Perioperative and Oncological Outcomes of Robot-assisted Versus Open Radical Cystectomy: A Propensity Score-matched Multicenter Study from Japan Authors: SHIMPEI YAMASHITA, MOTOHIRO YASUHARA, MASATOSHI HIGUCHI, SOHEI IWAGAMI, TAKAHITO WAKAMIYA, NAGAHIDE MATSUMURA, MASAYA NISHIHATA, YASUO KOHJIMOTO Published: 2026 ABSTRACT Background/AimThis study aimed to compare the perioperative and oncological outcomes of open radical cystectomy (ORC) and robot-assisted radical cystectomy (RARC) in a contemporary propensity score-matched multicenter cohort from Japan and to provide additional real-world evidence.Patients and MethodsWe retrospectively reviewed 438 patients who underwent radical cystectomy for bladder cancer at four Japanese institutions between March 2009 and December 2024. After excluding patients who underwent laparoscopic radical cystectomy, 397 patients were included. Propensity score matching was performed in a 1:1 ratio based on age, sex, body mass index, Charlson comorbidity index, pathological T stage, pathological nodal status, and type of urinary diversion, resulting in 123 matched pairs. Perioperative outcomes, postoperative complications, and recurrence-free survival (RFS) were compared between the ORC and RARC groups.ResultsAfter propensity score matching, baseline characteristics were generally balanced between the groups. The RARC group had a significantly longer operative time than the ORC group (461 min vs. 374 min, p<0.01). However, the RARC group had significantly lower estimated blood loss (240 ml vs. 1238 ml, p<0.01) and lower intraoperative transfusion rates (14% vs. 90%, p<0.01). The RARC group also had a significantly shorter postoperative hospital stay than the ORC group (23 vs. 28 days, p<0.01). Early and late postoperative complication rates were comparable between the groups. No significant difference in RFS was observed between the groups.ConclusionIn this Japanese cohort, RARC was associated with reduced blood loss, lower transfusion rates, and shorter postoperative hospitalization while maintaining postoperative complication rates and oncological outcomes comparable to those of ORC. Introduction Radical cystectomy with pelvic lymph node dissection is widely used for muscle-invasive bladder cancer and in some cases of high-risk non-muscle-invasive bladder cancer (1, 2). However, radical cystectomy is an invasive procedure and is associated with substantial perioperative morbidity and prolonged recovery (3, 4). As a minimally invasive alternative to open radical cystectomy (ORC), there has been a widespread adoption of robot-assisted radical cystectomy (RARC) in the last decade. Perioperative advantages of RARC, including reduced blood loss and shorter postoperative recovery, have been demonstrated in several studies (5-8). In addition, comparable oncological outcomes between ORC and RARC have been suggested in randomized trials and retrospective studies (9-11). However, the generalizability of these findings to routine clinical practice is uncertain because of substantial variation between institutions in patient selection, surgical experience, perioperative management, and urinary diversion strategies. Contemporary multicenter data from Asian populations remain limited. Therefore, this study compared the perioperative and oncological outcomes of ORC and RARC using a propensity score-matched multicenter cohort from Japan. Patients and Methods Study design and population. We retrospectively reviewed consecutive patients (March 2009-December 2024) at one of four Japanese tertiary referral institutions who underwent radical cystectomy for bladder cancer. We excluded patients who underwent laparoscopic radical cystectomy. The remaining patients were divided into ORC and RARC groups according to surgical approach. Clinical and pathological data were collected from institutional databases and medical records. This study was conducted in accordance with the Declaration of Helsinki and was approved by the institutional review boards of the participating institutions (approval number: 1671). Information regarding the study was publicly disclosed, and patient consent for participation was obtained via an opt-out approach in accordance with institutional policy. Variables and outcomes. Baseline variables included age, sex, body mass index, Eastern Cooperative Oncology Group performance status, Charlson comorbidity index, neoadjuvant therapy status, adjuvant therapy status, pathological T stage, pathological nodal status, type of urinary diversion, pelvic lymph node dissection status, and concomitant urethrectomy. Perioperative outcomes included operative time, estimated blood loss, intraoperative transfusion, intraoperative complications, time to first oral intake, time to first solid food intake, time to first ambulation, and length of postoperative hospital stay. We