The utility of individualized interactive 3D virtual models for improving patient and resident education in robotic partial nephrectomy

doi:10.21203/rs.3.rs-4980487/v1

Download PDF

Research Article

The utility of individualized interactive 3D virtual models for improving patient and resident education in robotic partial nephrectomy

https://doi.org/10.21203/rs.3.rs-4980487/v1

This work is licensed under a CC BY 4.0 License

Version 1

posted

You are reading this latest preprint version

Introduction:

Robotic partial nephrectomy (RAPN) requires careful planning due to high anatomical variability between patients presenting challenges during counseling on the surgery, anatomy, and risks. 3D digital modeling provides an opportunity to address this issue. Prior studies predominantly investigate the use of 3D printed models which have limitations including cost. This study seeks to assess the utility of an individualized interactive 3D virtual model for patient counseling and surgical trainee education for RAPN.

Methods:

Patient-specific virtual 3D models were derived from imaging of 32 patients to undergo RAPN at Thomas Jefferson University Hospital between October 2023 and April 2024. Patients filled out surveys pre and post model exposure to assess their understanding of their kidney, pathology, the surgery, and its risks. Surgical trainees filled out pre and post model surveys and post-operative surveys to assess the impact of the model on surgical understanding and patient counseling.

Results:

Patient understanding of their kidney, pathology, and surgery all significantly improved after viewing the model (p < 0.05) Both patients and trainees found the model to be helpful (averaging > 9 out of 10) across all assessed domains of patient counseling and education. The model was not found to be significantly associated with improved trainee confidence in anatomy and surgical planning.

Conclusion:

Imaging

Patient Education

Renal Cancer

Robotics

Robotic partial nephrectomy (RAPN) requires careful preoperative planning due to high anatomical variability between patients. This complexity presents challenges when counseling patients on the risks, benefits, and conduct of their procedure. 3D digital modeling provides an opportunity to enhance surgical planning^1,2, patient understanding³, and resident education.⁴

3D modeling technology has become more advanced and usage of patient-specific cross-sectional images to create individualized models has become feasible.⁵ Past studies have shown the utility of 3D printed models for surgical planning, patient counseling, and trainee education in RAPN.⁶ Due to limited access to 3D printers in standard clinical practice settings, the use of detailed virtual 3D models is an attractive option to reduce time and cost when creating models.

There have been prior investigations into the use of both 3D printed and virtual 3D models. A study by Wake and Rude (2017) showed surgeon confidence was increased by printed model use.⁷ In another study, patients and trainees consistently stated the printed models enhanced understanding of surgical anatomy.⁸

For virtual 3D models, one study showed that models helped improve anatomical clarity and confidence in surgical decisions.¹ In another study, virtual models helped improve intraoperative management.²

In this study, we aim to use detailed interactive virtual models in a greater range and number of patients to better understand the utility of interactive virtual 3D modeling in RAPN education for patients and surgical trainees for both the Da Vinci XI surgical system® or the Da Vinci SP surgical system® (Intuitive Surgical, Sunnyvale, CA).

Study Design and Population

This single-center, prospective study was approved by the Sidney Kimmel Cancer Center and the Thomas Jefferson University Institutional Review Board (IRB No. 2023–2311). Patients scheduled for RAPN were recruited between October 2023 and April 2024. Patients had computer tomography (CT) or magnetic resonance imaging (MRI) within 3 months of their surgery date. Informed consent was obtained prior to enrollment. Minors (< 18 years) and non-English speakers were excluded.

Model Creation

Materialise Mimics Innovation Suite was used for model creation. Segmentation for modeling included the renal cortex, medulla, urinary collecting system, and ureter, as well as the aorta, inferior vena cava, renal arteries, veins (up to tertiary branches), parasitic vasculature, and pathology (including solid lesions and cystic lesions of any Bosniak classification).

Segmentation was performed by a team of trained technicians in the Health Design Lab at Thomas Jefferson University through creation of ‘masks’ of each structure using coronal, sagittal, and axial images. The model was exported into Materialise 3-matic, and after components underwent the smoothing process, consistent colors were assigned to each of the anatomical structures (red for arteries, blue for veins, green for tumors, purple for cysts, yellow for the ureter and urinary collecting system, orange for renal medulla, white for renal cortex). The structures included in these models were often intercalated with one another, so the opacity of overlying structures (such as the medulla and renal cortex) was adjusted to provide a clear visualization of all components in their native anatomic configuration.

