Abstract
Introduction:
This systematic review and meta-analysis were performed to generate evidence on the complication rates between robot-assisted radical cystectomy and intracorporeal ileal conduit (RARC ICIC) and RARC and intracorporeal orthotopic neobladder (RARC ICONB).
Methods:
A systematic search of the PubMed, Scopus, and Web of Science databases was performed, and all the articles from inception up to June 30, 2024, were screened. Studies reporting on perioperative complications as per the Clavien–Dindo classification and comparing RARC ICIC with RARC ICONB were included. This systematic review was performed as per the Preferred Reporting Items for Systematic Reviews and Meta-analyses guidelines.
Results:
A total of 9 studies evaluating 999 patients were included in the final analysis, and all were retrospective analyses of prospectively maintained databases. The 30-day and 90-day complication rates, including the Clavien–Dindo I–II and III–V and the overall complication rate, were similar between RARC ICIC and RARC ICONB. RARC ICONB had a longer operative time (weighted mean difference – 69.62 min) and higher blood loss (weighted mean difference – 50.53 ml). Patients with stage pT4 and N1 were more in the RARC ICIC group as was the rate of positive surgical margin, which suggests an inherent selection bias while offering the procedure.
Conclusion:
This systematic review, which included only retrospective small-sized series, found that the 30-day and 90-day complication rates between RARC ICIC and RARC ICONB are similar; however, these results are marred by apparent selection bias while offering the procedure. Thus, larger, better-quality prospective randomized studies are required to provide high-quality evidence.
INTRODUCTION
Radical cystectomy is the surgical standard of care for patients with muscle-invasive bladder cancer, and robot-assisted radical cystectomy (RARC) is gaining widespread use. Recent randomized controlled trials have established the oncological safety of RARC equivalent to the open radical cystectomy, with few added advantages in the perioperative outcomes.[1] Still, despite the advances, radical cystectomy has fairly high complication rates. Urinary diversion component is the primary driver of complications after RARC, and intracorporeal diversions performed at high volume centers were found to have reduced risk of complications along with lower blood loss and transfusion rates when compared with extracorporeal diversions and have picked up widespread use as well.[2]
The two most commonly performed urinary diversions after radical cystectomy are ileal conduit and intracorporeal neobladder. A variety of factors, besides tumor-related factors, are considered while deciding on the choice of diversion, and each has certain advantages over the other.[3] An ileal conduit is considered to be faster, with fewer complications and has a shorter length of stay whereas a neobladder may preserve the normal voiding function but is associated with technical challenges and has a higher complication rate.[3,4]
Few recent studies have tried to compare the complications associated with ileal conduit and neobladder after the RARC and found conflicting results. The European multi-institutional series found a lower complication rate after RARC-ileal conduit compared with the RARC-neobladder, whereas the Asian RARC consortium reported similar complication rates between the two.[5,6] However, both these studies included both intracorporeal and extracorporeal diversions and did not focus only on the intracorporeal diversions which are gradually becoming the standard of care. Large volume comparative data on complication rates between RARC with intracorporeal ileal conduit (RARC ICIC) and RARC with intracorporeal orthotopic neobladder (RARC ICONB), which are required for proper counseling of the patients and procedure selection, are lacking, and only a handful of studies have compared the perioperative outcomes between RARC ICIC versus RARC ICONB. Thus, we planned to perform this systematic review and meta-analysis to generate evidence on the complication rates between RARC ICIC and RARC ICONB in patients with bladder cancer.
MATERIALS AND METHODS
Protocol and registration
The review was performed in accordance with the Preferred Reporting Items for Systematic Reviews and Meta-analyses (PRISMA) recommendations, and the study protocol was registered with the International Prospective Register of Systematic Reviews (PROSPERO; ID-CRD42024571961).[7]
Search strategy
A systematic search of the PubMed, Scopus, and Web of Science databases was performed, and all the articles from inception up to June 30, 2024, were screened. The search strategy included the following search terms: “Intracorporeal ileal conduit,” “intracorporeal neobladder,” and “robot-assisted radical cystectomy.” No limits were applied. The bibliography of the included studies was handsearched to further identify the potentially relevant articles. The search results were was uploaded into Rayyan software for screening, and duplicates were removed manually.[8]
Study selection and data extraction
Two authors (SY, TKA) separately reviewed the identified records to select the studies, and any discrepancies were resolved by discussion. The studies were initially screened by title and abstract, and full-length articles were retrieved for the shortlisted articles.
The eligibility criteria were defined using the patient, intervention, comparator, outcome, and study type (PICOS) approach. The inclusion criteria were (1) studies focusing on patients (male or female) with bladder cancer (P), (2) undergoing RARC with intracorporeal neobladder (irrespective of the type) (I), (3) undergoing RARC with intracorporeal ileal conduit (irrespective of the type – Bricker or Wallace) (C), and (4) evaluating perioperative outcomes (O; oncological outcomes [margin, T stage, N stage, recurrence], operative time, blood loss, length of hospital stay, overall complication rate [as per Clavien–Dindo classification], major complication rate, and other outcomes in (5) retrospective or prospective manner (S).
Studies that did not explicitly mention whether the diversion was intracorporeal or not, or did not report the complication rate classified as per the Clavien–Dindo classification, or studies that evaluated patients undergoing extracorporeal urinary diversions, laparoscopic or open cystectomy, or studies that included patients undergoing cystectomy for causes other than bladder cancer, or those that evaluated patients undergoing single-port robotic surgery or reported on cadaveric or animal models were excluded. Furthermore, non-original studies, including review articles, editorial comments, meeting abstracts or letters to the editor, and those not in the English language, were also excluded.
The outcomes measured were as follows: primary outcome –30- and 90-day complication rate as per the Clavien–Dindo classification. The secondary outcomes were demographic profile, perioperative outcomes (operative outcomes, blood loss, length of hospital stay), and oncological outcomes (pathologic stage, positive surgical margins, and recurrence rate).
