Abstract
To explore the therapeutic efficacy of intracorporeal urinary diversion (ICUD) and extracorporeal urinary diversion (ECUD) after robot-assisted radical cystectomy for bladder cancer through systematic review. This study systematically retrieved electronic databases of PubMed, Web of Science, Cochrane Library, CNKI, and Wanfang to include peer-reviewed studies comparing ICUD and ECUD after robot-assisted radical cystectomy for bladder cancer. Relative risk (RR) and mean difference (MD) were used to represent the pooled effect size and estimate its 95% confidence interval (CI). This study included 15 studies with good quality, involving 5,370 patients undergoing robot-assisted radical cystectomy for bladder cancer. Meta-analysis showed that ICUD reduced intraoperative bleeding volume by 64.12 ml (95% confidence interval [CI]: [−100.95, −27.29]), significantly decreased the risk of blood transfusion (RR: 0.40; 95% CI: [0.24, 0.68]) and gastrointestinal complications (odds ratio [OR]: 0.61; 95% CI: [0.47, 0.80]), with shorter postoperative time of exhaust (MD: −9.27; 95% CI: [−18.47, −0.08]) and oral intake (MD: −0.92; 95% CI: [−1.30, −0.54]). However, ICUD had a relatively longer surgical duration (MD: 30.84 min, 95% CI: [5.03, 56.66]). In addition, there was no statistically significant difference concerning the impact of ICUD and ECUD on the length of stay in the hospital (MD: −0.68d; 95% CI: [−1.79, 0.42]), overall complications (30-day: RR: 1.16; 95% CI: [0.93, 1.46]; and 90-day, RR: 0.85; 95% CI: [0.69, 1.04]) and readmission rate (30-day: RR: 0.96; 95% CI: [0.72, 1.27]; and 90-day: RR: 1.15; 95% CI: [0.80, 1.64]). ICUD after robot-assisted laparoscopic radical cystectomy for bladder cancer exhibits obvious positive effects, especially in reducing the risk of blood transfusion and gastrointestinal complications, and shortening postoperative time of exhaust and oral intake. The findings in this meta-analysis should be confirmed by multiple high-quality studies in the future.
Keywords
Introduction
Bladder cancer is one of the most common cancers in the world according to existing epidemiological surveys. It was estimated that there would be over 573,000 new cases of bladder cancer diagnosed in 2020 and about 212,000 deaths as a result of this disease worldwide (Wéber et al., 2024). The treatment of bladder cancer is of great significance in improving the prognosis of patients. Open radical cystectomy combined with pelvic lymphadenectomy is considered to be an important therapeutic option for muscle-invasive bladder cancer or high-risk non-muscle-invasive bladder cancer (Tong et al., 2019). However, open surgery may be accompanied by higher rates of postoperative complications and mortality (Su et al., 2019). Robot-assisted radical cystectomy has gradually become the mainstream option for bladder cancer along with the advancement of medical technology and good prognosis. The proportion of robot-assisted radical cystectomy in the United States increased from 0.6% in 2004 to 18.5% in 2012 (Hu et al., 2016).
The impact of urinary diversion on the morbidity of postoperative complications may be greater than that of cystectomy itself in patients undergoing radical cystectomy (Taber et al., 2024). Considering the complexity of surgery, extracorporeal urinary diversion (ECUD) is the preferred therapeutic option. While intracorporeal urinary diversion (ICUD) exhibits the advantages of small incisions and rapid recovery (Wilson et al., 2015). However, it requires a longer learning curve that may be a critical factor for its clinical application (Dell’Oglio et al., 2021). Despite this, robot-assisted radical cystectomy combined with ICUD, as an alternative to ECUD, is increasingly being widely used clinically. There have been reports on perioperative outcomes of ICUD and ECUD after robot-assisted radical cystectomy (Feng et al., 2020), yet with the absence of the latest large-scale studies (Dalimov et al., 2022; Morizane et al., 2024). Therefore, we conducted a systematic evaluation and meta-analysis by involving the latest reports to compare the efficacy of ICUD and ECUD in patients undergoing robot-assisted radical cystectomy to offer reliable evidence-based data for clinical practice.
Materials and Method
Retrieval Strategy
According to the PRISMA2020 statement (Page et al., 2021), a systematic retrieval was conducted on articles published in 5 relevant electronic databases in Chinese and English, including PubMed, Web of Science, Cochrane Library, CNKI, and Wanfang, from database inception to April 10, 2024. When searching English databases, the following retrieval strategy was applied: “bladder cancer,”“radical cystectomy”; “robotic assisted”; “urinary diversion,”“intracorporeal,” and “extracorporeal.” The retrieval was performed in Boolean operator fuzzy logic. Similar search terms were used for literature retrieval in Chinese databases. Furthermore, eligible articles were screened by reviewing the included literature and relevant references.
Inclusion and Exclusion Criteria
Inclusion criteria: (a) Chinese and English studies published in peer-reviewed journals; (b) studies performed with the inclusion of patients with muscle-invasive bladder cancer or high-risk non-muscle-invasive bladder cancer undergoing robot-assisted radical cystectomy; (c) studies with the establishment of an observation group applying ICUD; (d) studies with the establishment of a control group using ECUD; and (e) randomized controlled trials, cohort studies, or case-control studies.
Exclusion criteria: (a) studies performed with the inclusion of patients with other serious urinary system diseases or comorbidities, such as renal insufficiency, hematuria, and urinary system infections; (b) studies performed with the inclusion of patients who underwent cystectomy previously; and (c) studies with the lack of data for analysis.
Literature Screening and Data Extraction
Two researchers independently conducted literature screening. Disagreements between reviewers would be resolved by additional screening of the reference by a third researcher. Data extraction was completed using the standard data extraction table developed by the research team. The main data extracted were basic literature information (first author, publication time, and region), study design (type, sample size, study duration, and outcome endpoints), study population (mean age, percentage of male, %), intervention measures (urinary diversion methods), research results (surgical duration, intraoperative bleeding volume, blood transfusion rate, length of stay in the hospital, incidence of complications, and readmission rate).
Literature Quality Evaluation
We used the Newcastle-Ottawa Scale (NOS; Stang, 2010) to evaluate the quality of the included literature. Each study was scored in three categories–selection, comparability, and outcome. The maximum score of NOS was 9 points, and the study with a score of ≥7 points was considered high quality.
Statistical Analysis
Reman5.3 software was used for data analysis. The relative risk (RR) and mean difference (MD) were used to express the effect size of counting data and measurement data, respectively, and the 95% confidence interval (CI) was utilized to estimate the interval range of the effect size. Heterogeneity testing was performed using Cochran’s Q-tests and I2 statistics to determine the presence and magnitude of heterogeneity. The estimate from a random-effects meta-analysis is more conservative than that from a fixed-effect meta-analysis; hence, we used a conservative model to address the potential differences in effects across studies and populations (DerSimonian & Laird, 1986; Higgins & Altman, 2008). Measurement data with zero events was estimated using the Peto model, and the pooled effect size using the ratio estimation method (Xu et al., 2021). Subgroup analyses of complications and readmission risk were further conducted based on the time of evaluation. A sensitivity analysis was performed to evaluate potential sources of heterogeneity by excluding each study from the pooled estimate. Unless otherwise specified, the inspection level was set at 0.05.
Results
Literature Screening, Basic Characteristics of the Included Studies and Results of Literature Quality Evaluation
Searching in public electronic databases included 1,392 studies during the literature review, as shown in Figure 1. A total of 317 articles were included during the full-text review after excluding 411 duplicates and 664 unrelated studies. Finally, 15 eligible studies were included according to the inclusion and exclusion criteria pre-set in this study (Ahmed et al., 2014; Bertolo et al., 2019; Dalimov et al., 2022; Guru et al., 2010; Hussein et al., 2018; Kang et al., 2012; Kingo et al., 2017; Lenfant et al., 2018; Mistretta et al., 2021; Morizane et al., 2024; Pruthi et al., 2010; Pyun et al., 2016; Shim et al., 2020; Tan et al., 2019; Teoh et al., 2021).