Each model was validated by a radiologist before being finalized and uploaded to the Materialise Viewer software. Viewer allows for responsive 3D manipulation of the model on computer, phone, or tablet (Fig. 1).

Data Collection

Model review and data collection occurred on the day of surgery. A physician member of the surgical team counseled the patient on their surgery and disease using conventional imaging. Patients then received a “pre-model” survey consisting of four questions rated on a 10-point visual analog scale to assess their understanding of the kidney, their disease, the surgery being performed, and the risks and complications related to the surgery. Counseling was then repeated with the model and a “post-model” survey was administered using the same questions followed by four additional questions adapted from Bernhard et al.⁹ and Scott et al.⁶ All surveys were administered in person.

Trainees completed a “pre-model” survey of five multiple choice questions adapted from Wake et al.⁷ regarding operative planning (operative technique, operative approach, clamping of vessels) and a sixth question assessing confidence in their plan. The model was viewed, and a “post-model” survey of the same questions was completed. A “postoperative” survey consisted of five questions assessing the extent to which the model aided comprehension of the surgery and education of the patient.

Surgical Technique

All RAPN were performed by one of six attending surgeons using either the Da Vinci XI or SP. Port placement was standardized depending on approach. SP procedures were exclusively retroperitoneoscopic. Arterial clamping was performed using bulldog laparoscopic vascular clamps. After mass excision, a standard two-layer renorrhaphy was completed with a running base closure and a cortical closure with barbed suture and a sliding clip technique.

Outcomes and Statistical Analysis

The primary outcome was patient education, defined as the change in mean survey responses between pre- and post-model surveys. Secondary outcomes included trainee education, defined as positive change in mean confidence in surgical planning between pre- and post-model surveys, and trainee assessment of model utility using postoperative survey scores. Analysis included descriptive statistics to characterize cohorts and report trainee assessment of model utility. Sub-analyses were performed based on surgical platform (XI vs SP) and approach (trans- vs retroperitoneal). Comparison of pre- and post-model survey responses were performed using paired, one-sided t tests, with p < 0.05 considered significant. All statistical analyses were performed using JASP. We also provide qualitative description of instances where trainees changed preoperative planning parameters after viewing the model.

Patients

The studied cohort was comprised of 32 patients with 13 women and 19 men. Median age was 66.5 years old (range of 34 to 78). Average Nephrometry score was 6.7 (4–11). There were 21 transperitoneal cases and 11 retroperitoneoscopic cases. There was one conversion to open surgery but no conversions to radical nephrectomy. Of the completed robotic cases, 6 cases were performed with the Da Vinci SP system (all retroperitoneal) and 25 with the XI system (4 retroperitoneal). Patient characteristics are described in Table 1.

Table 1

Clinical characteristics of the study cohort
	Median (range)
Age (y)	66.5 (34–78)
Sex (N, %)
Female	13 (40.6)
Male	19 (59.4)
Body Mass Index (kg/m²)	31.4 (19.0–50.6)
Race (N, %)
Black or African American	11 (34.4)
Hispanic or Latino	20 (62.5)
White	1 (3.1)
Surgical Approach (N, %)
Robotic	31 (96.9)
XI	25 (78.1)
SP	6 (18.8)
Conversion from Robotic to Open	1 (3.1%)
Nephrometry Score	7 (4–11)
Kidney Side
Right	15 (46.9)
Left	17 (53.1)
Surgical Technique (N, %)
Transperitoneal	21 (65.6)
Retroperitoneal	11 (34.4)

Patient Response to Model

Patients reported better understanding of their anatomy and the proposed surgery across all four domains tested after viewing the virtual model (p < 0.05) (Fig. 2). Out of 10, there was an average 1.7-point increase in understanding of the kidney itself (6.7–8.4), a 1.7-point increase in understanding of their disease (6.7–8.4), a 1.5-point increase in understanding of the proposed surgery (7.4–8.9), and a 0.9-point increase in understanding of the risks and complications related to the surgery (7.5–8.4). Patients found the model helpful in all four domains (9.5, 9.4, 9.8, and 9.4 out of 10).

