Enter your email address below and we will send you your username
If the address matches an existing account you will receive an email with instructions to retrieve your username
You are prohibited from using or uploading content you accessed through this website into external applications, bots, software, or websites, including those using artificial intelligence technologies and infrastructure, including deep learning, machine learning and large language models and generative AI.
Upcoming Site Maintenance on Tuesday, May 28, 2024: Please note that some site functionality such as new user registrations, updates to user accounts, and article purchases will be unavailable during maintenance on this date.
Preoperative Immersive Virtual Reality Applied to Percutaneous Nephrolithotomy: A Prospective Randomized Clinical Study of Surgical Planning and Clinical Outcomes
Percutaneous nephrolithotomy (PCNL) is the procedure of choice for the management
of complex or large renal stones. A major challenge for the surgeon, however, is the
need to assimilate the nearly 2000 static images from a CT scan into a functional
mental image to enable surgical planning. Accordingly, we investigated the potential
of immersive virtual reality (iVR) to enhance surgical planning and its impact on
the outcomes among patients undergoing PCNL.
Materials and Methods:
Between 2019 and 2023, 175 patients undergoing PCNL were preoperatively randomized
into a CT-only group (N = 89) or a CT+iVR group (N = 86). CT scans were rendered into
iVR models that allowed the surgeon not only to visualize and manipulate each patient’s
relevant anatomy, but also to simulate the percutaneous approach to the proposed calyx.
Postoperative CT scans were defined as absolute stone free,
< 2 mm remnants, or 2.1 to 4 mm remnants.
Results:
Preoperative visualization of the iVR model resulted in a changed calyx of entry in
30% of cases. The CT+iVR group had a significant improvement in absolute stone-free
rate (33.70% vs 20.22%, P = .043) and overall < 4 mm remnant rate (62.79% vs 48.20%, P = .044). Clavien-Dindo II+IIIa complications were less in the iVR group (3.48% vs
12.30%, P = .03). The results were independent of the surgeon’s years of PCNL experience.
Conclusions:
Preoperative iVR model visualization benefited surgeons and patients alike. From a
surgical standpoint, viewing the iVR model resulted in a safer, more effective percutaneous
stone removal procedure.
JU Insight
Study Need and Importance
Unlike sports, surgeons have no coaches. After residency, improvement is based solely
on self-assessment and practice. The question arises of whether advances in imaging
can provide surgeons with a preoperative opportunity to better “see” and understand
the anatomy and thereby improve the surgical outcome. Accordingly, we sought to evaluate
immersive and interactive virtual reality (iVR) renderings of a urolithiasis patient’s
abdominal CT to see if viewing these images preoperatively with a 3D headset would
be of benefit (Figure).
What We Found
In this randomized study of 175 patients undergoing percutaneous stone removal, among
4 endourology fellowship-trained urologists, the brief, immediate preoperative viewing
of the patients’ iVR models resulted in an alteration in the selected calyx of entry
(30% of the time), an improved absolute stone-free rate, and a reduction in Clavien-Dindo
II and IIIa complications. Of note, this more sanguine outcome was recorded by each
individual surgeon regardless of their time in practice, which ranged from less than
10 years to 40 years.
Limitations
At this time, the biggest limitation for preoperative iVR is creating the models as
this is labor intensive, each requiring 1 to 3 hours.
Interpretation for Patient Care
With advances in artificial intelligence and computing, the time is near when any
CT scan may automatically be converted into iVR. For the patient, this translates
into a better understanding of their surgery and may result in a better outcome. Indeed,
some of the benefits we have seen for percutaneous stone removal have also been noted
in a pilot study in which preoperative viewing of patient-specific iVR models was
done before donor nephrectomy and laparoscopic/robotic partial nephrectomy. The next
question is whether in the future each surgeon have their own artificial intelligence
personal surgical coach providing a warm-up iVR opportunity and possibly making suggestions,
in and out of the operating theater, where improvements could be made.