Literature Screening Flow Diagram
Table 1 shows the baseline characteristics of the included studies. The included 15 studies conducted mainly from 2003 to 2021, were published from 2010 to 2024. Eight articles were multicenter studies, and 3 were prospective cohort studies. Furthermore, there were 5,370 patients undergoing robot-assisted radical cystectomy for bladder cancer among the 15 studies, 2,414 of whom received ICUD, and 2,956 used ICUD. The enrolled subjects of the study were older in age, with an average age of 60 years old, and the majority were males. In addition, the evaluation of literature quality revealed that the included studies had good quality (mean NOS: 7.53).
Characteristics of Eligible Studies and Results of Quality Assessment
Note. ICUD, Intracorporeal Urinary Diversion; ECUD, Extracorporeal Urinary Diversion; UD, Urinary Diversion; NOS, Newcastle-Ottawa Scale; BMI, body mass index.
Surgical Duration
Thirteen studies reported surgical duration for ICUD and ECUD, involving a total of 4,130 patients (ICUD: n=2,092, ECUD: n=2,038). The meta-analysis based on a random-effects model (Figure 2) showed that the surgical duration of patients with ICUD was 30.84 min longer than that of ECUD (95% CI: [5.03, 56.66]). The heterogeneity testing results revealed heterogeneity among the included studies (I2 = 99%, p < .00001). Meanwhile, sensitivity analysis revealed no significant reduction in heterogeneity after excluding any one of the included studies, without any change in the result of the pooled estimate.

