Abstract
Purpose:
To evaluate the efficacy of supine percutaneous renal access by robot-assisted (RA) fluoroscopy and ultrasound (US) guidance in terms of procedural outcomes and surgeon workload.
Methods:
We conducted a multicenter, randomized, controlled benchtop study involving 32 urologists using a renal phantom model. RA puncture was performed using the developed version of automated needle targeting with X-ray (ANT-X), which determines the direction of the needle. US puncture was performed under US guidance. The primary endpoint was the single-puncture success rate, and the secondary outcomes were the procedural time for each step, time of fluoroscopic exposure, and workload assessment.
Results:
The single-puncture success rates were 90.6% and 56.3% for RA and US punctures, respectively (p < 0.01). In RA puncture, the median device setup time was 120 seconds longer, the median total procedural time was 100 seconds longer, the median time of fluoroscopic exposure was 40 seconds longer, the median needle puncture time was 17 seconds shorter, and the distance from the target sphere was 1 cm shorter than those in US puncture (all p < 0.01). The mental and physical task workload, effort required by the surgeons, frustration felt by the surgeons, and overall National Aeronautics and Space Administration Task Load Index scores were lower in the RA puncture group than in the US puncture group (p = 0.01, p = 0.046, p < 0.01, p = 0.021, and p ≤ 0.01, respectively).
Conclusions:
RA puncture using ANT-X, which can also be used for puncture in the supine position, offers advantages over renal puncture in terms of accuracy and surgical workload.
Introduction
Percutaneous nephrolithotomy (PCNL) is an effective surgical procedure for removing large kidney stones. 1 Adequate renal puncture is required for the safety and efficacy of patient outcomes, and puncture errors may damage the surrounding tissues and organs. 2,3 This could be challenging, depending on the anatomical features of the renal collection system and the shape of the stones. 4,5
Currently, fluoroscopy- and ultrasound (US)-guided punctures are the primary PCNL methods. 6,7 The advantage of fluoroscopy-guided puncture is the visualization of the renal collecting system with which most urologists are familiar; therefore, percutaneous access, including selection of the renal trajectory, can be achieved relatively easily.
On the other hand, US puncture does not induce radiation exposure and provides additional information regarding the surrounding organs. However, the use of US needs certain training with ideal diary use in practice. 4,7,8 In recent years, the combined use of fluoroscopy and US guidance has been reported to increase puncture accuracy while reducing radiation exposure. 9 Moreover, ureteroscopy-assisted renal puncture is useful for shortening the puncture and surgical times and improving the stone-free rate. 10
Recently, a new robotic system for assisting renal puncture, automated needle targeting with X-rays (ANT-X), was developed. It has features, such as synchronization with fluoroscopy images and calculation of the optimal renal puncture line using artificial intelligence (AI). 11 –13 We previously reported that renal puncture using ANT-X with the bullseye method in mini-PCNL reduces the number of renal punctures required for successful access. 14 In addition, this result is more favorable for residents. However, this version of the ANT-X is an optimized model for guiding puncture only in the prone position.
Since Valdivia's report in 1987, PCNL in the supine position has also become widely used because of its advantages, such as optimal cardiovascular and airway control, no need of changing the patient's position, and better stone fragment washout. 15 However, the difficulty of the puncture increases in the supine position due to the limited area that can be punctured. To increase the success rate of renal puncture, it is necessary for the surgeon to accumulate experience and training.
Recently, a new version of the ANT-X using the bi-plane method, which can also be used for puncture in the supine position, has been developed. In the current study, to investigate the efficacy of ANT-X using the bi-plane method, we conducted a benchtop study using a phantom kidney model and compared robot-assisted (RA) fluoroscopy-guided puncture with US puncture.
Materials and Methods
Study design
This multicenter randomized controlled benchtop study using renal phantom models was conducted at Nagoya City University (NCU) and Wakayama Medical University (WMU) hospitals. Thirty-four urologists, including 19 and 15 from NCU and WMU hospitals, respectively, participated in this study. Being junior residents, senior residents, fellows, and attendings were defined as those with <2, 3 − 5, 6 − 10, and 10 years of urologic experience, respectively.
Block randomization was performed using an online software after adjusting for the participants' years of urologic experience (Fig. 1). Each participant was assigned to Group A or B. In Group A, two participants were unable to participate in the experimental schedule due to work inconvenience. In Group A, the participant first underwent a US puncture, followed by an RA puncture. In Group B, the sequences were the opposite.

