Intraoperative Performance of DaVinci Versus Hugo RAS During Radical Prostatectomy: Focus on Timing, Malfunctioning, Complications, and User Satisfaction in 100 Consecutive Cases (the COMPAR-P Trial)

Take Home Message The present prospective study compared 50 versus 50 unselected patients who underwent robotic prostatectomy performed with the DaVinci versus Hugo RAS system. Although more malfunctioning/troubleshooting events were recorded during Hugo RAS cases, the surgery outcomes, including the occurrence of intraoperative complications, were not statistically different. Longer operative time was recorded for Hugo RAS cases, likely explained by the meticulous care applied in the surgery due to the use of the novel platform. Still, the prospect of improvement within a relatively low number of procedures was evident. All these findings can contribute to the adoption of this new platform.


Introduction
The DaVinci surgical system (Intuitive Surgical, Sunnyvale, CA, USA) has long been the gold standard in roboticassisted surgery, dominating the field with its surgeonfriendly interface.Since its introduction, it has become synonymous with minimally invasive robot-assisted radical prostatectomy (RARP), known for its significant benefits.
After a 2-decade DaVinci monopoly, novel robotic platforms are now approved for clinical use.Such a striking event has generated the opportunity to open the market to competition with the expectation of reducing the costs of robotics [1].Among the systems emerging as other frontrunners for urology in the European Community, the Hugo RAS system (Medtronic, Minneapolis, MN, USA) is gaining popularity.This platform has introduced peculiar features such as modular robotic arms and an open console architecture [2].
However, to date, this platform has shown uneven adoption across institutions and is still supported by limited scientific evidence, primarily comprising a few retrospective and single-arm studies.Notably, as of now, no planned prospective comparative trials have been published, but further evidence in this regard is eagerly awaited because an understanding of the nuanced differences between platforms, encompassing efficacy, ergonomics, cost effectiveness, and learning curves, likely represents a crucial issue for health care providers and patients alike [3].
Our institution promoted a prospective head-to-head comparison of newly introduced platforms with the ''gold standard,'' represented by the DaVinci system.The project, granted by the regional health care system, encompasses various surgical procedures across different specialties, including prostatectomy, partial nephrectomy, stomach, liver, pancreatic resections, colectomy, hysterectomy, etc.The ''prostatectomy arm'' of the project has recently been concluded; the present paper is focused on the comparison between the Hugo RAS and DaVinci Xi systems for RARP, with the aim to provide granular and comprehensive data on intraoperative performance of the platforms.

Patients and methods
The Comparison of Outcomes of Multiple Platforms for Assisted Robotic surgery-Prostate (COMPAR-P) trial is a monocentric, postmarket study promoted by the Azienda Ospedaliera Universitaria Integrata (AOUI) of Verona (Italy) within a broader project concerning different procedures and specialties [4].The study received local ethical committee approval (4038CESC) and was registered on clinicaltrials.org(NCT05766163).
Enrollment began in March 2023.Unselected patients with organ-confined prostate cancer candidates to RARP at our department were allocated to one of the tested platforms, up to the completion of enrollment of 50 consecutive cases in each cohort.All patients signed informed consent: these documents were stored in a designated locker.The only exclusion criterion was the patient's refusal to consent to participate in the study.
The involved operative room personnel received dedicated intensive 3-d training at the ORSI Academy (Melle, Belgium).Two console surgeons (A.A. and V.D., with previous experience of >1000 and >500 DaVinci RARP [DV-RARP] interventions, respectively) performed the procedures.None of the surgeons carried out any procedure with the Hugo RAS before the recruitment for this study commenced.Three other surgeons, experienced in robotic assis-tance (A.V., R.R., and R.B.), were involved as table assistants.Anesthesiological protocol as well as patient positioning (supine position, legs joined and extended, and Trendelenburg 25°) was the same for DaVinci and Hugo RAS.Port placement, instead, varied according to the platform: for the DaVinci Xi system, four robotic trocars were put in line 2 cm above the umbilicus, plus two additional trocars (12 and 5 mm) triangulated to the robotic ones on the assistant side; for the Hugo RAS system, a W-shaped configuration was employed for the four robotic trocars, plus two additional trocars [5].Equivalent additional resources were used across the two platforms concerning insufflation and aspiration systems, robotic instruments (monopolar scissor, Maryland bipolar forceps, fenestrated grasp, and needle driver), assistant instruments, sutures, and clips.No variation in surgical technique was implemented depending on the platform [5].The laparoscopic approach was transperitoneal, using a 0°camera.An anterior antegrade dissection was performed, intra-, inter-, or extrafascially, according to clinical data; posterior and anterior reconstructions and anastomosis were done with barbed sutures.Lymph node dissection (LND) was indicated depending on the risk of lymph nodal invasion, as calculated by nomograms following an extended template [6].
The aim of the present study is the evaluation of the intraoperative performance of the DaVinci and Hugo RAS platforms, by assessing the following endpoints: 1. Timing and learning curve, as referred to the following phases of the procedure: (a) Setup phase, composed of the following steps: ''room configuration,'' ''patient positioning,'' ''port placement,'' ''draping,'' ''docking,'' ''undocking,'' and ''undraping'' (b) Console phase, composed of the following steps: ''opening of the umbilical-prevesical fascia,'' ''preparation of the Retzius space,'' ''dissection of the bladder neck,'' ''dissection of the seminal vesicles,'' ''dissection of the posterior plane,'' ''management of prostatic pedicles and eventual nerve sparing,'' ''dissection of the prostate apex and urethra,'' and ''posterior reconstruction and urethrovesical anastomosis''; a separate comparison was done for LND 2. Malfunctioning, defined as any technical abnormality that persistently impaired the function of the robot; the events were detailed in terms of the type, duration, and management.3. Complications, defined as any deviation from the regular intraoperative course of the intervention; the events were detailed in terms of the type and management, and graded according to the intraoperative adverse incident classification proposed by the European Association of Urology Ad Hoc Complications Guidelines Panel [7]; 4. Users' satisfaction, determined by a visual assessment scale (1-poor to 5-optimal): (a) First surgeon, assistant, and scrub nurse satisfaction concerning the whole procedure (b) First surgeon satisfaction with each robotic instrument (c) First surgeon satisfaction with the quality of the vision (image definition and depth perception), bimanual dexterity, instrument mobility (promptness, precision, and range of movements), force control, and usability of console commands Deidentified data were prospectively collected in a Research Electronic Data Capture (REDCap) dataset during surgery by a dedicated investigator who was not directly involved in the procedure.The REDCap could be opened by a dedicated password-coded computer stored within a locker.Password was changed periodically and managed exclusively by the involved investigators.