Meta-Analysis for the Comparison of Surgical Duration (min) Between ICUD and ECUD
Intraoperative Bleeding Volume
Intraoperative bleeding outcomes for ICUD and ECUD were reported in 12 studies, involving a total of 3,662 patients (ICUD: n=1,899, ECUD: n=1,763). The pooled effect size calculated based on the random-effects model (Figure 3) showed that compared with ECUD, ICUD reduced intraoperative bleeding volume by approximately 64.12 ml (95% CI: [−100.95, −27.29]). The heterogeneity testing results revealed heterogeneity among the included studies (I2 = 90%, p<.00001). Sensitivity analysis revealed no significant reduction in heterogeneity after excluding any one of the included studies, without any change in the result of the pooled estimate.

Meta-Analysis for the Comparison of Intraoperative Bleeding Volume (ml) Between ICUD and ECUD
Blood Transfusion Rate
Eight studies were related to blood transfusion rates for ICUD and ECUD, involving 3,602 patients in totally (ICUD: n=1,814, ECUD: n=1,788). The meta-analysis based on the random-effects model (Figure 4) indicated that the transfusion risk of ICUD patients was significantly lower than that of ECUD patients (RR: 0.40; 95% CI: [0.24, 0.68]). Heterogeneity among the included studies was observed in the heterogeneity testing (I2 = 72%, p=.0007). No significant reduction in heterogeneity was found in sensitivity analysis after excluding any one of the included studies, without any change in the result of the pooled estimate.

Meta-Analysis for the Comparison of Blood Transfusion Rate Between ICUD and ECUD
Length of Stay in the Hospital
Data related to the length of stay for ICUD and ECUD in the hospital was documented in 13 studies, with 5,312 patients included in total (ICUD: n=2,389, ECUD: n=2,923). The meta-analysis based on the random-effects model (Figure 5) showed no statistically significant difference in the length of stay between ICUD and ECUD (MD: −0.68; 95% CI: [−1.79, 0.42]). Heterogeneity among the included studies was observed in the heterogeneity testing (I2 = 97%, p<.00001). Sensitivity analysis revealed no significant reduction in heterogeneity after excluding any one of the included studies, without any change in the result of the pooled estimate.