Figure. Immersive virtual reality model. A, Overview of the 3D CT-based immersive virtual
reality model of the kidney parenchyma, collecting system, blood vessels, and skeletal
structures. B, Immersive virtual reality model manipulation is depicted; here the
renal collecting system is distinguished from the now ghosted renal parenchyma because
of prior contact of the user’s virtual hand with the renal parenchyma. C, The surgeon’s
virtual hand has pulled the parenchyma away from the collecting system. Red arrows
indicate the interpolar and lower pole stone burden. D, The model has been rotated
such that the access to the stone (yellow) is visualized through the surgeon’s percutaneous
calyx of choice (blue arrow). The surgeon can “peer” into the calyx and travel down
the infundibulum to see if it is truly in line with the stone from that particular
approach.
Over the past 50 years, percutaneous stone removal (PCNL) has evolved as the predominant
procedure for the treatment of large renal calculi.1,2 In this regard, a comprehensive understanding of the renal collecting system and
its relation to the stone as well as to the surrounding viscera is crucial for a successful,
safe procedure.3-7 The advent of CT scanning has greatly enhanced the urologist’s ability to understand
the renal anatomy and to select the calyx that would provide optimal access to the
calculus.8 Currently, a significant challenge lies in assimilating the nearly 2000 static images
of the 2-dimensional noncontrast CT (NCCT) scans into an accurate, mental 3D image.
Immersive interactive virtual reality (iVR) may act as a vital bridge between conventional
and innovative imaging modalities. iVR transforms the static, 2-dimensional NCCT depiction
of renal anatomy into a dynamic, 3D, interactive encounter.5-7 Our initial pilot studies with iVR, when retrospectively compared with an earlier
non-iVR PCNL cohort, revealed that after viewing the iVR model, surgeons were better
able to select the best calyx for entry and achieved better results.6 We report a prospective randomized clinical study designed to test the potential
impact of preoperative viewing of an iVR model on PCNL efficacy, effectiveness, and
safety, as well as its impact on the patient’s understanding and comfort with the
planned surgery.
METHODS
Study Design
To assess the impact of iVR, we created a 2-group, parallel, prospective, randomized
single-center clinical study. The study protocol was approved by our institutional
review board and was performed in accordance with the CONSORT (Consolidated Standards
of Reporting Trials) statement. All patients provided written informed consent before
enrollment. Data collection and follow-up were performed in accordance with the Health
Insurance Portability and Accountability Act. Participants were recruited between
April 2019 and May 2023; data were analyzed in August 2023.
Patient Enrollment and Randomization
Patients presenting with a non–lower pole stone > 2 cm or a lower pole renal stone
> 1 cm who were scheduled to undergo a PCNL procedure by any one of 4 University of
California, Irvine, endourologists were eligible for study inclusion.
After patient enrollment, permuted block randomization was performed.9 Consented patients understood that they would be randomized into either the CT+iVR
group or the CT-only group. Within blocks of different sizes (ie, 4, 6, or 8 patients),
a 1:1 randomization scheme was used to ensure treatment allocation balance. The randomization
and allocation concealment were performed by an investigator who was not involved
in patient enrollment. Moreover, 2 separate databases were created: a patient database
where the basic demographic information was kept, and a randomization database, which
held the data on which patient was allotted to which group. The randomization database
was stored in a password-protected Excel file to which the primary investigator did
not have access. By following these steps, the 2 criteria for adequate allocation
concealment described by Vickers10 were ensured: (1) researches were not able to predict which group a patient will
be randomized to, and (2) researchers were unable to change a patient’s allocation
after they were randomized.
Regardless of randomization, as part of the standard stone protocol at our institution,
CT imaging was obtained. No patients were required to undergo NCCT scans outside of
the standard of care either for the creation of the iVR models or for follow-up. It
is a common practice at our institution to gain percutaneous access with ureteroscopic
guidance. The CT or the iVR preoperative experience enabled the urologist to select
the most direct calyx to access the stone; the ureteroscope was only used to seek
out the target calyx and observe the puncture.