Study appraisal
For assessing the quality of nonrandomized comparative studies, the Newcastle–Ottawa quality assessment scale for cohort studies was used.[9] According to this scale, a study can be awarded the maximum of 9 stars, and the suggested threshold for a good quality study requires ≥3 stars in the selection domain, ≥1 stars in the comparability domain, as well as ≥2 stars in the outcome/exposure domain. A study is of fair quality if it receives 2 stars in the selection domain, ≥1 stars in the comparability domain, as well as ≥ 2 stars in the outcome/exposure domain, and a study is of poor quality if it receives 0 or 1 stars in the selection domain, 0 stars in the compatibility domain, or 0 or 1 stars in the outcome/exposure domain. Quality assessment was performed by two authors, and discrepancies were solved by discussion.
Statistical analysis
The categorical variables were reported as proportion and continuous variables as the weighted mean. Information on missing data was included, and the denominator was reported for each variable. Wherever required, the mean was calculated from the median and range using the method reported by Cai et al.[10] P < 0.10 was used to indicate statistical heterogeneity. A random effects model was used when heterogeneity was significant, and a fixed effects model when it was not. A two-sided P < 0.05 was considered statistically significant. All statistical analysis was performed using Review Manager v5.4 software designed for composing Cochrane Reviews (Cochrane Collaboration, Oxford, UK).
RESULTS
Study selection
The systematic search of PubMed, Scopus, and Web of Science databases yielded 9628 articles, of which 7221 were duplicates and 2407 abstracts were screened for inclusion. After screening the titles and abstracts of 2407 articles, 127 articles were selected for the full-text review. Out of these, 13 articles that provided head-to-head comparison data between intracorporeal orthotopic neobladder and intracorporeal ileal conduit were selected, and the rest that either provided pooled data on robotic cystectomy with intracorporeal diversions as a single group or used any other method to classify the data other than the type of intracorporeal diversions were excluded. On detailed analysis, 4 of these 13 articles were rejected – one did not explicitly specify whether the diversions were extra or intracorporeal,[6] in one, the complications were not detailed and were not presented as per the Clavien–Dindo classification,[11] one focused on late (>90 days) complications and did not classify them as per the Clavien–Dindo classification,[12] and the last focused on health-related quality of life of patients and lacked detailed presentation of complications as per the Clavien–Dindo classification.[13] The rest of the nine articles were selected for the final analysis.[14,15,16,17,18,19,20,21,22] Figure 1 shows the PRISMA flow diagram of this study.
Figure 1.
Preferred Reporting Items for Systematic Reviews and Meta-analyses 2020 flow diagram. RARC ICIC = Robot-assisted radical cystectomy and intracorporeal ileal conduit, RARC ICONB = Robot-assisted radical cystectomy and intra-corporeal orthotopic neobladdder, CD classification = Clavien–Dindo classification
Study characteristics
A total of nine studies comparing the RARC ICIC versus RARC ICONB were included in the final analysis; all were retrospective studies that analyzed data extracted from prospectively maintained databases. Two studies, Johnson et al.[14] and Collins et al.,[15] were reported from the same institute and had some overlap in the time frame and could have resulted in duplication of data to some extent. Similarly, articles by Abreu et al.[16] and Desai et al.,[17] were also reported from the same center by the same team over an overlapping time frame and could have led to the duplication of a part of the data. Thus, a separate analysis was planned after excluding the overlapping studies, as detailed below. The detailed risk of bias assessment of the included studies is presented in Supplementary Table 1.
Supplementary Table 1.
Risk assessment of comparative studies by Newcastle–Ottawa scale
| Study | Selection domain | Comparability domain | Outcome domain | Quality score | ||||
|---|---|---|---|---|---|---|---|---|
| Jonsson, 2011 [14] | 4 | 1 | 3 | 8/9; good | ||||
| Collins, 2013 [15] | 1 | 1 | 2 | 4/9; poor | ||||
| Abreu, 2014 [16] | 1 | 1 | 1 | 3/9; poor | ||||
| Desai, 2014 [17] | 1 | 1 | 2 | 4/9; poor | ||||
| Sim, 2015 [18] | 4 | 1 | 3 | 8/9; good | ||||
| Porreca, 2018 [19] | 4 | 1 | 3 | 8/9; good | ||||
| Noh, 2022 [20] | 4 | 1 | 3 | 8/9; good | ||||
| Rich, 2023 [21] | 2 | 1 | 2 | 5/9; fair | ||||
| Cochetti, 2023 [22] | 4 | 1 | 3 | 8/9; good |
The risk of bias was calculated based on the Newcastle–Ottawa scale. According to this scale, study can be awarded maximum of 9 stars. A study is of good quality study requires ≥3 stars in the selection domain, ≥1 stars in the comparability domain, as well as ≥2 stars in the outcome/exposure domain. A study is of fair quality if it receives 2 stars in the selection domain, ≥1 stars in the comparability domain, as well as ≥2 stars in the outcome/exposure domain and a study is of poor quality if it receives 0 or 1 stars in the selection domain, 0 stars in the compatibility domain or 0 or 1 stars in the outcome/exposure domain
Demographic profile
In the nine selected studies, a total of 999 patients underwent RARC; of them, 471 had ICIC and 528 had ICONB. The mean age of patients in the RARC ICIC group was 71.7 years compared with 60.1 years in the RARC ICONB arm (weighted mean difference – 10.53 years; 95% confidence interval [CI] 7.81–13.25, P < 0.00001). About 81.6% of the patients in the whole cohort were males, 77.4% (334/432) in the RARC ICIC arm versus 85.6% (370/432) in the RARC ICONB arm (P = 0.17). The mean BMI in the RARC ICIC arm was 25.92 kg/m2 compared with 26.17 kg/m2 in the RARC ICONB arm (P = 0.98). 20.1% (85/422) of the patients in the RARC ICIC arm received neoadjuvant chemotherapy compared with 22.2% (88/396) in the RARC ICONB arm (P = 0.61). The demographic profile is presented in Table 1.