Meta-Analysis for the Comparison of the Length of Stay in the Hospital (d) Between ICUD and ECUD
Time of Exhaust and Oral Intake
This study also compared the effects of ICUD and ECUD on postoperative time of oral intake and exhaust in patients. The meta-analysis based on random-effects model revealed that patients undergoing ICUD showed advanced time of oral intake (MD: −0.92; 95% CI: [−1.30, −0.54]; Figure 6) and exhaust (MD: −9.27; 95% CI: [−18.47, −0.08]; Figure 7). Besides, the heterogeneity testing suggested good homogeneity.

Meta-Analysis for the Comparison of the Time of Oral Intake (d) Between ICUD and ECUD

Meta-Analysis for the Comparison of the Time of Exhaust (h) Between ICUD and ECUD
Complications
Nine and eight studies reported the 30-day and 90-day risk of complications for ICUD and ECUD, respectively. The meta-analysis based on the random-effects model (Figure 8) revealed no statistically significant difference in the 30-day (RR: 1.16; 95% CI: [0.93, 1.46]) and 90-day (RR: 0.85; 95% CI: [0.69, 1.04]) complication risks between ICUD and ECUD. There existed heterogeneity among the included studies according to the heterogeneity testing (30-day complications: I2 = 81%; 90-day complications: I2 = 64%; overall: I2 = 80%). After excluding one study (Ahmed et al., 2014), the heterogeneity of the 90-day complication subgroup decreased to 23%, with a pooled effect size of 0.79 (95% CI: [0.67, 0.92]).

Meta-Analysis for the Comparison of Complications (30-Day and 90-Day) Between ICUD and ECUD
In addition, 7 studies provided data on the gastrointestinal complications of ICUD and ECUD, involving 2,504 patients. Considering the presence of zero event in one study (Pyun et al., 2016), the pooled effect size was estimated using odd ratio [OR] calculated based on the Peto model. Corresponding meta-analysis (Figure 9) revealed that the risk of gastrointestinal complications in ICUD was significantly lower than that in ECUD (OR: 0.61; 95% CI: [0.47, 0.80]). The heterogeneity testing suggested good homogeneity among the included studies (I2 = 0%, p=.81).

Meta-Analysis for the Comparison of Gastrointestinal Complications Between ICUD and ECUD
Readmission Rate
Five and four studies, respectively, reported the 30-day and 90-day readmission risks for ICUD and ECUD. The meta-analysis based on the random-effects model (Figure 10) indicated no statistically significant difference in the comparison of the 30-day (RR: 0.96; 95% CI: [0.72, 1.27]) and 90-day (RR: 1.15; 95% CI: [0.80, 1.64]) readmission risks between ICUD and ECUD. Good homogeneity among the included studies was observed based on the heterogeneity testing (30-day readmission: I2 = 41%; 90-day readmission: I2 = 42%; overall: I2 = 39%).