Outcomes
Primary outcomes include postoperative stone-free rates, defined on a 2-mm or 3-mm
NCCT scan obtained within 3 months after surgery, and complication rates as defined
by the Clavien-Dindo system. Residual fragments were classified as follows: Grade
A (no stone remnants), Grade B (≤2.0 mm remnants), and Grade C (2.1-4.0 mm remnants).
These outcomes were presented as percentages and assessed by univariate χ2 analysis.
Secondary outcomes included operative and fluoroscopy times, estimated blood loss,
stone volume, and volume of stone removed per minute of surgery. Blood loss was estimated
by calculating the difference between preoperative and postoperative hemoglobin levels
in g/dL. Renal function was evaluated through the measurement of serum creatinine
levels and estimated glomerular filtration rate. Estimated glomerular filtration rate
was calculated using the novel National Kidney Foundation and American Society of
Nephrology Task Force–endorsed CKD-EPI-2021 race-free equation.11 Preoperative and postoperative day 1 values were recorded. Values 3 to 6 months after
surgery were used to assess for acute or chronic renal impairment. These continuous
variables were presented as mean/SD or median/IQR. Multiple unpaired sample unequal
variance 2-tailed t tests were used.
As a secondary outcome of our study, we sought to adjust the effect of group allocation
(ie, CT-only vs CT+iVR) on stone-free rates (either absolute stone free or relative
stone free [RSF]) by the surgeon’s years of experience. As such, 2 multivariable logistic
regression models were built. The outcomes of the 2 multivariable models were absolute
stone-free status (defined as a binary “yes” or “no” categorical variable) or RSF
status (defined as a binary “yes” or “no” categorical variable). The assessed covariates
were determined a priori to the analysis based on the clinical judgment of the authors
and avoiding the stepwise approach model. The included covariates were group allocation
(CT-only vs CT+iVR) and surgeon experience (included in 4 categories: surgeon 1—40+
years of experience, surgeon 2—30+ years of experience, surgeon 3—10+ years of experience
and surgeon 4—<10 years of experience).
iVR Model Production and Display
Two or three-millimeter slice thickness NCCT scans were uploaded into an open-source
application, 3D Slicer (NIH, Bethesda, Maryland).12 Using a voxel-based virtual paintbrush, a research physician manually labeled the
relevant intrarenal and perirenal structures, slice by slice, using the generic anatomic
color-coding scale. The resulting 3D model was subsequently rendered into an iVR simulation.
The Oculus Quest 2 headset (Meta Platforms Inc, Menlo Park, California) facilitated
the fully immersive experience of 3D model visualization, whereas the touch controllers
allowed the user to manipulate the iVR model and enhanced the surgeon’s visual awareness
of the patient’s unique anatomy (Figure 1). The iVR models specifically were limited to the renal parenchyma and collecting
system with the proximal ureter, renal arterial and venous vasculature as well as
aorta and inferior vena cava, pleura, surrounding spine and ipsilateral ribs, and
liver or spleen dependent on the right or left kidney, respectively. The “immersive”
nature of the iVR model enabled the surgeon to rotate the kidney and its surrounding
anatomy such that the kidney could be viewed from any orientation; the aforementioned
anatomic structures could be individually viewed and manipulated. Furthermore, the
model allowed for the removal of the surrounding anatomy and renal parenchyma, to
provide a clearer view of the collecting system and its relation to the calculus (Figure 2).
Figure 1. Immersive virtual reality model creation framework. Thin slice (2-3 mm), completely
deidentified, noncontrast CT scans in the standard Digital Imaging and Communications
in Medicine format are uploaded into 3D Slicer. Using a voxel-based virtual paintbrush,
a research physician manually labeled the relevant intrarenal and perirenal anatomical
landmarks, slice by slice, using a generic anatomic color-coding scale. For ease of
use, segmentations are built from inside-out, starting with the renal vascular bed
(A) and collecting system (B) and then proceeding to the renal parenchyma (C) and
axial skeleton (D). The renal stone burden (E) and surrounding viscera (ie, spleen,
liver, pleura, bones) are the last segmented anatomical structures (F). The resulting
3D models (G) are subsequently rendered into an immersive virtual reality simulation
using the Oculus Quest 2 headset (H). These models allow the surgeon not only to visualize
and manipulate the relevant anatomy, but also to simulate the percutaneous approach
to the proposed calyx of choice. IVC indicates inferior vena cava.
