Comparative outcomes of KangDuo and da Vinci robot-assisted ureteral reimplantation: a retrospective cohort study

Ureteral reimplantation represents an essential reconstructive procedure for the management of complex distal ureteral strictures. Although conventional laparoscopic surgery is widely used and provides several advantages over open surgery, including reduced blood loss and shorter hospital stay ( 1 , 2 ), laparoscopic reconstructive procedures remain technically demanding owing to ergonomic constraints, restricted instrument articulation, and the challenges associated with precise intracorporeal free-hand suturing ( 3 , 4 ). The introduction of the da Vinci system has advanced reconstructive urology through stable three-dimensional visualization and wristed instrumentation, but its acquisition and maintenance costs have motivated evaluation of alternative platforms ( 5 , 6 ). In recent years, several novel robotic surgical platforms have been developed, incorporating technological innovations such as open-console architecture and modular system design ( 7 ). Among these platforms, the KangDuo Surgical Robot-01 (KD-SR-01) has shown encouraging clinical safety and efficacy in a range of urological procedures, including partial nephrectomy, pyeloplasty, and radical prostatectomy ( 8 - 10 ). Preliminary single-arm studies have suggested the technical feasibility of KD-SR-01-assisted ureteral reimplantation ( 11 ); however, direct comparative evidence against the established da Vinci platform remains limited. Because wider availability of alternative platforms could reduce institutional dependence on a single system and broaden access to robotic reconstruction, head-to-head comparative data are needed to evaluate their performance in technically demanding reconstructive procedures. We therefore conducted a retrospective cohort study to compare perioperative safety, surgical efficiency, and mid-term reconstructive outcomes between KD-SR-01-assisted and da Vinci-assisted ureteral reimplantation. We present this article in accordance with the STROBE reporting checklist (available at https://tau.amegroups.com/article/view/10.21037/tau-2026-0540/rc ).

Between January 2024 and December 2025, 162 patients underwent robot-assisted ureteral reimplantation in Peking University First Hospital, performed by an experienced surgeon. The choice of robotic platform was nonrandom and was determined solely by each patient’s voluntary preference after standardized counseling regarding both systems. Both platforms were available throughout the study period, and the operating surgeon did not recommend or assign either platform. Patient characteristics, perioperative data, and clinical outcomes were prospectively recorded in the Reconstruction of Urinary Tract: Technology, Epidemiology and Result (RECUTTER) database at preoperative assessment, operation, discharge, and each scheduled follow-up visit, and were verified against the electronic medical record ( 12 ). Eligibility was independently assessed by two investigators from clinical and imaging records. The study protocol is registered at www.chictr.org.cn (ChiCTR2400081949). The study was conducted in accordance with the Declaration of Helsinki and its subsequent amendments. The study was approved by the Ethics Committee of Peking University First Hospital (No. 2024-083-002), and individual consent for this retrospective analysis was waived. The inclusion criteria for this study were as follows: (I) aged 18–75 years; (II) a radiologically confirmed diagnosis of distal ureteral stricture; and (III) availability of comprehensive perioperative clinical records. Exclusion criteria were loss to follow-up (n=3), concomitant surgery (n=4), and bilateral surgical intervention (n=2). The patient selection process and study protocol are summarized in Figure 1 . Flowchart of participant selection. KD-SR-01, KangDuo Surgical Robot-01. All procedures were performed under general anesthesia with patients placed in a 30° Trendelenburg position. The pneumoperitoneum was maintained at a constant pressure of 14 mmHg. Trocar placement varied by system: the KD-SR-01 used a three-arm configuration, while the da Vinci utilized a standard four-arm setup. Additional assistant ports were placed as needed for suction and suture management. A transperitoneal approach was utilized. After incising the posterior peritoneum, the ureter was identified at the level of the iliac vessels and mobilized distally toward the lesion ( Figure 2A ). The distal ureter was then completely mobilized from its pelvic wall attachments until the dilated proximal end of the stricture was fully reached. The stenotic segment and surrounding fibrotic tissue were completely excised until adequate vascularity of the proximal ureteral stump and unobstructed urinary drainage were confirmed. During mobilization, meticulous dissection was