Trainee Response to Model

Trainee surveys were filled out for 25 patients by the fellow (14), a chief resident (18), and a junior resident (1). No significant difference was observed in trainee confidence in the proposed operative approach after viewing the model (9.2–9.4 out of 10, p = 0.13) (Table 2). When asked how much the virtual model helped with comprehension of anatomy and surgical planning, trainees rated an average 5.5 out of 10 with 5 indicating no effect.

Table 2

Trainee Rating of Confidence in the Proposed Operative Approach Before and After Viewing the Virtual 3D Model
	Overall		XI Robotic Platform		Single Port Robotic Platform
	Before Model	After Model	Before Model	After Model	Before Model	After Model
Mean ± SD	9.17 ± 1.39	9.40 ± 1.07	9.52 ± 0.90	9.57 ± 1.04	7.67 ± 2.07	8.67 ± 1.03
P value	0.064		0.357		0.055
			Transperitoneal Approach		Retroperitoneal Approach
			Before Model	After Model	Before Model	After Model
Mean ± SD			9.65 ± 0.49	9.70 ± 0.73	8.20 ± 2.04	8.80 ± 1.40
P value			0.358		0.056
Abbreviations: SD, standard deviation

Trainees rated the model 9.0, 9.2, and 9.1 out of 10 for explaining the disease, procedure, and surgical risks to the patient respectively. In one case, a trainee changed their initial recommendation from transperitoneal to retroperitoneal approach after viewing the model. Subsequently, retroperitoneal surgery was performed. In three cases, trainees who first indicated they would perform renal artery clamping revised their plan to perform selective arterial clamping (Fig. 1B). In one case, a trainee who recommended selective arterial clamping after viewing the imaging instead recommended renal artery clamping after viewing the model. In one case, a trainee who first recommended renal artery clamping indicated that they would perform renal artery and renal vein clamping after viewing the model, although vein clamping was not performed ultimately.

For XI cases, there was no significant difference observed in trainee confidence in the proposed operative approach before and after viewing the model (9.5–9.6 out of 10, p = 0.36). Trainee rating of how much the model helped with comprehension of the anatomy and surgical planning averaged 5.4 out of 10. For SP cases, there was no significant difference observed in trainee confidence in the proposed operative approach after viewing the model (7.7–8.7 out of 10, p = 0.06). Trainee rating of how much the model helped with comprehension of the anatomy and surgical planning averaged 6.2 out of 10.

For transperitoneal cases, there was no significant difference in trainee confidence in the proposed operative approach after viewing the model (9.7–9.7 out of 10, p = 0.36). Trainee rating of how much the model helped with comprehension of the anatomy and surgical planning averaged 5.2 out of 10. For retroperitoneoscopic cases, there was no significant difference observed in trainee confidence in the proposed operative approach after viewing the model (8.2–8.8 out of 10, p = 0.06). Trainee rating of how much the model helped with comprehension of the anatomy and surgical planning averaged 6.0 out of 10.

3D renal modeling has been shown to be a useful adjunct to surgical planning and education. 3D printed models are frequently studied but 3D printing comes with challenges including cost and difficulty representing complex intrarenal anatomy. Interactive virtual 3D modeling reduces cost compared to 3D modeling while rendering high complexity. Previous investigations on the use of 3D modeling for minimally invasive partial nephrectomy have focused on perioperative outcomes and surgical planning.^10,11 There is less evidence regarding the use of virtual 3D models for trainee and patient education.

This study showed that use of a cost-effective and detailed virtual 3D model was associated with significant improvements in patients’ understanding of the kidney itself, their disease, the procedure, and the associated risks and benefits. This demonstrates the utility of a virtual interactive model to accomplish what has been previously shown with 3D printed models.⁶ The model was perceived as helpful (> 9 out of 10) by both patients and providers across all assessed domains of counseling. These findings indicate that the model likely improved health literacy. As has been discussed in a prior similar study on 3D printed models, higher health literacy has been shown to correlate with better patient outcomes.^6,12

Although a previous study found that 3D printed model improved trainee confidence in the planned surgical approach, this was not observed in this study.⁶ Surveys were typically filled out by senior trainees who were well-prepared for the cases and had collaboratively reviewed the available cross-sectional imaging. Senior trainees may find review of the conventional imaging to be sufficient to establish confidence in the proposed surgical approach and specific patient anatomy.