evaluated postoperative complications via the Clavien–Dindo classification system. Early postoperative complications were defined as complications occurring within 30 postoperative days, whereas late postoperative complications were defined as complications occurring between postoperative days 31 and 90. Major complications were defined as those of Clavien–Dindo grade ≥3. Recurrence-free survival (RFS) was defined as the interval between surgery and recurrence or death from any cause, whichever occurred first. Statistical analysis. Propensity score matching was performed to minimize selection bias and balance baseline characteristics between the two groups. One-to-one nearest-neighbor matching without replacement was conducted using a caliper width of 0.2. Matching variables included age, sex, body mass index, Charlson comorbidity index, pathological T stage, pathological nodal status, and type of urinary diversion. Continuous variables were compared using the Mann–Whitney U-test, whereas categorical variables were compared using the chi-square test or Fisher’s exact test, as appropriate. RFS was estimated using the Kaplan–Meier method and compared using the log-rank test. All statistical analyses were performed using JMP Pro version 17 (SAS Institute Inc., Cary, NC, USA). A two-sided p-value <0.05 was considered statistically significant. Results Patient characteristics. The study flow and propensity score matching process are shown in Figure 1. During the aforementioned period, 438 patients underwent radical cystectomy. After exclusion of patients who underwent laparoscopic surgery, 397 patients were included in the final cohort, consisting of 259 ORC and 138 RARC patients. Propensity score matching yielded 123 matched pairs. Flow diagram of patient selection and propensity score matching. ORC: Open radical cystectomy; RARC: robot-assisted radical cystectomy. Baseline characteristics before and after propensity score matching are summarized in Table I. Before matching, several clinicopathological variables differed between the ORC and RARC groups. After matching, baseline characteristics were well balanced between the two groups, although the RARC group showed a significantly higher rate of concomitant urethrectomy (75% vs. 55%, p<0.01). Table IBaseline patient characteristics before and after propensity score matching.Continuous variables are shown in “median (quartile)” form. PSM: Propensity score matching; ORC: open radical cystectomy; RARC: robot-assisted radical cystectomy; BMI: body mass index; ECOG PS: Eastern Cooperative Oncology Group Performance Status; CCI: Charlson Comorbidity Index. Baseline patient characteristics before and after propensity score matching. Continuous variables are shown in “median (quartile)” form. PSM: Propensity score matching; ORC: open radical cystectomy; RARC: robot-assisted radical cystectomy; BMI: body mass index; ECOG PS: Eastern Cooperative Oncology Group Performance Status; CCI: Charlson Comorbidity Index. Perioperative outcomes. Perioperative and postoperative outcomes are summarized in Table II. The median operative time was significantly longer in the RARC group than in the ORC group (461 min vs. 374 min, p<0.01). However, estimated blood loss was significantly lower in the RARC group (240 ml vs. 1238 ml, p<0.01). Similarly, intraoperative transfusion rates were significantly lower in the RARC group than in the ORC group (14% vs. 90%, p<0.01). Intraoperative complications tended to occur more often in the ORC group than in the RARC group, although without statistical significance (10% vs. 3%, p=0.06). Table IIComparison of perioperative and postoperative outcomes between open radical cystectomy (ORC) and cystectomy robot-assisted radical (RARC) after propensity score matching.Major complications were defined as Clavien–Dindo grade ≥3 complications. Early complications were defined as those occurring within 30 postoperative days, and late complications as those occurring between postoperative days 31 and 90. Continuous variables are shown in “median (quartile)” form. ORC: Open radical cystectomy; RARC: robot-assisted radical cystectomy. Comparison of perioperative and postoperative outcomes between open radical cystectomy (ORC) and cystectomy robot-assisted radical (RARC) after propensity score matching. Major complications were defined as Clavien–Dindo grade ≥3 complications. Early complications were defined as those occurring within 30 postoperative days, and late complications as those occurring between postoperative days 31 and 90. Continuous variables are shown in “median (quartile)” form. ORC: Open radical cystectomy; RARC: robot-assisted radical cystectomy. Postoperative recovery was significantly faster in the RARC group, with earlier oral intake, solid food intake, and ambulation (all p<0.01). Postoperative hospital stay was