performed using cold scissors to strictly preserve the ureteral adventitia and its longitudinal blood supply. The proximal ureteral end was then spatulated for approximately 1.5–2.0 cm. In cases involving extensive scarring, near-infrared fluorescence (NIRF) imaging with indocyanine green (ICG) was employed to guide the precise demarcation and excision of the lesion. Representative intraoperative images of ureteral mobilization, psoas hitch, and Boari flap reconstruction. (A) Dissection and mobilization of ureter. (B) Psoas hitch technique. (C) Anastomosis of the spatulated ureter to the tailored bladder flap. (D) Completed Boari flap reimplantation. The Foley catheter was clamped and 200–300 mL of normal saline was instilled into the bladder to facilitate dissection from the anterior and lateral pelvic walls and to assess bladder mobility. If bladder mobilization was inadequate, a psoas hitch was performed by securing the mobilized bladder to the ipsilateral psoas tendon using three absorbable sutures ( Figure 2B ). For longer defects, the distance from the proximal ureteral stump to the psoas hitch fixation site was measured with a scaled stent, and an anterior bladder wall Boari flap was created 1–2 cm longer than the defect, with an apex width of 1–2 cm and a base width of 3–4 cm. The spatulated ureter was anastomosed side-by-side to the apex of the tailored bladder flap using 4–0 sutures ( Figure 2C ). The flap was subsequently tubularized using continuous 3–0 barbed sutures ( Figure 2D ). In selected cases requiring an anti-reflux mechanism, one of two anti-reflux ureterovesical anastomotic techniques was used according to the ureteral stump caliber. For normal or mildly dilated ureters, submucosal tunnel reimplantation was performed. After a seromuscular incision, a submucosal tunnel was developed in the bladder wall, with the tunnel length tailored to the ureteral diameter to maintain an approximate tunnel length-to-ureteral diameter ratio of 3:1 ( Figure 3A ). Mucosa-to-mucosa ureterovesical anastomosis was then performed over a double-J stent, followed by closure of the seromuscular layer ( Figure 3B ). For markedly dilated ureters, anti-reflux nipple reimplantation was performed by everting the distal 1–1.5 cm of the ureter to form a cuff-like intravesical nipple ( Figure 3C ), which was anastomosed to a full-thickness cystotomy ( Figure 3D ). Ureteral tailoring was performed before nipple formation when the distal ureteral lumen was excessively wide or megaureter existed. Representative intraoperative images of anti-reflux ureteral reimplantation techniques. (A) Formation of the submucosal tunnel, outlined by yellow dashed lines. (B) Completed submucosal tunnel reimplantation. (C) Formation of the ureteral nipple. (D) Completed nipple reimplantation. Most patients were hospitalized for 3–4 days until drainage tube removal. The urethral catheter was removed 2 weeks postoperatively, while the ureteral stent was removed at 2–3 months. For patients with nephrostomy, the tube was clamped 1–2 weeks postoperatively. Videourodynamic (VUD) assessment of the upper urinary tract was conducted 1 week after stent removal; the nephrostomy tube was subsequently withdrawn once VUD confirmed unobstructed contrast drainage and the absence of extravasation. The same follow-up protocol was applied to both groups: clinical evaluation, serologic testing, and renal ultrasonography at 3-month intervals; functional cine magnetic resonance urography (cine MRU) and computed tomography urography (CTU) were performed at 3 and 6 months postoperatively, respectively. The primary clinical outcome was mid-term surgical failure, defined as unresolved symptoms (flank pain, fever, or recurrent urinary tract infection) or radiographic obstruction on cine MRU or CTU, with or without progressive deterioration of split renal function during follow-up, necessitating continued drainage or further intervention ( 13 ). Secondary outcomes were operative time, estimated blood loss, postoperative hospital stay, postoperative serum creatinine (sCr), postoperative estimated glomerular filtration rate (eGFR), and postoperative complications. Perioperative complications were categorized according to the Clavien-Dindo classification. The full cohort of 153 patients constituted the primary analytic population. For the unadjusted comparisons, continuous variables were presented as mean ± standard deviation (SD) or median [interquartile range (IQR)], depending on the normality of the data, and group comparisons were performed using either an independent samples t-test or the Mann-Whitney U test. Categorical variables were expressed as frequency and percentage (%), and group differences were analyzed using the Pearson’s Chi-squared test or Fisher’s exact test, as appropriate. For the primary adjusted analyses, continuous outcomes were modeled using multivariable