Statistical analysis
The statistical analysis was performed according to guidelines [8].The sample size was established according to study feasibility [4].Both the Shapiro-Wilk test and the graphical assessment were adopted for assessing data distribution.
The mean and standard deviation (SD), or the median and interquartile range (IQR) were used for reporting continuous variables based on data distribution.Frequencies and percentages were adopted for dichotomous data.In the case of normal distribution, the Student t test was chosen to evaluate the differences between the two groups, whereas the Mann-Whitney U test was deemed suitable for the nonparametric variables.The differences in the case of discrete variables were evaluated using Pearson's v 2 test.To assess the learning curve for Hugo RAS, the single and overall times evaluated were reported graphically along the progression from the first to the 50th case, and compared with the corresponding DaVinci cases.Additionally, to define the learning curve for Hugo RAS, the cumulative summation analysis (CUSUM) was used.This method represents data from consecutive procedures, transforming raw data into a cumulative sum of differences between single values and the overall mean, described graphically by a curve: the breakpoint from the ascending to the descending portion indicates the number of cases when the transition from a learning to a proficiency phase occurs [9].All the tests were two sided, with statistical significance set at p 0.05.Analyses were performed using STATA 18.0 (1996-2024; StataCorp LLC, Lakeway Dr, College Station, TX, USA).The following STATA syntax was adopted to run the tests: swilk, histogram, ttest, tabstat, ranksum, tabulate, chi2, and column.For the CUSUM estimation, Microsoft Excel 2023 (Microsoft 2024, Redmond, WA, USA) was used.

Results
Fifty and fifty patients were enrolled during the study period to be submitted to DV-RARP and RARP performed by the Hugo RAS (H-RARP) platform, respectively.Of interest, each of the involved console surgeons performed 25 cases with Hugo RAS.
The comparison of baseline features showed that groups were balanced, with differences limited to Charlson Comorbidity Index, cN+ rate-higher in the DV-RARP group, and Briganti (2018) score [6]-higher in the H-RARP group (Table 1).
The graphical longitudinal analysis of timing-from the first to the 50th procedure-showed an almost flat line for the DV-RARP group, both for the setup and the console phase.Conversely, in the H-RARP series, the setup time duration declined modestly along with experience, while the console time duration declined markedly and tended to the DV-RARP values (Fig. 2; more details are provided in the Supplementary material).The analysis of the learning curve by the CUSUM graphs showed a breakpoint at 22 and 17 H-RARP procedures for the setup and console phases, respectively (Fig. 3).
Respectively, for the DV-RARP and H-RARP groups, four (time consumption 1-2 min) and 20 (time consumption up to 45 min) events of malfunctioning, as well as two and three events of intraoperative complications were reported (detailed in the Supplementary material).The median (IQR) estimated blood loss was 200 (150-300) and 300 (150-400) ml, respectively (p = 0.1).