Table 1.
Characteristics of studies included in the study
| First author, year | Type of study | Number of patients | Mean age of patients (years) | Percentage of male patients | Mean BMI (kg/m2) | Operative time (min) | Blood loss (mL) | Length of stay (days) | Risk of bias | |||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Jonsson, 2011 [14] | Retrospective | |||||||||||||||||
| ICIC | 9 | 69.00±9.83 | 55.5 | 23.67±2.11 | 449.03±96.58 | 951.93±2067.57 | 34.23±56.28 | 8/9; good | ||||||||||
| ICONB | 36 | 56.67±9.40 | 91.6 | 25.67±3.92 | 504.8±114.5 | 783.46±566.97 | 17.54±26.44 | |||||||||||
| Collins, 2013 [15] | Retrospective | |||||||||||||||||
| ICIC | 43 | 69.9±6.7 | 74.4 | 24.8±3.1 | 324.07±96.86 | 395.25±647.50 | 24.09±54.54 | 4/9; poor | ||||||||||
| ICONB | 70 | 59.8±9.0 | 88.5 | 26.1±3.4 | 449.74±114.94 | 624.30±480.0 | 15.26±18.53 | |||||||||||
| Abreu, 2014 [16] | Retrospective | |||||||||||||||||
| ICIC | 57 | N/A | 75.4 | N/A | 414.60±75.16 | 276.78±140.01 | 10.71±13.87 | 3/9; poor | ||||||||||
| ICONB | 46 | N/A | 89.1 | N/A | 486.20±122.16 | 314.43±315.85 | 11.07±11.78 | |||||||||||
| Desai, 2014 [17] | Retrospective | |||||||||||||||||
| ICIC | 19 | 71.67±13.58 | 84.2 | 28.65±5.10 | 413.99±99.77 | 298.04±181.20 | 16.28±25.15 | 4/9; poor | ||||||||||
| ICONB | 18 | 58.67±12.55 | 72.2 | 28.58±5.83 | 472.23±162.29 | 388.45±558.40 | 14.55±8.49 | |||||||||||
| Sim, 2015 [18] | Retrospective | |||||||||||||||||
| ICIC | 28 | 75.37±7.21 | 89.2 | 26.93±3.35 | 347.62±83.87 | 404.97±268.29 | 18.94±9.90 | 8/9; good | ||||||||||
| ICONB | 73 | 61.03±8.84 | 78.0 | 25.86±3.41 | 460.44±92.66 | 353.62±161.09 | 22.81±12.36 | |||||||||||
| Porreca, 2018 [19] | Retrospective | |||||||||||||||||
| ICIC | 11 | N/A | N/A | N/A | 368±35.0 | 384±54.2 | 7.36±2.4 | 8/9; good | ||||||||||
| ICONB | 13 | N/A | N/A | N/A | 412±19.0 | 425±61.3 | 8.3±2.0 | |||||||||||
| Noh, 2022 [20] | Retrospective | |||||||||||||||||
| ICIC | 85 | 70.51±8.25 | 85.8 | 24.90±1.06 | 293.79±196.67 | N/A | N/A | 8/9; good | ||||||||||
| ICONB | 118 | 60.63±10.51 | 83.0 | 24.47±1.03 | 362.59±82.62 | N/A | N/A | |||||||||||
| Rich, 2024 [21] | Retrospective | |||||||||||||||||
| ICIC | 190 | 71.18±2.54 | 73.6 | 26.61±1.41 | 273.82±11.75 | 270.54±68.64 | 5.00±0.39 | 5/9; fair | ||||||||||
| ICONB | 71 | 63.95±1.79 | 92.9 | 26.38±1.44 | 340.67±17.08 | 343.10±86.20 | 4.00±0.46 | |||||||||||
| Cochetti, 2023 [22] | Retrospective | |||||||||||||||||
| ICIC | 29 | N/A | N/A | N/A | 267.2±50.7 | 380.3±241.3 | N/A | 8/9; good | ||||||||||
| ICONB | 83 | N/A | N/A | N/A | 306.6±67.3 | 406.1±103.5 | N/A |
The risk of bias was calculated based on the Newcastle–Ottawa scale. According to this scale, study can be awarded maximum of 9 stars. A study is of good quality study requires ≥3 stars in the selection domain, ≥1 stars in the comparability domain, as well as ≥2 stars in the outcome/exposure domain. A study is of fair quality if it receives 2 stars in the selection domain, ≥1 stars in the comparability domain, as well as ≥2 stars in the outcome/exposure domain and a study is of poor quality if it receives 0 or 1 stars in the selection domain, 0 stars in the compatibility domain or 0 or 1 stars in the outcome/exposure domain. BMI=Body mass index, ICIC=Intra-corporeal ileal conduit, ICONB=Intra-corporeal orthotopic neobladdder, N/A=Not available
Perioperative outcomes
The mean operative time in the RARC ICIC arm was 350.23 min as compared with 421.6 min in the RARC ICONB arm (weighted mean difference: −69.62 min; 95% CI − 86.48–−52.76, P < 0.00001) [Figure 2a]. The mean blood loss in the RARC ICIC arm was 420.2 ml compared with 454.8 ml in the RARC ICONB arm (weighted mean difference: −50.53 ml, 95% CI − 81.76–−19.30, P = 0.002) [Figure 2b]. The mean length of stay in the RARC ICIC arm was 16.6 days compared with 13.3 days in the RARC ICONB arm (weighted mean difference: −0.08, 95% CI − 1.71–1.55, P = 0.93) [Figure 2c].