linear regression with HC3 heteroscedasticity-robust standard errors, incorporating clinically relevant covariates that directly influence surgical difficulty, including ureteral stricture length, preoperative double-J stent placement, preoperative nephrostomy, previous failed reconstruction, preoperative urinary tract infection, and reconstructive technique; results are reported as adjusted differences with 95% confidence intervals (CIs). Postoperative complications and surgical failures were analyzed using parsimonious Firth penalized logistic regression models to reduce small-sample bias and address potential separation, with adjustment for ureteral stricture length and previous failed reconstruction. Adjusted odds ratios (ORs) with profile-likelihood 95% CIs are reported. Propensity score matching (PSM) was retained as a secondary sensitivity analysis. Propensity scores were estimated using a multivariable logistic regression model with robotic platform as the dependent variable. Covariates were selected a priori based on clinical relevance and included age, sex, body mass index (BMI), laterality, stricture etiology, stricture length, preoperative sCr, preoperative urinary tract infection, preoperative double-J stent or nephrostomy, previous failed reconstruction, and reconstructive technique. Patients were matched 1:1 using nearest-neighbor matching without replacement with a caliper width of 0.2 SD of the logit of the propensity score. Baseline balance after PSM was assessed using two-sided P values, with P>0.05 indicating adequate balance. Following PSM, continuous variables were compared using the paired t-test or the Wilcoxon signed-rank test, and categorical variables were compared using McNemar’s test. No data were missing for the exposure, prespecified covariates, or outcomes. All statistical analyses were conducted using SPSS Statistics version 27.0 (IBM Corp., Armonk, NY, USA) and R version 4.4.2 (R Foundation for Statistical Computing, Vienna, Austria). All tests were two-sided, and a P value <0.05 was considered statistically significant.

The baseline characteristics of the patients are summarized in Table 1 . Of the 162 patients assessed, 9 were excluded and 153 patients were included in the final analysis, comprising 51 patients in the KangDuo group (KD group) and 102 patients in the da Vinci group (DV group). Most baseline characteristics were comparable between the two groups, including age, sex, BMI, laterality, preoperative serum creatinine, preoperative urinary tract infection, history of preoperative double-J stent placement and nephrostomy, surgical technique, and psoas hitch. However, the DV group had a significantly longer ureteral stricture length than the KD group [5.0 (3.0–10.0) vs. 4.0 (3.0–5.0) cm, P=0.01]. In addition, the proportion of patients with previous failed reconstruction was significantly higher in the DV group than in the KD group [32/102 (31.4%) vs. 7/51 (13.7%), P=0.03]. Data are presented as mean ± standard deviation, n (%), or median (interquartile range). BMI, body mass index; DJ, double-J; DV, da Vinci; KD, KangDuo; PSM, propensity score matching; sCr, serum creatinine; UVJO, ureterovesical junction obstruction. As shown in Table 2 , all procedures were successfully completed without conversion to conventional laparoscopy or open surgery. In the six-variable multivariable model, KD-SR-01 remained associated with a longer operative time (adjusted difference, 13.91 minutes; 95% CI: 1.94–25.88; P=0.02). No clear platform-associated differences were identified for estimated blood loss (adjusted difference, −2.01 mL; 95% CI: −8.92 to 4.91; P=0.57), postoperative hospital stay (−0.14 days; 95% CI: −0.97 to 0.68; P=0.73), postoperative sCr (2.77 µmol/L; 95% CI: −4.57 to 10.10; P=0.46), or postoperative eGFR (−2.81 mL/min/1.73 m 2 ; 95% CI: −9.31 to 3.70; P=0.40). Data are presented as mean ± standard deviation, n (%), or median (interquartile range). CI, confidence interval; DV, da Vinci; eGFR, estimated glomerular filtration rate; KD, KangDuo; sCr, serum creatinine. All patients completed the standardized follow-up assessments at 3 and 6 months postoperatively, followed by annual evaluations thereafter. The median follow-up time was comparable between the KD and DV groups [16.0 (13.0–21.0) vs. 17.0 (12.2–21.8) months, P=0.61]. No severe complications, defined as Clavien-Dindo grade ≥ III events, occurred in either group. Postoperative complications occurred in 10 patients: 8 urinary tract infections, 1 acute kidney injury, and 1 case of vesicoureteral reflux-related symptoms; all were managed conservatively. Three patients reached the primary outcome of surgical failure, and subsequently underwent conservative management or reintervention. The corresponding mid-term surgical success rates were 98.0% (50/51) in the KD group and 98.0% (100/102) in the DV group. In the Firth penalized