Discussion
The present study prospectively collected granular data to compare the DaVinci and Hugo RAS robotic platforms for performing RARP.It confirmed the expected difference in timing favoring the DaVinci system, both during the setup and the console phase.This can partly be related to the need to get confidence with a new ''surgical instrument'' but potentially also to the intrinsic differences between the platforms.
The graphical longitudinal analysis of the setup and console timing for DaVinci cases revealed an almost flat line, confirming that the team involved had well surpassed the learning curve.The same analysis for the Hugo RAS setup timing showed a mild decline along with experience, although it did not reach the efficiency level of DaVinci values.The breakpoint from learning to proficiency was identified in 22 cases.This finding suggests that the modular configuration with independent arms required additional efforts, particularly related to the need for separate draping/undraping and a more laborious docking process in terms of arms' angles and port positioning.
The modest satisfaction indexes reported by the assistant surgeon and scrub nurse align reasonably with the additional work required, as well as the greater discomfort experienced due to the table side being ''crowded'' by the larger arms of the Hugo RAS.In a recent publication focus- ing on the perspectives of the bedside assistant, other authors emphasized the significance of her/his active role.This active involvement is crucial to ensuring the success of Hugo RAS pelvic procedures and preventing issues such as crushes, collisions, and other technological concerns that could negatively impact surgical performance [10].
In contrast, the decline in console times was steeper, with an earlier breakpoint set at 17 cases, and tended to overlap the DaVinci times toward the end of our experience.This indicates that once the system is docked, an expert surgeon can duplicate the surgical gestures adopted with the DaVinci with similar effectiveness.
However, it should be noted that this result also entails a certain degree of adaptation, as evidenced by the first surgeon's indicators of satisfaction regarding the overall per-formance of the platform, quality of vision and movement, and single robotic instrument.Although absolutely encouraging-and higher than the scores from the table personnel, these values were lower than those reported for DaVinci.
A crucial element to be considered is the significantly higher number of malfunctioning events recorded during Hugo RAS cases (20 vs 4).Remarkably, half of these events, including platform battery supply alarm, system power failure, conflict of arms, scissors rupture, malfunctioning of Maryland bipolar forceps, and failed calibration of the fourth arm, required over 5 min for resolution, significantly impacting the procedure flow and the total procedure time.
Overall, our data suggest that the new platform was ''accepted'' earlier by the console surgeon, while various ''behind the scene'' issues could complicate the assistance to  RARP for other operating room personnel, despite their skill, training, and full dedication.
Our data align with previous experiences reporting that, for a surgeon with consistent DaVinci experience, skills are transferable soon from DaVinci to Hugo RAS when performing RARP and other pelvic procedures [11][12][13].
A few publications dealt with the preliminary experiences of RARP performed with Hugo RAS.The group headed by Dr. Mottrie described the first five RARP interventions performed with the Hugo RAS platform [2].The authors reported a median console time of 120 min (IQR 110-150), which is very similar to that recorded from our experience.On the contrary, the authors judged both ports' placement and docking as ''straightforward and rapid'': we share the same perception based on our experience, but the more comprehensive assessment of the setup and the prospective comparison with DaVinci probably render our findings more solid.At the same time, the more intensive training done by this group [14], practicing in both dry and wet laboratory settings and testing the best operative setup on human cadavers, led to a faster reduction of some troubleshooting that we recorded in our series.It should be acknowledged that the training experienced by our group represents ''real-life'' conditions for most urologists.Still, we suggest that for the implementation of new technologies, mandatory laboratory-based training to proficiency should be considered, prioritizing patient safety over accepting such phenomena as reflective of real-life conditions.Other groups have published their very initial experience with H-RARP [15][16][17], but only Ou et al [18] described the need to find optimization in ports' placement and docking.The authors observed some intraoperative issues that slowed the procedure flow.The intraoperative pauses for troubleshooting were mostly due to the mispositioning of the docking ports, which caused collisions of the robotic arms.Dr. Mottrie's team followed with the publication of more consistent case series, including up to 112 patients, which confirmed the data reported at the start of the experience [19,20].Conversely, just a couple of papers attempted a comparison between the Hugo and DaVinci platforms, although in a retrospective manner.The first came from India, with 17 versus 17 patients compared, without finding significant differences [21].The second, again from Mottrie's group, analysed 378 versus 164 patients who underwent DV-RARP versus H-RARP [22].The median total operative time was slightly longer for Hugo RAS procedures than for DaVinci surgeries-180 (IQR 150-200) versus 165 (IQR 130-200) min, a difference that was more pronounced in procedures without LND.Interestingly, our experience also confirms that LND did not exacerbate the differences between the platforms, resulting in comparable durations regardless of the robot used, likely because of the requirement of minimal assistance and instrument exchange.It is interesting to note that, with respect to the surgical learning curve for operative time with Hugo RAS, Mottrie and team did not find any association between increasing surgical experience and the time to complete surgery, possibly because the larger number of cases masked the impact of the learning curve or, alternatively, due to some bias in the recording of times related to the retrospective study design.
Nevertheless, despite the system's novelty at our institution and the technology disparity, it is crucial to emphasize that all surgeries were completed successfully within acceptable operative times and with comparable intraoperative complication rates to the fourth-generation DaVinci platform.This major finding confirms that, since its first generation, the Hugo RAS platform has the potential to be introduced rapidly into the clinical routine of a highvolume institution with an experienced team.Furthermore, we believe that some users could particularly appreciate some peculiar characteristics of the Hugo RAS platform, namely, the open console, different three-dimensional image quality of the console screen, sharpness of the mono- poly shears, and efficiency of the monopolar coagulation.However, it is important to note that the user's perception of these features is highly subjective.
We acknowledge the main limitations of the present study, including the relatively small sample size and the limited generalizability of our findings.The study was underpowered to detect significant differences between treatment groups in some secondary endpoint outcomes, specifically intraoperative complications.This limitation is attributed to their relative rarity in the setting of a surgical trial conducted at a tertiary institution by expert surgeons.
We emphasize that the comparison was between the first 50 H-RARP cases and the last 50 DV-RARP cases performed at our institution.This distinction likely accounts for the observed differences in setup and console times between the treatment cohorts.While this aspect may be a common limitation in studies of this nature, we believe that our methodology remains robust, showcasing the strength of our overall effort.Furthermore, our results directly fit those of the surgeons experienced in DaVinci surgery who are contemplating a shift to the Hugo RAS platform.This scenario is likely to be one of the most common contexts for the adoption of this new robotic platform.
On the contrary, we underscore that the specialized expertise of the surgeons involved may restrict the direct applicability of the outcomes to settings with varying levels of surgical proficiency or experience.Introducing the Hugo RAS platform into a robotic-naïve center might present more challenges.To the best of our knowledge, there is a paucity of knowledge about this specific setting.As such, almost all the urological surgeons who started their experience with Hugo RAS had prior consistent experience with DaVinci robotic surgery.It is interesting to note that a cross-sectional study investigated surgeons of different robotic backgrounds who participated in a hands-on session with the Hugo RAS simulator (none of them had prior expertise with the system) and found that prior robotic console expertise improved basic skills at the Hugo RAS simulator.This undoubtedly has implications for skill transference across different platforms [23].We strongly believe that, for centers/surgeons who are robotic surgery naïve, it would be advisable to invest in robust training programs, collaborate closely with experienced trainers or centers already using the Hugo RAS platform, and ensure a gradual integration process to optimize the transition.