Figure 2.
Forest plots of (a) operative time, (b) blood loss, (c) length of hospital stay
Oncological outcomes
The mean LN yield in the RARC ICIC arm was 30.03 versus 28.61 in the RARC ICONB arm (P = 0.99). The recurrence rate in the RARC ICIC arm was 25.2% (25/99) versus 21.8% (43/197) in the RARC ICONB arm (P = 0.52) [Supplementary Figure 1 (1.4MB, tif) ]. The rate of positive surgical margins was higher in the RARC ICIC arm at 5.4% (23/422) versus the RARC ICONB arm at 2.7% (11/396) (P = 0.05) [Supplementary Figure 1 (1.4MB, tif) ].
Percentage of patients with pT1 (15% (58/385) versus 14.5% (58/399)) (P = 0.69), pT2 (21% (81/385) versus 23.5% (94/399)) (P = 0.99), and pT3 (15.8% (61/385) versus 20% (80/399)) (P = 0.89) were similar between the RARC ICIC and RARC ICONB arms, whereas the percentage of patients with pT4 (15.0% (58/385) versus 4.2% (17/399) (P < 0.00001) and N1 (24.6% (109/442) versus 17% (76/445) (P = 0.003) disease was higher in the RARC ICIC arm as compared with RARC ICONB arm [Supplementary Figure 1 (1.4MB, tif) ]. The pathological and oncological outcomes are presented in Table 3.
Table 3.
Analysis of pathological and oncological outcomes between intracorporeal ileal conduit and intracorporeal orthotopic neobladdder groups
| First author, year | Percentage of pT1 | Percentage of pT2 | Percentage of pT3 | Percentage of pT4 | Percentage of N+ | Lymph node yield (mean±mea | Percentage of recurrence | Percentage of positive surgical margins | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Jonsson, 2011 [14] | ICIC - 22.2 | ICIC - 11.1 | ICIC - 0 | ICIC - 11.1 | ICIC - 33.3 | ICIC - 26.671n,.70 | ICIC - 33.3 | ICIC - N/A | ||||||||
| ICONB - 16.6 | ICONB - 19.4 | ICONB - 13.8 | ICONB - 2.7 | ICONB - 16.6 | ICONB - 23.031n,.70 | ICONB - 19.4 | ICONB - N/A | |||||||||
| Collins, 2013 [15] | ICIC - 9.3 | ICIC - 23.2 | ICIC - 13.9 | ICIC - 25.5 | ICIC - 27.9 | ICIC - 20.9i11.6 | ICIC - 16.2 | ICIC - 11.6 | ||||||||
| ICONB - 10 | ICONB - 28.5 | ICONB - 11.4 | ICONB - 2.8 | ICONB - 14.2 | ICONB - 20.6B11. | ICONB - 18.5 | ICONB - 1.4 | |||||||||
| Abreu, 2014 [16] | ICIC - N/A | ICIC - N/A | ICIC - N/A | ICIC - N/A | ICIC - 29.8 | ICIC - 39.33, 2014 | ICIC - N/A | ICIC - 7.0 | ||||||||
| ICONB - N/A | ICONB - N/A | ICONB - N/A | ICONB - N/A | ICONB - 19.5 | ICONB - 38.30, 2014 | ICONB - N/A | ICONB - 0 | |||||||||
| Desai, 2014 [17] | ICIC - 10.5 | ICIC - 31.5 | ICIC - 36.8 | ICIC - 10.5 | ICIC - 42.1 | ICIC - 44.00, 2014 | ICIC - 31.5 | ICIC - 10.5 | ||||||||
| ICONB - 11.1 | ICONB - 44.4 | ICONB - 11. | ICONB - 5.5 | ICONB - 22.2 | ICONB - 33.00, 2014 | ICONB - 33.3 | ICONB - 0 | |||||||||
| Sim, 2015 [18] | ICIC - 0 | ICIC - 32.1 | ICIC - 28.5 | ICIC - 25.0 | ICIC - 21.4 | ICIC - 17.6050151 | ICIC - 32.1 | ICIC - 14.2 | ||||||||
| ICONB - 8.2 | ICONB - 32.8 | ICONB - 31.5 | ICONB - 8.2 | ICONB - 23.2 | ICONB - 23.5450151 | ICONB - 23.2 | ICONB - 6.8 | |||||||||
| Porreca, 2018 [19] | ICIC - 27.2 | ICIC - 54.5 | ICIC - 0 | ICIC - 0 | ICIC - 18.1 | ICIC - 31.718a, | ICIC - N/A | ICIC - N/A | ||||||||
| ICONB - 23.0 | ICONB - 46.1 | ICONB - 15.3 | ICONB - 0 | ICONB - 0 | ICONB - 33.2B8a, | ICONB - N/A | ICONB - N/A | |||||||||
| Noh, 2022 [20] | ICIC-17.6 | ICIC - 17.6 | ICIC - 29.4 | ICIC - 14.1 | ICIC - 22.3 | ICIC - N/A | ICIC - N/A | ICIC - 1.1 | ||||||||
| ICONB - 16.1 | ICONB - 16.1 | ICONB - 25.4 | ICONB - 5.0 | ICONB - 20.3 | ICONB - N/A | ICONB - N/A | ICONB - 2.5 | |||||||||
| Rich, 2024 [21] | ICIC - 16.8 | ICIC - 17.8 | ICIC - 21.0 | ICIC - 13.1 | ICIC - 22.1 | ICIC - N/A | ICIC - N/A | ICIC - 3.6 | ||||||||
| ICONB - 21.1 | ICONB - 14.0 | ICONB - 14.0 | ICONB - 1.4 | ICONB - 8.4 | ICONB - N/A | ICONB - N/A | ICONB - 2.8 | |||||||||
| Cochetti, 2023 [22] | ICIC - N/A | ICIC - N/A | ICIC - N/A | ICIC - N/A | ICIC - N/A | ICIC - N/A | ICIC - N/A | ICIC - N/A | ||||||||
| ICONB - N/A | ICONB - N/A | ICONB - N/A | ICONB - N/A | ICONB - N/A | ICONB - N/A | ICONB - N/A | ICONB - N/A |
ICIC=Intracorporeal ileal conduit, ICONB=Intracorporeal orthotopic neo-bladdder, N/A=Not available, SD=Standard deviation
Complications
Thirty-day complications: the 30-day Clavien I–II complication rate in the RARC ICIC arm was 29.1% (82/281) versus 28.8% (132/457) in the RARC ICONB arm (P = 0.59) [Figure 3a]. The 30-day Clavien III–V complication rate in the RARC ICIC arm was 20.6% (58/281) versus 15.5% (71/457) in the RARC ICONB arm (P = 0.09) [Figure 3b]. The 30-day overall complication rate in the RARC ICIC arm was 49.4% (139/281) versus 44.4% (203/457) in the RARC ICONB arm (P = 0.76) [Figure 3c]. The complication rates of all the studies are reported in Table 2.