Meta-Analysis for the Comparison of Readmission Rate (30-Day and 90-Day) Between ICUD and ECUD
Discussion
This study was designed as a systematic review and meta-analysis based on the latest cohort studies comparing ICUD and ECUD after robot-assisted radical cystectomy for bladder cancer, with the purpose of offering critical evidence-based data for clarifying the effect of both strategies. The present meta-analysis included 15 eligible articles involving 5,370 patients undergoing robot-assisted radical cystectomy for bladder cancer. Consequently, ICUD can significantly reduce intraoperative bleeding volume, lower the risk of blood transfusion, and reduce the average postoperative time of oral intake and exhaust, yet with prolonged surgical duration by about half an hour when compared with ECUD. Further sensitivity analysis suggested that ICUD can decrease the overall risk of 90-day complications postoperatively. However, there was no statistically significant difference concerning the impact of ICUD and ECUD on the length of stay in the hospital and readmission risk.
Results in this study were consistent with those reported previously. For instance, Feng et al. (2020), for the first time, compared and analyzed the perioperative outcomes of ICUD and ECUD after robot-assisted radical cystectomy for bladder cancer. Corresponding meta-analysis of 9 studies revealed that ICUD could reduce intraoperative bleeding volume (MD: −90.5, 95% CI: [−131.26, −49.74]) and gastrointestinal complications (RR: 0.65; 95% CI: [0.45, 0.93]), yet without statistically significant association of ICUD with blood transfusion rate and surgical duration. In addition, similar results were found in an analysis of 13 retrospective studies carried out by Cai et al. (2021).
In our study, ICUD was associated with lower bleeding volume and blood transfusion rate. As for possible reasons, compared with open radical cystectomy, robot-assisted radical cystectomy for bladder cancer has the advantage of reducing the bleeding volume and blood transfusion rate (Satkunasivam et al., 2019; Tzelves et al., 2019), which were also observed in this study. Moreover, pneumoperitoneum pressure can create a better surgical field of view and operating condition, and reduce blood flow and vascular dilation intraperitoneally, thus reducing intraoperative bleeding (Kostakopoulos et al., 2022). According to previous studies (Harraz et al., 2024; Zhang et al., 2020), blood transfusion may increase the risk of tumor recurrence and death after radical cystectomy, and reducing blood transfusion may be beneficial for tumor patients. Besides, the good surgical field of view and operating condition can facilitate lesion removal, and alleviate the extent of surgical trauma and tissue damage, thereby accelerating intestinal function recovery and shortening the postoperative time of exhaust (Shim et al., 2020).
Furthermore, ICUD also has a positive effect on decreasing the incidence of postoperative complications, with the risk of gastrointestinal complications reduced to 0.61 (95% CI: [0.47, 0.80]). In a prospective study carried out by Bochner et al. (2015), the most common complications in radical cystectomy for bladder cancer were intestinal complications and infection; undoubtedly, complications related to intestinal incision were one of the biggest issues in Urology, and also an important reason for prolonging the length of stay in the hospital. Ahmed et al. (2014) also demonstrated that prolonged exposure of the peritoneum extracorporeally might increase the risk of intestinal inflammation, which would lead to intestinal complications such as intestinal paralysis or obstruction. In addition, shortening the extracorporeal exposure time of abdominal organs can reduce the intraoperative stress response to benefit patients’ recovery (Knox et al., 2013).
Nevertheless, this study still has the following limitations. First, this meta-analysis was conducted based on observational studies, all of which were cohort studies, without the inclusion of randomized controlled studies, leading to the potential weakening of the causal inference power of this study. Second, studies included in this meta-analysis did not elaborate on the impact of surgeon experience on the results, which may be an important reason for the heterogeneity among the included studies. As a result, this study failed to further evaluate the effect of surgeons’ learning curve on the therapeutic outcome. Besides, the quality of life and other longer-term prognostic outcomes were not compared among studies included in this analysis.
Conclusion
In conclusion, ICUD after robot-assisted radical cystectomy for bladder cancer is safe and feasible, especially in reducing the risk of blood transfusion and gastrointestinal complications, and shortening the postoperative time of exhaust and oral intake. It can be popularized clinically for muscle-invasive bladder cancer or high-risk non-muscle-invasive bladder cancer patients with surgical indications. Considering the existing limitations in this study, further analysis based on multiple prospective studies conducted by experienced surgeons is still required to validate the results of the current study.
Footnotes
Author Contributions
SA conceived of the study. LS, YL, and LR participated in its design and data analysis and statistics. KZ and MZ helped to draft the manuscript. All authors read and approved the final manuscript.
Declaration of Conflicting Interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author(s) received no financial support for the research, authorship, and/or publication of this article.
Ethics Approval and Consent to Participate
An ethics statement is not applicable because this study is based exclusively on published literature.
Availability of Data and Materials
All data generated or analyzed during this study are included in this published article.