Study design. Thirty-four urologists, consisting of 19 urologists from NCU Hospital and 15 urologists from WMU Hospital, participated in this study. Each participant was assigned to Group A or Group B after adjusting for the participants' years of urologic experience. In Group A, two participants were unable to participate in the experimental schedule due to work inconvenience. In Group A, the participant first conducted US puncture, followed by RA puncture. In Group B, these sequences were reversed. NCU = Nagoya City University; RA = robot-assisted; US = ultrasound; WMU = Wakayama Medical University.
After completing the punctures with 32 participants, an additional investigation was conducted with 8 participants to examine the three-dimensional (3D) spatial relationship between both punctures. These eight participants performed both US and RA punctures. All procedures were performed in the shockwave lithotripsy room at NCU Hospital and in the operative room at WMU Hospital.
The primary endpoint was the single-puncture success rate, and the secondary outcomes were the procedural time of each step, time of fluoroscopic exposure, and workload assessment.
Procedures
Assuming a puncture in the supine position, the puncture was performed from the side of the phantom. The participants were asked to choose one of five spheres contained within the phantom body as the target. When the tip of the needle entered the target sphere, the puncture was considered successful. The needle tip position was confirmed by fluoroscopy during the RA puncture, whereas US was used for assessment during the US puncture.
RA puncture was performed using the developed version of the ANT-X, which determines the direction of the needle using the bi-plane method. The participants determined the approximate insertion point and direction of the needle toward the target on the side of the phantom (Fig. 2a). The direction in which the needle should be advanced was determined using the bi-plane method under fluoroscopic guidance (Fig. 2b).

Procedure in RA puncture. Participants determined the approximate needle insertion point and direction toward the target on the side of the phantom
The needle was then advanced along this line (Fig. 2c). For the US puncture, the participants confirmed the location of the target sphere by applying a US probe (ARIETTA 70; Hitachi, Ltd., Tokyo, Japan) from the side of the phantom (device setup), and the needle was subsequently inserted toward the target sphere (puncture). During the puncture, a needle guide attachment was used to standardize US puncture skills.
Data collection
The target sphere location, single-puncture success rate, device setup time, needle puncture time, total procedural time, and total fluoroscopic exposure time were recorded. After a successful puncture, the distance between the tip of the needle and the center of the target sphere (distance from the target sphere) was measured using fluoroscopy.
After the participants completed the puncture, the workload of the procedure was assessed using the National Aeronautics and Space Administration Task Load Index (NASA-TLX). 16,17 The NASA-TLX is a subjective questionnaire for workload assessment that includes six subscales: mental, physical, and temporal task demands; performance; effort; and frustration.
In the additional investigations, the distance between the puncture site and needle tip and the angle between the puncture needle and phantom surface were measured. These measurements were recorded in two orientations: the angle with the C-arm vertical to the ground and the angle with the C-arm tilted at 45°.
Statistical analysis
Categorical variables are presented as frequencies (%), and continuous variables are expressed as medians (interquartile ranges). Categorical variables were compared using Fisher's exact test, and continuous variables were compared using the Mann–Whitney U test. All statistical analyses were performed using EZR for R software (R Project 3.6.3). 18 Statistical significance was set at p < 0.05.
Results
The characteristics of the participants are summarized in Table 1. The number of junior residents, senior residents, fellows, and attendings was 7 (20.6%), 11 (32.4%), 5 (14.7%), and 11 (32.4%), respectively. The number of participants in Group A was 2 (12.5%), 6 (37.5%), 3 (18.8%), and 5 (31.3%), respectively. In Group B, the number of participants was 5 (27.8%), 5 (27.8%), 2 (11.1%), and 6 (33.3%), respectively. There was no significant difference in participants' clinical experience between the two groups (p = 0.749).
Characteristics of the Participants
A total of 32 sets of RA and US punctures were performed. A comparison of the procedural outcomes between RA and US punctures is shown in Table 2. There was no significant difference in the location of the target sphere between the two puncture methods. The single-puncture success rate was higher in the RA than in the US puncture group (90.6% vs 56.3%, p < 0.01). In the RA puncture group, the median device setup and fluoroscopic exposure times were 120 and 40 seconds longer than those in the US puncture group, respectively (all p < 0.01).
Comparison of Procedural Outcomes Between Robot-Assisted and Ultrasound Punctures
RA = robot-assisted; US = ultrasound.
However, the median needle puncture time was 17 seconds shorter, and the distance from the target sphere was 1 cm shorter in the RA puncture group (both p < 0.01). The results of additional investigation performed by eight participants are shown in Supplementary Table S1.
