Teaching and evolution in surgical planning of robotic partial nephrectomy with 3D printing models

Abstract

This study aims to analyze the contribution of 3D printing to the teaching and development of robotic partial nephrectomy (RNP) for renal tumors. Evaluation of results in planning the technique, which may vary depending on the professional performing it. A prospective study involving 32 patients who were susceptible to undergoing RPN for renal tumor. Surgical planning (SP) by senior, junior, and resident surgeons was analyzed based on three variables: (1) Tomography (2D), (2) Three-dimensional (3D) tomography, and (3) With a 3D printed model. Three consecutive pre-surgery SP questionnaires were administered. A usability test was completed after surgery. The cohort has a high R.E.N.A.L. score (41% intermediate and 21% high). The use of the 3D printed model changes the SP relative to the CT image by 25.60% and 20.11% with the 3D CT image. The SP factor that varies the most with 3D printing is that related to vascular anatomy (43.84% - 56.82%), followed by assessment of tumor depth (37.67% - 43.18%). Senior urologists modified the vascular hilum passage by up to 63.16%, compared to less than 12.5% by junior urologists. Feelings of security (55.3%), stress (4%), assessment of vascular anatomy (58.9%), tumor depth (36.45%), and surgical time (30.4%) improved. 3D printing influences planning regardless of the professional level, improves safety, and decreases surgical time. Senior surgeons significantly modify the SP more when assessing vascular anatomy by proposing more advanced surgeries.

Supplementary Information

The online version contains supplementary material available at 10.1007/s11701-025-03007-8.

Introduction

The incidence of renal cancer (European eighth highest incidence tumor) increases 2–3% per year [1]. Surgery for renal tumors aims to achieve oncological control while preserving as much nephron mass as possible. In this context, robotic partial nephrectomy (RPN) has been increasingly promoted and has become the preferred option for selected T1 and T2 renal tumors. The advanced technical features of robotic assistance, enables us to progressively manage more complex cases. The teaching of this technique remains challenging for surgeons, as traditional surgical planning (SP) is predominantly based on two-dimensional images from computed tomography (CT) scans or magnetic resonance imaging (MRI). Considering the kidney’s complex anatomy, the integration of three-dimensional (3D) technology allows a more precise anatomical comprehension, offering substantial educational value for both experienced surgeons and those in training.

Studies conducted at academic centers have shown a reduction in surgical variability among professionals when using 3D printed models compared to relying on CT imaging models alone [2].

This tool enhances surgical planning, minimizes intraoperative decision-making, shortens operative time, and enables precise management of complex cases through guided surgery techniques applied during the preoperative phase [3]– [4]. 3D printed models offer new learning opportunities for trainees and serve as valuable tools for surgeons rehearsing highly complex or novel procedures [5].

The benefit of using these models increases with small and intra-renal tumors and for residents with less knowledge and experience [6].

The scoring system, called the R.E.N.A.L. score (RS), is used at our center to assess the most surgically relevant anatomical features of renal tumors [7].

This study aims to analyze the contribution of 3D printing to surgical education and the evolution of RPN for renal tumors, as well as to assess its influence on surgical planning across different levels of professional experience (senior, junior, and resident surgeons).

Materials and methods

A prospective cohort of 32 patients with indication for RPN who underwent CT imaging, a 3D PDF reconstruction of their scan, and had a 3D printed kidney model, were analyzed in this study. The medical variables of the cases were examined, including demographic data such as age, sex, and body mass index (BMI), as well as the case complexity according to the R.E.N.A.L. score.

On the other hand, prior to surgery, the SP, surgical approach, and tumor characterization, were evaluated through questionnaires completed by 3 senior urologists (> 15years of experience), 1 junior urologist (< 5years of experience), and 4 resident surgeons who participated as assistants. Ease-of-use questionnaires were administered after each surgery.