Figure 3.
Forest plots of (a) 30-day Clavien–Dindo I–II complications, (b) 30-day Clavien–Dindo III–V complications, (c) 30-day overall complications
Table 2.
Analysis of complication rate between intracorporeal ileal conduit and intracorporeal orthotopic neobladdder groups
| First author, year | Incidence of 30-day Clavien I–II complications (%) | Incidence of 30-day Clavien III–V complications (%) | Total 30-day complication rate (%) | Incidence of 90-day Clavien I–II complications (%) | Incidence of 90-day Clavien III–V complications (%) | Total 90-day complication rate (%) | ||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Jonsson, 2011 [14] | ICIC - 11.1 | ICIC - 33.3 | ICIC - 44.4 | ICIC - 11.1 | ICIC - 22.2 | ICIC - 33.3 | ||||||
| ICONB - 19.4 | ICONB - 19.4 | ICONB - 38.8 | ICONB - 16.6 | ICONB - 16.6 | ICONB - 33.3 | |||||||
| Collins, 2013 [15] | ICIC - 32.5 | ICIC - 53.4 | ICIC - 86.0 | ICIC - 0 | ICIC - 23.2 | ICIC - 23.2 | ||||||
| ICONB - 12.8 | ICONB - 31.4 | ICONB - 44.2 | ICONB - 12.8 | ICONB - 21.4 | ICONB - 34.2 | |||||||
| Abreu, 2014 [16] | ICIC - 42.1 | ICIC - 22.8 | ICIC - 64.9 | ICIC - N/A | ICIC - N/A | ICIC - N/A | ||||||
| ICONB - 41.3 | ICONB - 10.8 | ICONB - 52.1 | ICONB - N/A | ICONB - N/A | ICONB - N/A | |||||||
| Desai, 2014 [17] | ICIC - 42.1 | ICIC - 26.3 | ICIC - 68.4 | ICIC - 47.3 | ICIC - 31.5 | ICIC - 78.9 | ||||||
| ICONB - 66.6 | ICONB - 5.5 | ICONB - 72.2 | ICONB - 66.6 | ICONB - 16.6 | ICONB - 83.3 | |||||||
| Sim, 2015 [18] | ICIC - 21.4 | ICIC - 14.2 | ICIC - 35.7 | ICIC - 0 | ICIC - 7.1 | ICIC - 7.1 | ||||||
| ICONB - 28.7 | ICONB - 27.3 | ICONB - 56.1 | ICONB - 1.3 | ICONB - 15.0 | ICONB - 16.4 | |||||||
| Porreca, 2018 [19] | ICIC - 9 | ICIC - 0 | ICIC - 9.0 | ICIC - 0 | ICIC - 0 | ICIC - 0 | ||||||
| ICONB - 15.3 | ICONB - 7.6 | ICONB - 23.0 | ICONB - 3.0 | ICONB - 7.6 | ICONB - 38.4 | |||||||
| Noh, 2022 [20] | ICIC - 24.7 | ICIC - 8.2 | ICIC - 32.9 | ICIC - 12.9 | ICIC - 15.2 | ICIC - 28.2 | ||||||
| ICONB - 33 | ICONB - 5.0 | ICONB - 38.1 | ICONB - 15.2 | ICONB - 10.1 | ICONB - 25.4 | |||||||
| Rich, 2024 [21] | ICIC - N/A | ICIC - N/A | ICIC - N/A | ICIC - 40 | ICIC - 11.5 | ICIC - 51.5 | ||||||
| ICONB - N/A | ICONB - N/A | ICONB - N/A | ICONB - 29.5 | ICONB - 16.9 | ICONB - 46.4 | |||||||
| Cochetti, 2023 [22] | ICIC - 24.1 | ICIC - 10.3 | ICIC - 34.4 | ICIC - 31 | ICIC - 3.4 | ICIC - 34.4 | ||||||
| ICONB - 27.7 | ICONB - 10.8 | ICONB - 38.5 | ICONB - 43.3 | ICONB - 6 | ICONB - 49.3 |
ICIC=Intracorporeal ileal conduit, ICONB=Intracorporeal orthotopic neobladdder
Ninety-day complications: the 90-day Clavien I–II complication rate in the RARC ICIC arm was 25.6% (106/414) versus 22.1% (107/482) in the RARC ICONB arm (P = 0.27) [Figure 4a]. The 90-day Clavien III–V complication rate in the RARC ICIC arm was 13.5% (56/414) versus 13.4% (65/482) in the RARC ICONB arm (P = 0.95) [Figure 4b]. The 90-day overall complication rate in the RARC ICIC arm was 39.1% (162/414) versus 35.6% (172/482) in the RARC ICONB arm (P = 0.35) [Figure 4c].