The single-puncture success rate was equivalent to that of the study conducted by 32 participants. In addition, there were no significant differences in the distance between the puncture site and needle tip and the angle between the puncture needle and phantom surface for both punctures in both orientations. These results show that there are no significant differences in the 3D spatial relationship between both punctures.
Table 3 presents the results of the NASA-TLX scores. The mental and physical task workloads were 59.2% and 86.0% lower in the RA puncture group than in the US puncture group, respectively (p = 0.01 and 0.046, respectively). Moreover, the surgeons' effort and frustration felt were 73.0% and 80.0% lower in the RA puncture group than in the US puncture group, respectively (p < 0.01, p = 0.021, respectively). Overall, the NASA-TLX scores were 58.8% lower in the RA puncture group than in the US puncture group (p < 0.01).
Comparison of National Aeronautics and Space Administration Task Load Index Between Robot-Assisted and Ultrasound Punctures
TLX = Task Load Index.
Table 4 describes the subgroup analysis results comparing the outcomes of RA and US puncture among residents, fellows, and attendings. In the fellows and attendings groups, no significant differences were observed between the two puncture methods in terms of the single-puncture success rate, median needle puncture time, and distance from the target sphere.
Results of the Subgroup Analysis, Comparing Outcomes Across Experience Levels Between Robot-Assisted and Ultrasound Punctures
However, focusing on junior and senior residents, RA puncture demonstrated a higher single-puncture success rate (94.4% vs 44.4%, p < 0.01), a 20-second shorter median needle puncture time (p < 0.01), and a 1-cm shorter distance from the target sphere (p = 0.031) than US puncture.
In both the residents and the fellows and attendings, as was the case in the overall analysis, the median device setup, median total procedure, and fluoroscopic exposure times were longer, and the overall TLX score was lower for RA puncture than for US puncture.
Discussion
The present study compared the efficacy of RA and US puncture using a phantom model with regard to procedural outcomes and surgeon workload using the NASA-TLX. The results demonstrated that RA puncture had a higher single-puncture success rate and required longer device setup, total procedure, and fluoroscopic exposure times than US puncture.
Moreover, RA puncture had a shorter needle puncture time and distance from the target sphere and lower mental and physical task workload, effort and frustration scores, and overall TLX scores than US puncture. These findings suggest that RA puncture offers advantages over renal puncture in terms of accuracy, safety, and surgical workload.
Recently, the use of robotic systems has increased in various surgical fields, including urology. 19,20 The use of robotic systems in renal puncture has been reported in several studies and has been shown to improve the accuracy and safety of the procedure. 21 –23 Ritter et al. reported punctures in urology using 3D laser-guidance with the Uro Dyna-CT (Siemens Healthcare Solutions, Erlangen, Germany). 24
This guidance enables the visualization of the morphology of the target renal calyx and can be adapted for complex punctures; however, it requires extensive equipment. Moreover, Lima et al. reported kidney punctures using guidance by the 3D navigation with electromagnetic sensors and simultaneous confirmation using a ureterorenoscope. 25
In this guidance, extensive equipment is not required; however, it cannot be used in cases where the kidney stones are large or in the presence of staghorn stones, as the electromagnetic sensor cannot be inserted into the target renal calyx. The ANT-X robotic system has unique features, such as compact size, ease of use, and automation of the needle puncture trajectory toward target calyx visualized by contrast agent provided by the AI platform. 11 –14
Our previous study using ANT-X with the bullseye method in a clinical setting showed that in US puncture, it was necessary to change the surgeon in 14.3% of cases due to difficulty in determining the puncture route, whereas there was no need to change the surgeon during RA puncture. 14 This previous version of ANT-X with the bullseye method can only be used in the prone position, whereas the device used in the current study utilizes the bi-plane method to enable renal access in the supine position.
The previous model of ANT-X used in the prone position was placed over the phantom or patient. The device used in this study for the supine position, as shown in Figure 2, was positioned on the side of the phantom, and the placement could be approximate. In actual surgical use, it is positioned on the patient's upper body in a twisted posture and ANT-X is placed on the side of the body.
Moreover, in the bullseye method, the puncture point is determined by the initially set C-arm angle; however, in the bi-plane method, the puncture point can be determined from a wide range. Further, CT image reading is not necessary; instead, real-time fluoroscopy images are captured on-site, and AI calculates the optimal puncture trajectory. Indeed, thoracic complications and bleeding have been reported as potential risks of renal puncture in the supine position. 26,27
However, as the current study demonstrated that robotic guidance using the bi-plane method could achieve sufficient renal puncture accuracy, it may reduce the complications observed in renal puncture in the supine position. Further, our study indicated that robotic guidance may be more beneficial for residents.