models, the robotic platform was not associated with postoperative complications (adjusted OR, 0.91; 95% CI: 0.20–3.49; P=0.90) or with surgical failure (adjusted OR, 1.20; 95% CI: 0.09–10.93; P=0.87); the wide confidence intervals reflect the small number of events. The PSM sensitivity analysis retained 46 pairs. All baseline variables were well balanced between the two groups after matching, with no statistically significant differences observed across the matched covariates. Operative time remained longer with KD-SR-01 (155.5±33.8 vs. 136.0±40.4 minutes; P=0.01) ( Table 3 ). No statistically significant matched differences were observed for estimated blood loss, postoperative hospital stay, postoperative sCr, postoperative eGFR, complications, or surgical failure. Data are presented as mean ± standard deviation, n (%), or median (interquartile range). DV, da Vinci; eGFR, estimated glomerular filtration rate; KD, KangDuo; PSM, propensity score matching; sCr, serum creatinine.

Robot-assisted surgery has expanded the role of minimally invasive approaches in complex distal ureteral reconstruction by providing stable three-dimensional visualization and articulated instrumentation ( 14 ). However, direct comparative evidence regarding newly developed robotic platforms remains limited. In this study, we compared perioperative and mid-term outcomes between KD-SR-01-assisted and da Vinci-assisted ureteral reimplantation. After multivariable adjustment in the full cohort, KD-SR-01-assisted surgery was associated with a longer operative time, whereas estimated blood loss, postoperative length of stay, postoperative renal function, complication profile, and mid-term surgical success were comparable between the two platforms. All procedures were completed without conversion, and no Clavien-Dindo grade III or higher complications occurred. To our knowledge, this is the first comparative study evaluating KD-SR-01-assisted versus da Vinci-assisted ureteral reimplantation. The present findings extend previous single-arm evidence supporting the feasibility and safety of the KD-SR-01 platform in urologic reconstructive surgery. Han et al. reported favorable perioperative and functional outcomes for KD-SR-01-assisted ureteral reimplantation in a single-center, single-arm pilot cohort ( 11 ), but the lack of a comparator group limited interpretation of whether these outcomes reflected platform performance, surgeon experience, or case selection. Benchmarked against the da Vinci experience, in which Marien et al. achieved radiographic and symptomatic success in 98% and 97% of 250 consecutive robotic upper urinary tract reconstructions ( 15 ) and a contemporary single-center series reported stricture recurrence in only 10% of 63 robotic ureteral reconstructions ( 16 ), the mid-term success rates of both groups in the present cohort were of a similar order, indicating that neither platform compromised the durability of repair. Comparative studies in other urologic procedures have similarly suggested that KangDuo-assisted surgery can achieve perioperative and functional outcomes comparable to those of the da Vinci system, although operative time may be longer in some settings ( 17 - 19 ). Our findings are consistent with this pattern: KD-SR-01 appears capable of supporting technically demanding reconstructive procedures, while operative efficiency may still be influenced by platform maturity, surgeon familiarity, and procedural configuration. The longer operative time in the KD-SR-01 group should be interpreted carefully. Ureteral reimplantation is a technically demanding reconstructive procedure, and operative time alone may not fully reflect the quality of reconstruction or the durability of surgical success. In this study, the difference in operative time was approximately 14 minutes and was not accompanied by increased blood loss, prolonged hospitalization, deterioration in postoperative renal function, higher complication rates, or reduced surgical success. Several factors may explain this finding. First, although the surgeon had experience with both systems, cumulative experience with the da Vinci platform was greater, which may have conferred a platform-specific efficiency advantage ( 20 ). Consistent with this, a recent comparative review of next-generation robotic platforms observed that early inefficiencies with emerging systems may prolong non-surgical operative time until the surgical team reaches proficiency ( 6 ). Second, KD-SR-01 procedures were performed using a three-arm configuration, whereas da Vinci procedures were performed with a four-arm configuration; the additional robotic arm may facilitate traction, exposure, and suturing during pelvic reconstruction. Third, given the relatively recent clinical adoption of the KD-SR-01 platform, a more methodical operative workflow may have been followed during anastomosis, Boari