Figure 2. Immersive virtual reality model. A, Overview of the 3D CT-based immersive virtual
reality model of the kidney parenchyma, collecting system, blood vessels, and skeletal
structures. B, Immersive virtual reality model manipulation is depicted; here the
renal collecting system is distinguished from the now ghosted renal parenchyma because
of prior contact of the user’s virtual hand with the renal parenchyma. C, The surgeon’s
virtual hand has pulled the parenchyma away from the collecting system. Red arrows
indicate the interpolar and lower pole stone burden. D, The model has been rotated
such that the access to the stone (yellow) is visualized through the surgeon’s percutaneous
calyx of choice (blue arrow). The surgeon can “peer” into the calyx and travel down
the infundibulum to see if it is truly in line with the stone from that particular
approach.
On the morning of surgery, the surgeon reviewed the NCCT scans and completed a Likert-scale
questionnaire designed to assess their spatial understanding of the stone’s position
within the collecting system and their intended calyx of entry. Subsequently, if the
patient was randomized to the CT+iVR group, the surgeon then interacted with the iVR
model and completed a second Likert-scale questionnaire to assess if viewing the iVR
model had in any way altered their initial operative plan.
Patient Evaluation
Patients completed a preoperative questionnaire (1 = strongly disagree; 5 = strongly
agree) to assess their understanding and comfort with their planned PCNL procedure.
This was administered initially after viewing the NCCT scans and then a second time
after reviewing the iVR model, as facilitated by either an attending physician or
a research fellow.
Statistical Analysis
Analysis was performed using IBM SPSS Statistics, version 29.0 (IBM Corp, Armonk,
New York). Sample size was determined based on an a priori power analysis (Supplemental
Material, https://www.jurology.com).
RESULTS
The study had a CONSORT score of 22 out of 22.13 From 2019 to 2023, a total of 175 patients undergoing PCNL were preoperatively randomized
into the CT-only group (n = 89) or CT+iVR group (n = 86; Figure 3). Demographic data, preopoerative and postoperative variables, are summarized in
Table 1. There were no differences between groups when accounting for patient age, body mass
index, American Society of Anesthesiologists score, median stone burden (quantified
as cumulative stone length), stone density, or lithotripsy modality used (Table 1).
Abbreviations: ASA, American Society of Anesthesiologists Physical Status Classification;
eGFR, estimated glomerular filtration rate; Ho:YAG, holmium: yttrium aluminum garnet;
iVR, immersive virtual reality; POD1, postoperative day 1; POM3, postoperative month
3.
P values calculated with unpaired sample unequal variance 2-tailed t test.
P values calculated using χ2 test.
For both groups, 75% of postoperative NCCT scans were performed within 2 weeks after
the PCNL procedure. When comparing the time to postoperative NCCT scan between the
2 groups using an unpaired sample, unequal-variance, 2-tailed t test, we found no statistically significant difference between the CT-only and CT+iVR
groups with regard to the timing of the postoperative CT scan (mean difference: 1.8
days; 95% CI for the difference-5.7-9.3 days).
The CT+iVR group had a statistically significant improvement in absolute stone-free
rate (34% vs 20%, P = .04) and RSF rate (< 4 mm; 63% vs 47%, P = .04; Figure 4). Of note, the mean volume of stone ablated/minute of surgery was not significantly
higher in the CT+iVR vs CT-only group (mean difference: −16 mm3; 95% CI for the difference −25 to 7.3 mm3).
Figure 4. Postoperative main outcomes: stone-free rates and Clavien-Dindo complications. *P value < .05. iVR indicates immersive virtual reality.