Acquisition of images and 3D models

The process of obtaining renal anatomical models for surgical planning began with a contrast (Iomeron 400) CT scan with multidetector and < 2 mm slice thickness, in arterial phase (ROI located in the abdominal aorta) and venous phase (70 s after injection). In some patients of this study, an excretory phase was also performed 10 min after injection in order to be able to segment the excretory pathway.

These DICOM images were subjected to a segmentation process (average time 30–35 min), using the Syngovia Frontier 3D Printing software (Siemens Healthineers). The kidney, lesion, renal arteries and veins, and the excretory pathway were segmented separately using different tools within the anatomical viewer of the software, such as region growth, hand-drawn object definition, or the Hounsfield thresholding tool. Subsequently, the segmented files were post-processed with Autodesk Meshmixer^® software (version 3.5.474, 2017 Autodesk, Inc., San Rafael, California, USA) in STL format. In addition to standard procedures for cleaning and adapting meshes, this pre-processing reduced the triangles comprising the kidney mesh to generate a pattern and create a “wire mesh” structure. This structure enabled the visualization of internal elements like blood vessels or the urinary tract. The use of PolyJet technology (Objet260 Connex, Stratasys, Germany), which deposits microdroplets of UV-curable liquid photopolymer, enabled the assignment of different colors and shore hardnesses to distinguish structures: kidney — soft transparent material mixed with pink; vein — blue rigid material; artery — pink rigid material; urinary tract — soft transparent material mixed with blue; tumor — solid soft material (Agilus30 FLX935 (3.6 kg), VeroMagenta RGD851 (3.6 kg), VeroCyan RGD843 (3.6 kg), Support SUP705 (3.6 kg)). These final anatomical models were printed on Biobizkaia Health Research Institute, 3D Printing and Bioprinting Laboratory.

Evaluation questionnaires

A surgical planning survey (S), which consisted of 14 questions that covered everything from planning and approach to tumor characterization was designed (Table 1). This questionnaire was progressively administered to surgeons based on the following criteria: S1: Urologists considered only CT images; S2: in addition to CT images, it included the 3D image in PDF format and S3: CT images, the 3D image in PDF, and the printed 3D model. From a total of 146 questionnaires, the average number of different answers and the average percentage of differing answers between S1-S2, S1-S3, and S2-S3 were analyzed. The number and percentage of differing responses between seniors, junior and residents were also compared using the Wilcoxon test. After the RPN, a ease-of-use questionnaire (S4) of the 3D printing model, consisting of 10 questions, was conducted (Fig. 1).

Table 1.

Surgical planning survey (S). S1: survey with TC-2D; S2: survey with TC-3D; S3: survey with printing 3D; Q: question; ICG: indocyanine green

Surgical planning survey (S1,S2,S3)
Q1 Placement of robotic trocars
Lateral displacement of trocar ports					Cranial displacement of trocar ports			Standard						Other
Q2 Dissection of lower pole of the kidney and the ureter
Yes								No			Other
Q3 Mobilization of the upper pole of the kidney
Yes					No			Other
Q4.1 Renal hilum dissection
From ureter to hilum					From upper pole to hilum			Directly to hilum						Other
Q4.2 Order of renal hilum dissection
Artery, vein	Multiple artery, vein				Artery, multiple vein			Multiple artery and vein		Vein, artery	Multiple vein, artery				Vein, multiple artery				Multiple vein and artery
Q5 Tumor delimitation
Complete dissection of anterior and posterior perirrenal fat					Complete dissection of cranial or distal pole				Dissection less than a pole				Other
Q6 Clamping
Artery				Selective artery					Entire hilum					Other
Q7.1 Tumor extensión
2 cm		3 cm		4 cm			5 cm		6 cm		7 cm			> 7 cm				Other
Q7.2 Tumor extension depth
1 cm			2 cm			3 cm			4 cm			> 4 cm					Other
Q7.3 Relationship between tumor and renal hilum
Artery			Vein			Artery and vein			No relationship					Other
Q7.4 High risk of injury urinary tract
Yes				No					Other
Q7.5 Tumor excision technique
Enucleation			Enucleoresection			Polar nephrectomy			Wedge partial nephrectomy							Other
Q8 Need for intraoperative ultrasound
Yes				No					Other
Q9 Need for Indocyanine green
Yes				No					Other

Fig. 1.