also significantly shorter in the RARC group (23 days vs. 28 days, p<0.01). Early, major, and late postoperative complication rates were comparable between the two groups. Oncological outcomes. The median follow-up durations were 47 months [interquartile (IQR)=14-92 months] and 31 months (IQR=17-55 months) in the ORC and RARC groups, respectively. Disease recurrence was observed in 38 (31%) and 34 patients (28%) in the ORC and RARC groups, respectively. Kaplan–Meier analysis demonstrated no significant difference in RFS between the groups (Figure 2). The 3-year RFS rates were 60% and 65% in the ORC and RARC groups, respectively (log-rank p=0.17). Kaplan–Meier curves for recurrence-free survival after propensity score matching. ORC: Open radical cystectomy; RARC: robot-assisted radical cystectomy. Discussion In the present multicenter propensity score-matched study from Japan, RARC was associated with significantly lower blood loss, lower intraoperative transfusion rates, and shorter postoperative hospitalization compared with ORC. The relatively high intraoperative transfusion rate observed in the ORC group may partly reflect differences in surgical era and perioperative transfusion practices during the earlier study period but estimated blood loss remained substantially lower in the RARC group. Despite these perioperative advantages, postoperative complication rates and recurrence-free survival were comparable between the two surgical approaches. Our findings are generally consistent with previous reports evaluating minimally invasive and open approaches to radical cystectomy. RARC has been shown to be associated with reduced blood loss and transfusion requirements in several randomized trials and meta-analyses, although operative time is typically longer than that of ORC (5-10). In the phase III RAZOR trial, Parekh et al. demonstrated non-inferior progression-free survival after RARC compared with ORC, but they also reported lower blood loss and transfusion rates in the robotic cohort (9). In addition, the CORAL trial, which compared open, robotic, and laparoscopic radical cystectomy, demonstrated comparable perioperative complication profiles and early oncological outcomes among the three approaches (12). Similarly, Bochner et al. reported comparable oncological outcomes and postoperative complication rates between ORC and RARC in a randomized trial (10, 13). More recently, several real-world studies have further supported the perioperative benefits and oncological feasibility of RARC in routine clinical practice. Hanna et al. reported lower estimated blood loss and shorter postoperative hospital stay following RARC without significant differences in postoperative complications or oncological outcomes (11). Similarly, Yamaguchi et al. compared robot-assisted and laparoscopic radical cystectomy in a Japanese multicenter cohort and reported lower blood loss and comparable oncological outcomes following RARC, while also demonstrating a higher lymph node yield (14). Likewise, a prospective multicenter study demonstrated comparable complication rates and oncological outcomes between ORC and RARC, while confirming favorable perioperative outcomes for the robotic approach (15). In addition, a nationwide propensity score-matched analysis from Sweden by Mortezavi et al. showed lower blood loss, lower transfusion rates, shorter hospital stay, and fewer high-grade complications after RARC, while maintaining acceptable oncological outcomes (16). However, reports from Asian populations are relatively limited. Differences in body habitus, pelvic anatomy, perioperative management, and patient backgrounds between Western and Asian populations may influence perioperative and oncological outcomes after radical cystectomy, highlighting the importance of region-specific real-world data (15, 16). A recent Korean multicenter propensity score-matched study similarly demonstrated lower blood loss and shorter hospitalization after RARC while maintaining comparable recurrence-free survival outcomes (17). Nevertheless, contemporary multicenter real-world data from Asian cohorts are still scarce. The present study further extends these findings in several ways. First, it represents a relatively contemporary multicenter cohort from Japan reflecting real-world clinical practice in an Asian population. Given the aforementioned potential differences between Western and Asian populations, region-specific data may provide clinically important insights into the feasibility and outcomes of RARC. Second, postoperative recovery parameters, including time to oral intake, solid food intake, and ambulation, were evaluated in addition to conventional perioperative variables. Third, propensity score matching was performed to minimize imbalance in baseline patient and pathological characteristics between the two surgical