Figure 4.
Forest plots of (a) 90-day Clavien–Dindo I–II complications, (b) 90-day Clavien–Dindo III–V complications, (c) 90-day overall complications
Two pairs of studies had overlapping data, as described above. To overcome the limitation of overlapping data, we selected the study with a larger sample size of the two pairs of studies with overlapping data, omitted the data from the study with a smaller size, and reanalyzed the overall complication rates. Analysis of 30-day overall complications was performed after excluding data from studies by Jonsson et al.[14] and Desai et al.,[17] and the 30-day overall complication rate was still found to be similar between the RARC ICIC and RARC ICONB groups (48.2% (122/253) versus 43.6% (176/403), P = 0.79). Analysis of the 90-day overall complications was repeated after excluding data from the study by Jonsson et al.[14] only, as the article by Abreu et al.[16] had not reported on 90-day complications, and the 90-day overall complication rate was still found to be similar between RARC ICIC and RARC ICONB (39.2% (159/405) versus 35.8% (160/446), P = 0.30).
Only three of the nine included studies presented rates of individual complications. The rate of urinary tract infection was higher after RARC ICONB as compared to RARC ICIC (19.4% (39/201) versus 6.4% (9/139), P = −0.0009), whereas the rates of postoperative ileus (RARC ICIC – 17.9% [25/139] versus RARC ICONB – 17.4% [35/201], P = 0.77) and ureteroenteric stricture (RARC ICIC – 2.1% [3/139] versus RARC ICONB – 4.9% [10/201], P = 0.21) were similar between the two groups. The rates of reoperation for intestinal obstruction (RARC ICIC – 2.3% [1/43] versus RARC ICONB – 2.8% [2/70], P = 0.86) and need of per-cutaneous nephrostomy (RARC ICIC – 23.2% [10/43] versus RARC ICONB – 11.4% [8/70], P = 0.10) were also similar between the two groups, whereas the rate of any complication requiring resurgery (RARC ICIC – 23.2% [10/43] versus RARC ICONB – 8.5% [6/70], P = 0.04) was higher after RARC ICIC [Supplementary Figure 1 (1.4MB, tif) ].
DISCUSSION
There is a lack of large-size studies comparing complications of RARC ICIC and RARC ICONB, which are gradually becoming the standard of care. Such data are required in patient counseling and selecting one procedure over the other. This systematic review and meta-analysis evaluated all published studies in the English language comparing RARC ICIC and RARC ICONB and describing the complications as per the Clavien–Dindo classification system. A total of 9 studies were included, and the patients in the RARC ICONB arm were younger by 10 years, had longer operating time by 69 min, had higher blood loss by 50 ml, and tended to have a lower number of patients with pT4 and N1 disease. The review found that 40%–50% of the patients had some complication or the other, and the 30-day and 90-day complication rates were similar between the RARC ICIC and RARC ICONB arms. Considering the limitations of the review, these findings are hypothesis-generating at best and need confirmation with prospective randomized studies.
Two important studies on the same topic, i.e., studies by Tappero et al. and Wong et al. had to be excluded as Tapero et al. included both extracorporeal and intracorporeal diversions without providing separate data for the both and the study by Wong et al. did not specify whether the diversions were extracorporeal or intra-corporeal.[5,6] However, these two studies still remain relevant as they are multicenter large studies evaluating complications of RARC. Tappero et al. noted a similar overall complication rate of 57% in the IC arm and 60% in the ONB arm; however, the rates of genitourinary (11% vs. 24%, P < 0.001) complications were higher in the ONB arm.[5] Furthermore, the rate of overall urinary diversion-related complications was slightly higher in the ONB arm, although this did not reach the statistical significance (51% vs. 56%, P = 0.08). The type of urinary diversion was also found to be an independent predictor of urinary diversion-related complications on the multivariate analysis. On the other hand, Wong et al. noted a slightly higher incidence of postoperative complications in the ONB arm that did not reach statistical significance (30.8% vs. 42.3%, P = 0.072). However, the incidence of high-grade complications was similar between the two groups (16.8% vs. 18.0%). In the current systematic review, the overall 30-day and 90-day complication rates, as well as, the high-grade complications, were found to be similar between the RARC ICIC and RARC ICONB arms.
