. Gondo T, Yoshioka K, Nakagami Y, et al. Robotic versus open radical cystectomy: prospective comparison of perioperative and pathologic outcomes in Japan. Jpn J Clin Oncol 2012;42:625–31. 35. Sung HH, Ahn JS, Seo SI, et al. A comparison of early complications between open and robot-assisted radical cystectomy. J Endourol 2012;26:670–5. 36. Nepple KG, Strope SA, Grubb RL, 3rd, Kibel AS. Early oncologic outcomes of robotic vs. open radical cystectomy for urothelial cancer. Urol Oncol, http:// dx.doi.org/10.1016/j.urolonc.2011.06.009. 37. Herr H, Lee C, Chang S, Lerner SBladder Cancer Collaborative Group. Standardization of radical cystectomy and pelvic lymph node dissection for bladder cancer: a collaborative group report. J Urol 2004;171:1823–8. 38. Cheng L, Weaver AL, Leibovich BC, et al. Predicting the survival of bladder carcinoma patients treated with radical cystectomy. Cancer 2000;88:2326–32. 39. Hellenthal NJ, Hussain A, Andrews PE, et al. Surgical margin status after robot assisted radical cystectomy: results from the International Robotic Cystectomy Consortium. J Urol 2010;184:87–91. 40. Khan MS, Shah SS, Hemel A, Rimington P, Dasgupta P. Robotic-assisted radical cystectomy. Int J Med Robot 2008;4:197–201. 41. Hayn MH, Hellenthal NJ, Hussain A, et al. Does previous robot-assisted radical prostatectomy experience affect outcomes at robot-assisted radical cystectomy? Results from the International Robotic Cystectomy Consortium. Urology 2010;76:1111–6. 42. Wu WC, Smith TS, Henderson WG, et al. Operative blood loss, blood transfusion, and 30-day mortality in older patients after major noncardiac surgery. Ann Surg 2010;252:11–7. 43. Rhee JJ, Lebeau S, Smolkin M, Theodorescu D. Radical cystectomy with ileal conduit diversion: early prospective evaluation of the impact of robotic assistance. BJU Int 2006;98:1059–63. 44. Khan MS, Challacombe B, Elhage O, et al. A dual-centre, cohort comparison of open, laparoscopic and robotic-assisted radical cystectomy. Int J Clin Pract 2012;66:656–62.