We felt that the model was particularly helpful for retroperitoneoscopic single-port cases. Especially in the single port cases, where renal manipulation and global anatomic perspective are more challenging to achieve, the model was felt to aid in operative confidence. While the results were not statistically significant, there was a trend toward significance in improvement of trainee confidence in the proposed operative approach in retroperitoneoscopic cases regardless of platform and for single port cases specifically when compared to transperitoneal and XI cases respectively.

To our knowledge, this is the first 3D modeling study that has involved a significant proportion of SP retroperitoneoscopic partial nephrectomy. Due to the heterogeneity that this introduced in operative time, mass complexity, and surgical approach, we did not feel that an analysis of perioperative metrics or comparison to a historical control cohort, as has been performed in similar studies, would provide a comparable control. We acknowledge that our institution was early in our experience with the SP platform. Nevertheless, further studies specifically investigating 3D model usage in single port robotic surgery can be considered.

The protocol used in this most recent study allows for cost improvement. A 3D printed model technique was previously investigated at our institution to create cost-effective models at a material cost of $30–50 per model.⁶ This study did not consider additional 3D printing-related labor costs incurred for computer-aided design and print processing/assembly (approximately $200 and $40 respectively), bringing the true cost to produce a physical 3D printed model to about $270–290 per model based on our previously published protocol. Additionally, the previous study protocol included the segmentation of four discrete structures in each case (renal parenchyma, renal artery, renal vein, and tumor/pathology). This study’s protocol aimed for a higher level of model specificity including segmentation of seven structures (renal parenchyma, renal artery, renal vein, tumor/pathology, medulla, collecting system, ureter) and additional fine structures, as relevant and differentiable (ex. smaller arteries/veins supplying the pathology) without needing to consider printability of these fine structures. This increased detail likely translated to an increase in segmentation time (5 hours in this study vs 2 hours in the prior study) although the segmentation staff were different and the studies were not designed to correlate complexity and segmentation time. In the current protocol, eliminating the physical model may preserve much of the benefit, while reducing the cost and complexity of execution. By eliminating the design, 3D printing, and assembly of a physical model, we gained a time benefit and could create models for cases closer to the scheduled surgery.

This study has other important limitations. Creation of virtual models, although cost-effective compared to printed models, incurs costs including labor to perform segmentation. While model development algorithms have not yet met the necessary precision threshold, this process could be automated in the future. Additionally, all trainees completing surveys were staff of the institution administering the study, which may have created response bias. Since the model was shown the day of surgery, the counseling with conventional imaging and then the model was iterative, which may have improved post-model survey results, in part, from repetition.

This study shows that an interactive 3D virtual model is a useful adjunct to patient education prior to RAPN from both the patient and provider experience and can serve as a less resource-intensive alternative to 3D printed modeling for this application. 3D modeling may be particularly helpful to the surgeon in cases where excellent surgical exposure is more difficult to obtain, such as with retroperitoneoscopic single port robotic surgery for which it may aid in helping the surgeon or trainee better conceptualize their operative approach. Virtual interactive 3D modeling is a cheap and straightforward intervention that should be considered as part of standard patient surgical counseling workflow for small renal masses.

RAPN Robotic partial nephrectomy

CT Computer tomography

MRI Magnetic resonance imaging

Funding: No funds, grants, or other support was received.

Declaration of Competing Interests: The authors have no relevant financial or non-financial interests to disclose.

Author Contributions: ARH (Methodology; Data Collection; Data curation; Formal Analysis; Writing – Original Draft; Writing – Review and Editing). RR (Methodology; Data curation; Formal Analysis; Writing – Original Draft; Writing – Review and Editing). RM (Methodology; Writing – Original Draft; Writing – Review and Editing). SB (Methodology; Writing – Original Draft; Writing – Review and Editing). IBL (Data Collection; Data Curation; Writing – Review and Editing). BHI (Data Collection; Data Curation; Writing – Review and Editing). MD (Data Collection; Data Curation; Writing – Review and Editing). RP (Conceptualization, Methodology; Writing – Review and Editing; Supervision). CDL (Conceptualization, Methodology; Writing – Review and Editing; Supervision).