Two other studies that were excluded were by Presicce et al. and Abozaid et al.[12,13] Presicce et al. reported on long-term (>90 days) complications after RARC and intracorporeal urinary diversions and reported a similar rate of urinary tract infection (IC – 34% vs. ONB – 38%, P = 0.312), incisional hernia (IC – 16% vs. ONB – 21%, P = 0.640), ureteroileal stricture (IC – 12% vs. ONB – 15%, P = 0.658), and bowel occlusion (IC – 4% vs. ONB – 1%, P = 0.435). However, the rate of stone formation was higher in the ONB arm (IC – 0% vs. ONB – 10%, P = 0.047), whereas the rate of progression to CKD ≥ 3a was higher in the IC arm (IC – 58% vs. ONB – 44%, P = 0.001). On the other hand, Abozaid et al. primarily focused on the health-related quality of life in patients undergoing RARC and intracorporeal diversion and noted a similar 90-day major complication rate between IC and ONB (16.6% vs. 13.5%, P = 0.712). They also reported that the role functioning and fatigue scores of the health-related questionnaire returned to the baseline after 6 months in both the arms, but the physical functioning score did not return to the baseline even after 12 months in the IC arm, whereas it returned to the baseline in the ONB arm at 6 months. Both these studies did not meet the inclusion criteria of our study and thus were excluded from the analysis despite providing valuable comparative data for patients and physicians attempting to choose between RARC ICIC versus RARC ICONB. Other important aspects when counseling patients for IC versus ONB are the incontinence rates after the ONB, which are reported to be around 10% at 12 months after the surgery and the long-term functional outcomes and both of these were not assessed in this review.[23]
As far as the perioperative outcomes are concerned, RARC ICONB tends to have longer operative time than the RARC ICIC and this has been confirmed by multiple studies.[12,13] However, the amount of blood loss and the length of the postoperative hospital stay have been reported as either similar or higher after RARC ICONB in various studies.[12,13] The amount of blood loss is slightly more after RARC ICONB by around 50 ml, which is statistically significant but is not clinically significant as the transfusion rates tend to be similar between the two.[12] Similarly, for the length of postoperative stay, few studies report a longer length of stay after RARC ICONB by 2–5 days, but in the current systematic review, we could not find a difference in the length of the postoperative hospital stay.[12,13]
In the majority of the literature, the patients who were offered RARC ICONB tended to be younger by about 6–10 years than those being offered RARC ICIC and had lower chances of having pT4 and N + disease.[12,13] These findings resonate with those found in the current systematic review and can explain the higher rates of positive surgical margins noted in our review and the poorer disease-free survival noted by Wong et al. in the RARC ICIC arm.[13] Although selecting younger patients and those with lower-stage disease makes clinical sense, the difference in the age and disease status may have affected the complication rates and could have been responsible for a lack of difference in the complication rates between the ICIC and ICONB groups in our review. Properly designed studies, that use either randomization or propensity matching, in the future are required to mitigate this selection bias. Another important shortcoming of this review is the possibility of duplication of a small amount of data, as two pairs of studies were published from the same institute with overlapping time frames. Although we have performed an additional meta-analysis by excluding one of these studies and found no effect on the outcomes, duplication of data can confound the results, further downgrading the quality of the systematic review.
Although a systematic review generates the highest quality of evidence, the quality of evidence generated usually depends on the quality of the studies included. The current systematic review included only retrospective studies with their inherent risks and biases. Furthermore, there appears to be a selection bias in the patients offered ICIC and ICONB based on the age and the stage of the disease, with the percentage of younger patients and those with lower-stage disease higher in the ICONB arm. Furthermore, the long-term data, including the continence rates, functional outcomes, and long-term complications, including the metabolic ones, which form an important aspect of discussion while selecting IC versus ONB, have not been covered by this review. Besides, although the majority of the studies were of good quality, some were of poor quality and there were some concerns about the duplication of the data also. The overall quality of evidence (as per GRADE recommendation) for most of the outcomes reported in this systematic review is very low in view of the lack of randomized trials, apparent selection bias in offering the procedure, and heterogeneity among some of the reported outcomes, and there is a need for properly designed randomized control trials to better assess the complication rates between RARC ICIC and RARC ICONB.
CONCLUSION
This systematic review, which included retrospective small-sized series only, found that the 30-day and 90-day complication rates between RARC ICIC and RARC ICONB are similar. However, these results are marred by apparent selection bias while offering the procedure, and thus, the findings are hypothesis-generating, and larger better quality prospective randomized studies are required to provide high-quality evidence.
Conflicts of interest
There are no conflicts of interest.
Forest plots of (a) age distribution, (b) recurrence rate, (c) positive surgical margin, (d) distribution of pT4, (e) distribution of N1, (f) urinary tract infection rates, (g) need of resurgery, (h) 30-day overall complication rate after excluding studies with overlapping data, (i) 90-day overall complication rate after excluding studies with overlapping data
Funding Statement
Nil.
REFERENCES
- 1.Khetrapal P, Wong JK, Tan WP, Rupasinghe T, Tan WS, Williams SB, et al. Robot-assisted radical cystectomy versus open radical cystectomy: A systematic review and meta-analysis of perioperative, oncological, and quality of life outcomes using randomized controlled trials. Eur Urol. 2023;84:393–405. doi: 10.1016/j.eururo.2023.04.004. [DOI] [PubMed] [Google Scholar]