flap tubularization, and anti-reflux reconstruction. This difference should nonetheless not be dismissed: although it did not compromise perioperative safety or reconstructive quality, an additional 14 minutes per case affects operating-room throughput and cost, and whether the gap narrows with accumulated experience and four-arm configurations warrants prospective evaluation. Importantly, after multivariable adjustment for the baseline imbalances in stricture length and previous failed reconstruction, and consistently in the PSM analysis, the two platforms showed comparable perioperative safety and mid-term reconstructive outcomes. This similarity is clinically meaningful because durable success after distal ureteral reconstruction depends not only on completing the anastomosis, but also on complete excision of diseased ureter, preservation of ureteral vascularity, avoidance of anastomotic tension, and appropriate reconstruction of the bladder-ureter interface ( 15 ). The comparable success rates observed in this study suggest that, in experienced hands, KD-SR-01 can support the essential reconstructive steps required for complex distal ureteral repair. Beyond achieving comparable perioperative and mid-term outcomes, the KD-SR-01 platform differentiates itself through its ergonomic design philosophy and sophisticated multimodal integration. By employing an open-console architecture, the system inherently mitigates the cervical strain associated with immersive platforms while fostering a more collaborative, communicative intraoperative environment ( 21 , 22 ). This hardware configuration is further leveraged by a dual-screen setup, which allows for the concurrent display of preoperative 3D reconstructions and real-time ICG fluorescence, a capability instrumental in demarcating stricture margins and identifying anatomical landmarks in complex redo cases ( 23 - 25 ). In contrast to the da Vinci Firefly system, where surgeons must frequently toggle between white-light and fluorescence modes, the KD-SR-01 provides simultaneous NIRF and white-light visualization. This parallelism ensures uninterrupted surgical manipulation and superior spatial orientation by maintaining a constant view of the surrounding anatomy during fluorescence-guided navigation. Furthermore, the system’s universal compatibility with commercially available 3D endoscopic platforms not only mitigates costs associated with specialized equipment but also provides high-definition optical clarity ( 11 , 26 ). Mechanically, the integration of force sensors and cross-laser guidance optimizes the docking workflow, while the synchronous rotation of the three-arm beam facilitates complex maneuvers without necessitating cart repositioning ( 17 , 18 ). The next-generation four-arm KangDuo SR2000 platform represents a meaningful evolution of the robotic system and may improve operative efficiency relative to the three-arm KD-SR-01 configuration, thereby broadening its applicability in complex reconstructive urology ( 19 ). Finally, its robust performance in 5G-enabled telesurgery underscores the system’s potential for remote mentoring and expanding access to specialized surgical expertise ( 18 , 27 ). This study has several limitations. First, platform allocation was nonrandom and based solely on patients’ voluntary choice, without surgeon-directed selection. However, the da Vinci group had longer strictures and more previous failed reconstructions, which may reflect patients’ platform preferences and therefore represents a potential source of selection bias. Although multivariable adjustment and propensity score matching addressed measured differences, residual confounding cannot be excluded. Second, all procedures were performed by a single experienced surgeon at a high-volume tertiary center, which minimized inter-surgeon variability but may limit generalizability to other institutions and surgeons at different stages of the learning curve. Third, the limited cohort size and only three surgical failures resulted in imprecise estimates and insufficient statistical power for definitive comparisons of surgical success between the two platforms. Future multicenter randomized controlled trials involving surgeons with varying levels of experience are warranted to comprehensively assess the KD-SR-01 robotic system in terms of long-term therapeutic outcomes, quality of life analysis, economic efficiency, and learning curves in comparison with other robotic platforms.

The KD-SR-01 system achieves perioperative safety and mid-term clinical outcomes comparable to the da Vinci platform for complex ureteral reimplantation. Despite a marginally longer operative time, its integrated multimodal imaging and ergonomic advantages position this emerging platform as a feasible alternative in the global landscape of robotic reconstructive urology.

The article’s supplementary files as