approaches. Although the RARC group demonstrated significantly improved perioperative outcomes, postoperative complications were comparable between the two groups. Similar findings have been reported in previous randomized and observational studies (9-11). One possible explanation is that while robotic surgery may reduce surgical invasiveness and blood loss, postoperative morbidity after radical cystectomy is also strongly influenced by patient frailty, comorbidities, bowel manipulation, urinary diversion type, and perioperative management. Regarding oncological outcomes, we observed no significant difference in recurrence-free survival between ORC and RARC. This finding is generally consistent with previous studies that reported oncological equivalence between the two surgical approaches (9, 10, 13). However, interpretation of oncological outcomes should be made cautiously because perioperative systemic therapies, including neoadjuvant and adjuvant treatments, as well as multidisciplinary management strategies for bladder cancer, have substantially evolved during the study period. Taken together, these findings suggest that RARC can achieve acceptable oncological control when performed in appropriately selected patients. Study limitations. First, it was retrospective in nature, and residual selection bias may have remained despite propensity score matching. Second, the study period spanned more than 15 years, so era bias may have affected the results, as ORC was more commonly performed during the earlier study period whereas RARC became increasingly adopted in later years. In addition, perioperative management and systemic therapies for bladder cancer evolved substantially during the study period and may have influenced oncological outcomes. Third, surgeon experience and learning curve effects were not evaluated and may have influenced perioperative outcomes. Fourth, although pelvic lymph node dissection status was analyzed, detailed information regarding the extent of pelvic lymph node dissection was unavailable. Furthermore, detailed pathological information regarding variant histology was not available in the present multicenter database. Previous studies have demonstrated that the presence, subtype, and extent of variant histology may influence oncological outcomes after radical cystectomy (18-20). Therefore, residual confounding related to pathological heterogeneity cannot be completely excluded. Finally, the median follow-up duration was shorter in the RARC group because robotic surgery was introduced later during the study period, potentially limiting long-term oncological comparisons between the two approaches. Nevertheless, the present study also has several strengths; it represents a contemporary multicenter real-world cohort from Japan and provides additional evidence regarding the perioperative and oncological outcomes of RARC in an Asian population. In addition, we performed propensity score matching to reduce imbalance in baseline clinicopathological characteristics between the two surgical approaches. Furthermore, detailed postoperative recovery parameters, including oral intake, solid food intake, and ambulation, were evaluated in addition to conventional perioperative outcomes. These findings provide additional real-world evidence regarding the perioperative and oncological outcomes of RARC in contemporary Japanese clinical practice. In conclusion, this multicenter propensity score-matched study from Japan demonstrated that RARC was associated with lower blood loss, lower intraoperative transfusion rates, and faster postoperative recovery than ORC, although operative time was longer. Postoperative complication rates and recurrence-free survival were comparable between the two approaches. These findings provide additional real-world evidence supporting the feasibility of RARC in contemporary Japanese clinical practice. Conflicts of Interest The Authors declare no conflicts of interest in relation to this study. Authors’ Contributions Shimpei Yamashita: Conceptualization, Methodology, Formal analysis, Investigation, Resources, Data Curation, Visualization, Writing-Original Draft, Visualization. Motohiro Yasuhara, Masatoshi Higuchi, Sohei Iwagami and Takahito Wakamiya: Resources. Yasuo Kohjimoto: Writing-Review & Editing, Supervision, Project administration. Acknowledgements The Authors acknowledge proofreading and editing by Benjamin Phillis, a Board-Certified Editor in the Life Sciences (BELS), at Wakayama Medical University. Funding No funding was received for this study. Artificial Intelligence (AI) Disclosure The authors used ChatGPT (OpenAI) solely to assist with English-language translation and editing of the manuscript. All scientific content, interpretations, and final wording were reviewed and approved by the authors, who take full responsibility for the manuscript.