- 2.Katayama S, Mori K, Pradere B, Mostafaei H, Schuettfort VM, Quhal F, et al. Intracorporeal versus extracorporeal urinary diversion in robot-assisted radical cystectomy: A systematic review and meta-analysis. Int J Clin Oncol. 2021;26:1587–99. doi: 10.1007/s10147-021-01972-2. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Lee RK, Abol-Enein H, Artibani W, Bochner B, Dalbagni G, Daneshmand S, et al. Urinary diversion after radical cystectomy for bladder cancer: Options, patient selection, and outcomes. BJU Int. 2014;113:11–23. doi: 10.1111/bju.12121. [DOI] [PubMed] [Google Scholar]
- 4.Nieuwenhuijzen JA, de Vries RR, Bex A, van der Poel HG, Meinhardt W, Antonini N, et al. Urinary diversions after cystectomy: The association of clinical factors, complications and functional results of four different diversions. Eur Urol. 2008;53:834–42. doi: 10.1016/j.eururo.2007.09.008. [DOI] [PubMed] [Google Scholar]
- 5.Tappero S, Dell’Oglio P, Cerruto MA, Sanchez Salas R, Buisan Rueda O, Simone G, et al. Ileal conduit versus orthotopic neobladder urinary diversion in robot-assisted radical cystectomy: Results from a multi-institutional series. Eur Urol Open Sci. 2023;50:47–56. doi: 10.1016/j.euros.2023.01.009. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Wong CH, Ko IC, Kang SH, Kitamura K, Horie S, Muto S, et al. Long-term outcomes of orthotopic neobladder versus ileal conduit urinary diversion in robot-assisted radical cystectomy (RARC): Multicenter results from the Asian RARC consortium. Ann Surg Oncol. 2024;31:5785–93. doi: 10.1245/s10434-024-15396-5. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Moher D, Liberati A, Tetzlaff J, Altman DG, PRISMA Group Preferred reporting items for systematic reviews and meta-analyses: The PRISMA statement. Ann Intern Med. 2009;151:264–9. doi: 10.7326/0003-4819-151-4-200908180-00135. W64. [DOI] [PubMed] [Google Scholar]
- 8.Johnson N, Phillips M. Rayyan for systematic reviews. J Electron Resour Librariansh. 2018;30:46–8. [Google Scholar]
- 9.Wells GA, Shea B, O’Connell D, Peterson J, Welch V, Losos M, et al. The Newcastle-Ottawa Scale (NOS) for Assessing the Quality of Nonrandomized Studies in Meta-Analyses. The Ottawa Hospital. Available from: https://ohri.ca/programs/clinical_epidemiology/oxford.htm. (accessed on 30th August 2024)
- 10.Cai S, Zhou J, Pan J. Estimating the sample mean and standard deviation from order statistics and sample size in meta-analysis. Stat Methods Med Res. 2021;30:2701–19. doi: 10.1177/09622802211047348. [DOI] [PubMed] [Google Scholar]
- 11.Iqbal U, Houenstein HA, Elsayed AS, Jing Z, James G, Hussein AA, et al. Ileal conduit versus neobladder: A propensity score-matched analysis of the effect on renal function. Int J Urol. 2022;29:158–63. doi: 10.1111/iju.14747. [DOI] [PubMed] [Google Scholar]
- 12.Presicce F, Leonardo C, Tuderti G, Brassetti A, Mastroianni R, Bove A, et al. Late complications of robot-assisted radical cystectomy with totally intracorporeal urinary diversion. World J Urol. 2021;39:1903–9. doi: 10.1007/s00345-020-03378-7. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Abozaid M, Tan WS, Khetrapal P, Baker H, Duncan J, Sridhar A, et al. Recovery of health-related quality of life in patients undergoing robot-assisted radical cystectomy with intracorporeal diversion. BJU Int. 2022;129:72–9. doi: 10.1111/bju.15505. [DOI] [PubMed] [Google Scholar]
- 14.Jonsson MN, Adding LC, Hosseini A, Schumacher MC, Volz D, Nilsson A, et al. Robot-assisted radical cystectomy with intracorporeal urinary diversion in patients with transitional cell carcinoma of the bladder. Eur Urol. 2011;60:1066–73. doi: 10.1016/j.eururo.2011.07.035. [DOI] [PubMed] [Google Scholar]
- 15.Collins JW, Tyritzis S, Nyberg T, Schumacher M, Laurin O, Khazaeli D, et al. Robot-assisted radical cystectomy: Description of an evolved approach to radical cystectomy. Eur Urol. 2013;64:654–63. doi: 10.1016/j.eururo.2013.05.020. [DOI] [PubMed] [Google Scholar]
- 16.Abreu AL, Chopra S, Azhar RA, Berger AK, Miranda G, Cai J, et al. Robotic radical cystectomy and intracorporeal urinary diversion: The USC technique. Indian J Urol. 2014;30:300–6. doi: 10.4103/0970-1591.135673. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Desai MM, de Abreu AL, Goh AC, Fairey A, Berger A, Leslie S, et al. Robotic intracorporeal urinary diversion: Technical details to improve time efficiency. J Endourol. 2014;28:1320–7. doi: 10.1089/end.2014.0284. [DOI] [PubMed] [Google Scholar]
- 18.Sim A, Balbay MD, Todenhöfer T, Aufderklamm S, Halalsheh O, Mischinger J, et al. Robot-assisted radical cystectomy and intracorporeal urinary diversion – Safe and reproducible? Cent European J Urol. 2015;68:18–23. doi: 10.5173/ceju.2015.01.466. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Porreca A, Chessa F, Romagnoli D, Salvaggio A, Cafarelli A, Borghesi M, et al. Robot assisted radical cystectomy with totally intracorporeal urinary diversion: Initial, single-surgeon’s experience after a modified modular training. Minerva Urol Nefrol. 2018;70:193–201. doi: 10.23736/S0393-2249.17.02970-8. [DOI] [PubMed] [Google Scholar]
- 20.Noh TI, Shim JS, Kang SG, Cheon J, Pyun JH, Kang SH. The learning curve for robot-assisted radical cystectomy with total intracorporeal urinary diversion based on radical cystectomy pentafecta. Front Oncol. 2022;12:975444. doi: 10.3389/fonc.2022.975444. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Rich JM, Cumarasamy S, Ranti D, Lavallee E, Attalla K, Sfakianos JP, et al. Contemporary outcomes of patients undergoing robotic-assisted radical cystectomy: A comparative analysis between intracorporeal ileal conduit and neobladder urinary diversions. Asian J Urol. 2023;10:446–52. doi: 10.1016/j.ajur.2023.06.002. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Cochetti G, Paladini A, Del Zingaro M, Ciarletti S, Pastore F, Massa G, et al. Robot-assisted radical cystectomy with intracorporeal reconstruction of urinary diversion by mechanical stapler: Prospective evaluation of early and late complications. Front Surg. 2023;10:1157684. doi: 10.3389/fsurg.2023.1157684. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Nayak AL, Cagiannos I, Lavallée LT, Morash C, Hickling D, Mallick R, et al. Urinary function following radical cystectomy and orthotopic neobladder urinary reconstruction. Can Urol Assoc J. 2018;12:181–6. doi: 10.5489/cuaj.4877. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Forest plots of (a) age distribution, (b) recurrence rate, (c) positive surgical margin, (d) distribution of pT4, (e) distribution of N1, (f) urinary tract infection rates, (g) need of resurgery, (h) 30-day overall complication rate after excluding studies with overlapping data, (i) 90-day overall complication rate after excluding studies with overlapping data