The present study demonstrated that RA puncture had a longer device setup and total procedural time than US puncture. The current ANT-X device adopts a bi-plane method, which calculates the direction in which the needle should be advanced. In this method, the approximate needle insertion points and their directions toward the target were first determined artificially.
Subsequently, two different planes were set by ANT-X using fluoroscopy; each plane consisted of the target and direction of the needle. Finally, the line of intersection where the two planes met was derived as the direction in which the needle should be advanced. These processes involve multiple steps, mainly compensation of needle movement for the other plane, which result in long setup and total procedural times.
However, once targeting is completed, the time required for puncture could be shorter, and a reliable puncture with a high success rate would be possible. Moreover, one could mitigate this time limitation in the actual clinical setting since surgeons could carry on other preparations during the device setup time with the cooperation of the medical engineering staff who prepares the device simultaneously.
In this study, RA puncture had a longer fluoroscopic exposure time than US puncture. The puncture was performed while confirming the needle tip position with a probe during the US puncture, and radiation was only used to confirm the position of the needle tip after puncturing, assessing whether the puncture was successful. In contrast, since the participants were not familiar with fluoroscopy-guided puncture, some fluoroscopy confirmation during the RA puncture was performed in a live, continuous mode, which might be reduced by following the as low as reasonably achievable (ALARA). 28 Adherence to the ALARA protocol reduces the radiation exposure time.
The NASA-TLX is a widely used tool for evaluating workload and is often used in urology research. 16,17 The current study demonstrated lower mental and physical task workloads, effort and frustration scores, and overall TLX scores in the RA puncture group than in the US puncture group. These results indicate that the accuracy and ease of puncture provided by the RA technique could reduce the mental and physical burden of the procedure, despite the fact that RA puncture involves complex technical skills, including multiple steps in device setup and the use of fluoroscopy.
Moreover, the current study demonstrated that RA puncture had a higher single-puncture success rate; however, there was no difference between the two puncture types in NASA-TLX performance evaluation, which reflects the work performance evaluated by the surgeons themselves. In the US puncture group, 44% of the participants could not puncture successfully in a single session. However, all of them succeeded in puncturing the second time; therefore, they probably did not underestimate their work performance.
The present study had some limitations, including the use of a phantom model rather than a clinical setting, a small sample size, and limited assessment only for percutaneous access. Further studies with larger sample sizes and in clinical settings are required to confirm the efficacy and safety of RA puncture and assess its impact on the outcomes of PCNL.
Moreover, a comparison between two different imaging modalities may not reflect the impact of robotic assistance on fluoroscopic guidance. However, we designed this study in accordance with current practice, which provides US guidance for evaluating the usefulness of RA punctures. Further, in this study, we compared RA puncture with the US puncture that we usually perform. However, US-guided puncture and fluoroscopy-guided puncture are fundamentally different modalities.
Ideally, a comparison between RA puncture and freehand fluoroscopy-guided puncture may be more desirable. Finally, workload assessment compared the entire process, including device setup, and could not differentiate between device setup and puncture. Despite these limitations, we believe that this multicenter prospective benchtop study provides credibility in demonstrating the efficacy of supine RA puncture using the bi-plane method.
Conclusion
This prospective multicenter benchtop study found that RA puncture using ANT-X with the bi-plane method, which can also be used for puncture in the supine position, had a higher puncture success rate and lower task workload than US puncture. These findings suggest that RA puncture offers advantages over renal puncture in terms of accuracy and surgical workload.
Footnotes
Acknowledgments
The authors thank the participating doctors at the NCU Hospital and WMU Hospital. NDR Medical Technology provided the ANT-X device and the phantom models used in this study without any further financial support.
Authors' Contributions
T.S.: project design and modification and article writing and editing. K.T.: project conception and design and article reviewing. K.T.: data collection. M.I.: data collection. K.K.: data collection. R.D.: project design and modification. S.Y.: project design and modification. R.U.: data management. S.H.: data management. R.A.: data analysis. A.O.: data analysis. T.Y.: article reviewing and editing.
Ethical Approval
Written informed consent for publication of their details was obtained from the study participant.
Author Disclosure Statement
No competing financial interests exist.
Funding Information
No funding was received for this article.
Supplementary Material
Supplementary Table S1
Abbreviations Used
References
Supplementary Material
Please find the following supplemental material available below.
For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.
For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.