Workflow of questionnaires before and after using 3D printed model

Results

Patient cohort

The mean age of the 32 patients was 65 (59.5–70), with an average BMI of 28.5 (24.4–33.9). 79% were male. The tumor’s laterality was similar, with 45% on the left and 55% on the right. The R.E.N.A.L. score indicated that 38% of the patients had a low score (n = 11), 41% an intermediate score (n = 12), and 21% a high score (n = 6). Two preselected patients were excluded from the study because NPR was considered too complex for 3D segmentation assessment.

Biomodels

The processing of patient-specific models prior to 3D printing took an average of 95 ± 10 min, covering both segmentation and mesh processing from DICOM files. The average cost of manufacturing the 3D kidney models, printed with PolyJet technology, was €380.43 ± 96.36, which included the depreciation of the printing equipment. Regarding post-processing of the sample, which included support removal, cleaning, validation, labelling, and shipping, the time was approximately 52 ± 5 min. All printed models were validated dimensionally through comparison with the STL model with manual measurements using calipers. The 3D printed model showed high fidelity to its digital version, with an average deviation of less than 1% along all three spatial axes. During the study, the 3D model was selected in mesh form and printed with a ductile material that allowed for easier handling.

Variability in pre-surgical assessments

When performing pre-surgical assessment using different imaging modalities, the variability in surgical planning reached up to 25% when comparing conventional (2D) CT images with 3D printed models.

A comprehensive analysis was conducted focusing on those surgical planning questions that showed the most significant variation (more than 20%) when comparing 2D, 3D PDF and 3D printed models in surgical strategy decision-making. Figure 2 shows the 2D CT image, 3D PDFs and 3D printed models of two different cases from this study.

Fig. 2.

A The use of a 3D model changes SP improving surgery resolution and surgical time. 20% of surgeons that used 3D models changed the planning to “no clamping” (under option “others”); 18% changed it when using 3PDF images and no one with CT Scans. B The 3D model, with mesh material, was useful in this case with multiples tumors, allowing the determination of the depth of the malignant tumor and the absence of urinary tract involvement. The answer “ high risk of injury urinary tract” is changed to “no” only with 3D model (S3)

The questions (Q) exhibiting the greatest discrepancies were studied to determine whether these differences were influenced by the level of surgical experience (senior, junior, and resident urologists) (Appendix 1). Three questions showed statistically significant differences. Firstly, in the assessment of the renal hilum (Q4.2), senior urologists showed marked variation when using 3D PDF (61.36%) and 3D printed models (63.16%) compared to junior urologists, who almost did not change their preoperative planning (p = 0.005). Conversely, questions related to tumor excision extension (Q7.1) and potential urinary tract opening (Q7.4) led the juniors to change their planning by more than 50%, while senior surgeons showed significantly lower rates of change (p = 0.014 and p = 0.016, respectively) when exposed to 3D technologies (Table 2).

Table 2.

Comparison among questionnaires S1, S2, S3. A complete questionnaire: number and percentage of changed answers. B main questions with changed answers by level of expertise. R: residents; S: seniors; J: juniors. P: p-value. Q4.2: order of renal hilum dissection; Q5: tumor delimitation; Q7.1: tumor extension; Q7.2: tumor extension depth; Q7.4: high risk of injury urinary tract

	S1 vs. S2					S1 vs. S3					S2 vs. S3
A. Complete questionnaire
n changed answers	mean (SD)
	2.73 (2.23)					3.59 (2.14)					2.82 (1.73)
% changed answers	mean (SD)
	19.52 (15.93)					25.60 (15.31)					20.11 (12.35)