Conclusions
The present prospective study compared 50 versus 50 unselected patients who underwent DA-RARP versus H-RARP.Although more malfunctioning/troubleshooting events were recorded during Hugo RAS cases, the outcomes of surgery, including the occurrence of intraoperative complications, were not statistically different.Longer operative time was recorded for Hugo RAS cases, likely explained by the meticulous care applied in the surgery due to the use of the novel platform.Still, the prospect of improvement within a relatively low number of procedures was evident.All these findings can contribute to the adoption process of this new platform.
Society of Anesthesiologists; BMI = body mass index; CAD = coronary artery disease; ED = erectile dysfunction; IIEF-5 = five-item version of the International Index of Erectile Function; IPSS = International Prostate Symptom Score; IQR = interquartile range; ISUP = International Society of Urological Pathology; MRI = magnetic resonance imaging; PCa = prostate cancer; PI-RADS = Prostate Imaging Reporting and Data System; PSA = prostate-specific antigen; QoL = quality of life; SD = standard deviation; * missing data.E U R O P E A N U R O L O G Y O P E N S C I E N C E 6 3 ( 2 0 2 4 ) 1 0 4 -1 1 2

Fig. 1 -
Fig. 1 -Segmentation of DaVinci-and Hugo RAS-related timing expressed as median duration.The setup phase is at the extremes of the yellow dashed lines.

Fig. 2 -
Fig. 2 -The median duration of the (A) setup and (B) console phases is shown for each procedure of the series (1-50).

Fig. 3 -
Fig. 3 -(A) CUSUM curve for the overall operative time (setup + console) of the Hugo RAS platform.Respective CUSUM curves of (B) setup and (C) console times are shown separately.CUSUM = cumulative summation analysis.