Ethics Approval: This study was performed in line with the principles of the Declaration of Helsinki. Approval was granted by the Sidney Kimmel Cancer Center and the Institutional Review Board of Thomas Jefferson University Hospital (10/19/2023/IRB No. 2023-2311).

Consent to Participate: Informed consent was obtained from all individual participants included in the study.

Hyde ER, Berger LU, Ramachandran N, et al. Interactive virtual 3D models of renal cancer patient anatomies alter partial nephrectomy surgical planning decisions and increase surgeon confidence compared to volume-rendered images. Int J Comput Assist Radiol Surg. 2019;14(4):723-732. doi:10.1007/s11548-019-01913-5
Bianchi L, Barbaresi U, Cercenelli L, et al. The Impact of 3D Digital Reconstruction on the Surgical Planning of Partial Nephrectomy: A Case-control Study. Still Time for a Novel Surgical Trend? Clin Genitourin Cancer. 2020;18(6):e669-e678. doi:10.1016/j.clgc.2020.03.016
Pandrangi VC, Gaston B, Appelbaum NP, Albuquerque FC, Levy MM, Larson RA. The Application of Virtual Reality in Patient Education. Ann Vasc Surg. 2019;59:184-189. doi:10.1016/j.avsg.2019.01.015
Murgitroyd E, Madurska M, Gonzalez J, Watson A. 3D digital anatomy modelling – Practical or pretty? The Surgeon. 2015;13(3):177-180. doi:10.1016/j.surge.2014.10.007
Rengier F, Mehndiratta A, von Tengg-Kobligk H, et al. 3D printing based on imaging data: review of medical applications. Int J Comput Assist Radiol Surg. 2010;5(4):335-341. doi:10.1007/s11548-010-0476-x
Scott ER, Singh A, Quinn AM, et al. The use of individualized 3D-printed models on trainee and patient education, and surgical planning for robotic partial nephrectomies. J Robot Surg. 2022;17(2):465-472. doi:10.1007/s11701-022-01441-6
Wake N, Rude T, Kang SK, et al. 3D printed renal cancer models derived from MRI data: application in pre-surgical planning. Abdominal Radiology. 2017;42(5):1501-1509. doi:10.1007/s00261-016-1022-2
Silberstein JL, Maddox MM, Dorsey P, Feibus A, Thomas R, Lee BR. Physical Models of Renal Malignancies Using Standard Cross-sectional Imaging and 3-Dimensional Printers: A Pilot Study. Urology. 2014;84(2):268-273. doi:10.1016/j.urology.2014.03.042
Bernhard JC, Isotani S, Matsugasumi T, et al. Personalized 3D printed model of kidney and tumor anatomy: a useful tool for patient education. World J Urol. 2016;34(3):337-345. doi:10.1007/s00345-015-1632-2
Amparore D, Piramide F, De Cillis S, et al. Robotic partial nephrectomy in 3D virtual reconstructions era: is the paradigm changed? World J Urol. 2022;40(3):659-670. doi:10.1007/s00345-022-03964-x
Piramide F, Kowalewski KF, Cacciamani G, et al. Three-dimensional Model–assisted Minimally Invasive Partial Nephrectomy: A Systematic Review with Meta-analysis of Comparative Studies. Eur Urol Oncol. 2022;5(6):640-650. doi:10.1016/j.euo.2022.09.003
Baker DW, Parker RM, Williams M V., Clark WS. Health literacy and the risk of hospital admission. J Gen Intern Med. 1998;13(12):791-798. doi:10.1046/j.1525-1497.1998.00242.x

No competing interests reported.

Download PDF

Version 1

posted

You are reading this latest preprint version

The utility of individualized interactive 3D virtual models for improving patient and resident education in robotic partial nephrectomy

Status:

Version 1

Abstract

Introduction:

Methods:

Results:

Conclusion:

Figures

Introduction

Methods

Study Design and Population

Model Creation

Data Collection

Surgical Technique

Outcomes and Statistical Analysis

Results

Discussion

Conclusion

Abbreviations

Declarations

References

Additional Declarations

Status:

Version 1