B. Main questions with Changed answers % (n)
Professional	Global	R	S	J	p	Global	R	S	J	p	Global	R	S	J	p
Q4.2	43.84 (64)	39.36 (37)	61.36 (27)	0 (0)	0.001	56.82 (75)	58.62 (51)	63.16 (24)	0 (0)	0.005	43.08 (56)	44.71 (38)	45.95 (17)	12.5 (1)	0.23
Q5	30.14 (44)	25.53 (24)	40.91 (18)	25 (2)	0.19	36.36 (48)	32.18 (28)	42.11 (16)	57.14 (4)	0.26	21.54 (28)	16.47 (14)	29.73 (11)	37.5 (3)	0.13
Q7.1	22.60 (33)	24.47 (23)	20.45 (9)	12.5 (1)	0.74	35.61 (47)	41.38 (36)	18.42 (7)	57.14 (4)	0.014	30.00 (39)	30.59 (26)	27.03 (10)	37.5 (3)	0.82
Q7.2	37.67 (55)	37.23 (35)	36.36 (16)	50 (4)	0.77	43.18 (57)	43.68 (38)	42.11 (16)	42.86 (3)	1.0	36.92 (48)	34.12 (29)	48.65 (18)	12.5 (1)	0.10
Q7.4	19.86 (29)	13.83 (13)	27.27 (12)	50 (4)	0.016	20.45 (27)	21.84 (19)	18.42 (7)	14.29 (1)	0.83	20.00 (26)	20 (17)	16.22 (6)	37.5 (3)	0.39

Surgeons’ views on the functionality and usability of 3D models

57 ease-of-use questionnaires (S4) were completed by different surgeons (65,21% seniors and junior, and 34.78% resident surgeons) and data were assessed in global. 91.1% reported using the 3D printed model also during surgery. The 3D printed model increased surgeon’s sense of security by 55.3%, leading to a noticeable reduction in surgical stress. There was a notable improvement in the assessment of vascular anatomy, with 58.9% of participants strongly agreeing that the 3D models enhanced their understanding. Additionally, 36.45% stated achieving a better assessment of the tumor depth. Surgical time decreased moderately by 30.4% (Appendix II).

Discussion

The application of 3D technologies in RPN is the method with the highest number of publications in urology. The number of cases increases every year. In our series of 32 models, we observed a significant variation in surgical planning when using 3D printed models compared with 2D images. Dissecting the hilum, releasing the vessels in the correct order, and assessing the number of arteries or veins were the most influenced surgical steps when using 3D imaging, and particularly 3D printed models. This improvement was due to a better spatial understanding of the anatomy. This step may represent the most critical factor in ensuring surgical safety, especially during tumor excision. Moreover, it represents a significant technical evolution, encompassing procedures without clamping, with selective clamping, and with full clamping. At this stage, senior urologists often adjusted their surgical planning when using 3D printed models (63% versus < 12.5%), while junior surgeons generally maintained their initial approach.

Another frequent modification of the surgical plan involved the reassessment of tumor size, depth, and surface extent, constituting the second major surgical decision point. In our series, the use of 3D printed models notably enhanced assessment at this stage, particularly prompting junior urologists to adapt their approach compared with senior urologists (50% against 27.22%).

The characterization of the physical location and dimensions of the tumor in relation to healthy parenchyma, intrarenal vascularization, and the collecting system improved with 3D printed models. This means that complex surgery can be performed by less skilled surgeons using 3D technology rather than 2D images. In our case, surgeons had comprehensive imaging options to observe changes in surgical planning. It is notable that senior urologists altered the overall surgical strategy more extensively, likely due to a greater ability to perform more complex procedures, such as surgery without clamping or with selective clamping. Resident surgeons varied slightly less than the seniors, indicating good knowledge of the adjustable aspects of surgery using the 3D model. In our scenario, the residents act as assistants and do not operate at the console, but they adapt to the changes according to the senior urologist’s considerations. As described in the literature, there was a reduction in surgical variability among urology staff when using 3D printed models compared with relying on CT 2D imaging models [2]. A limitation of the statistical analysis was that only one junior urologist participated, which may have introduced bias and confounded the comparative results. Furthermore, the use of multiple comparison correction factors may become necessary with a larger sample size.

Although partial nephrectomy is increasingly indicated, with a growing tendency to perform it even in T2 stage tumors, its recommendation in cases with R.E.N.A.L. scores of 7 or higher [8] remains a matter of debate. 3D technology enabled us to expand indications in borderline cases, allowing for safer decisions in high-complexity cases (62% of our series) selected for RPN [6]. Conversely, the model sometimes led to deciding against partial surgery.

The findings of this study were consistent with previous publications [3]–[4], which demonstrate that 3D printing of renal models enhances SP, minimizes intraoperative decision-making, shortens operative time, and enables precise management of complex cases through guided surgery techniques during the preoperative phase. Furthermore, we agreed with the authors [6, 9] in recognizing that 3D printed medical models provide valuable opportunities for medical education, benefiting both trainees and experienced surgeons preparing for highly complex or unprecedented procedures.

Even resident surgeons were able to perform surgery or SP with high R.E.N.A.L. score, 21% of cases in our cohort, using a 3D printed model [9].

All this contributed to improve urological surgeons’ confidence and therefore patients’ too, which will undoubtedly improve both oncological and functional surgical results in the future.

The physical 3D models also improve urologists’ ability to communicate with patients, allowing them to explain not only the tumor characteristics, but also the planned surgical approach and the various available techniques that exist [10]–[11]. In this study, to preserve objectivity in responses across the different imaging modalities, it was only assessed on the day before the intervention. Consequently, 3D printing was not employed to explain the procedure to patients, unlike other published studies that have proposed this application. Notably, previous reports suggest that the effectiveness of 3D printed models in enhancing patient understanding has been inconsistent [12].

Today, it remains a technology that requires the radiologist’s time and involves a lengthy printing process, although the cost of printing is decreasing (2019: 1000 dollars compared to 2025: less than 500 euros in our field). Variability in cost is also influenced by the choice of materials and the complexity of the anatomical structures involved.

Selecting an appropriate, lightweight 3D model constructed with mesh material allows for an excellent internal evaluation of the vessels and urinary tract. However, the sample size of this study (32 patients) was relatively small, with a limited number of surgeons in each group. Therefore, further studies are needed to fully elucidate the potential of 3D technologies compared to 2D imaging not only in surgical planning, but also in clinical and oncological outcomes13.

Conclusions

Regardless of the surgeon’s experience, all professionals agreed that this 3D technology enhances confidence in surgical planning, reduces surgical time, and even alleviates surgical stress. It is utilized both in planning and during the procedure.

The assessment of vascular anatomy, as well as the in-depth evaluation for seniors and resident surgeons, was more effective with the 3D printed model. This enables us to propose more complex surgeries in cases of intermediate and high R.E.N.A.L scores, as well as more advanced surgeries such as selective clamping and non-clamping surgery. On the other hand, it also helps us to dismiss cases for partial nephrectomy due to their high risk.

Therefore, the 3D printing model aids in the training of residents, the technical evolution of seniors, and the future technical development of juniors.

Supplementary Information

Below is the link to the electronic supplementary material.

Author contributions

AL: Protocol/project development, data collection, data analysis, manuscript writing/editing. CDA, RIG, IAS, MU and MVB: Protocol/project development. DD, MS, IL, JG and CZ: Protocol/project development and data collection. IAM: Manuscript writing/editing.

Funding

The translation and publication have been funded by the IIS Biobizkaia, with funds from the Research Commission of the OSI BILBAO BASURTO.

Data availability

The data supporting the findings of this study are available from the corresponding author, [A.L.], upon reasonable request.

Declaration

Competing interests

The authors declare no competing interests.

Ethical approval

The study has obtained the approval of the Basque Ethics Committee (PS2024024). In accordance with an institutional guidelines. All men signed an institutional review board-approved, protocol-specific informed consent form before study entry. The study was performed in accordance with the ethical standards established in the 1964 Declaration of Helsinki and its later amendments or comparable ethical standards. Anonymized data was used.