In 2 multivariable logistic regression models, preoperative iVR visualization maintained
a statistically significant impact on the objective clinical outcomes, when stratifying
absolute stone-free (P < .04) and RSF (P = .008) rates by surgeon’s experience; irrespective of surgical experience, outcomes
benefited from preoperative iVR visualization (Table 2). This finding corroborated the subjective postoperative surgeon’s assessment as
recorded on the Likert-scale questionnaire. A preoperative iVR experience enhanced
each surgeon’s understanding of the perirenal anatomy, stone’s size, stone’s location,
and optimal transcalyceal access (all P < .01; Table 3). After viewing the iVR, the calyx of entry for optimal stone ablation was altered
in 30% of cases.
Table 2. Multivariate Binary Logistic Regression Analysis of Stone-Free Outcomes, Stratified
by Group (CT-Only/CT+Interactive Virtual Reality) and Surgeon Experience (Surgeon
1-4)
Variables in the equation
95% CI for OR
B
SE
Wald
df
Sig
OR
Lower
Upper
Absolute stone free
Group allocation
0.753
0.367
4.214
0.040
2.124
1.035
4.360
Surgeon
6.249
4
0.181
Surgeon 1
−0.380
1.470
0.067
0.796
0.684
0.038
12.194
Surgeon 2
−0.884
1.556
0.322
0.570
0.413
0.020
8.723
Surgeon 3
−1.303
1.462
0.795
0.373
0.272
0.015
4.765
Surgeon 4
−1.314
1.662
0.625
0.429
0.269
0.010
6.986
Constant
−0.377
1.451
0.067
0.795
0.686
Relative stone free
Group allocation
0.883
0.334
6.983
0.008
2.419
1.256
4.657
Surgeon
5.997
4
0.199
Surgeon 1
0.788
1.483
0.282
1
0.595
2.198
0.120
40.244
Surgeon 2
0.805
1.545
0.272
0.602
2.238
0.108
46.208
Surgeon 3
0.106
1.464
0.005
0.942
1.112
0.063
19.595
Surgeon 4
−0.754
1.631
0.214
0.644
0.471
0.019
11.496
Constant
−0.442
1.458
0.092
0.762
0.643
Abbreviations: OR, odds ratio.
Variable(s) entered on step 1: group, surgeon.
Table 3. Surgeons’ Questionnaire Answers Postoperatively Comparing the Utility of CT Alone
With That of Reviewing the Virtual Reality Model Preoperatively
95% CI of the difference
Questionnaire
Mean
Lower
Upper
P value
1. The VR model helped me navigate the intrarenal anatomy (vs CT scan)
−1.5
−1.7
−1.2
< .001
2. The VR model correlated well with regard to the calyx chosen for percutaneous access
(vs CT scan)
−1.1
−1.4
−0.8
< .001
3. The VR model correlated well with regard to the renal pelvis intrarenal collecting
system (vs CT scan)
−1.1
−1.4
−0.8
< .001
4. The VR model correlated well with regard to the location of the stone (vs CT scan)
−0.9
−1.1
−0.7
< .001
5. The VR model correlated well with regard to the size of the stone (vs CT scan)
−1.2
−1.4
−1.0
< .001
6. The VR model correlated well with the severity of hydronephrosis (vs CT scan)
−0.8
−1.0
−0.6
< .001
Abbreviations: VR, virtual reality.
There was no difference in overall complication rates between the 2 groups; however,
Clavien II+IIIa complications were significantly lower in the CT+iVR group (3.5% vs
12.4%, P = .03; Figure 4, Table 4).
Table 4. Comparison of Clavien II+IIIa Complications Between the CT-Only and CT+Immersive Virtual
Reality Group
The postoperative questionnaire was administered to both CT+iVR and CT-only cases;
a statistically significant difference existed only between the pre-iVR and post-iVR
answers. From a preoperative patient education perspective, iVR visualization resulted
in an enhanced understanding of stone size, location, and shape as well as the rationale
for PCNL recommendation (all P < .001). The iVR experience did not correlate with a decrease in patient concern
(P = .4) or patient worry (P = .5) about the upcoming procedure (Table 5).
Table 5. Patient’s Questionnaire Answers Before and After Immersive Virtual Reality Model Review
95% CI of the difference
Questionnaire
Mean
Lower
Upper
P value
1. Understanding stone location before and after the VR model
−0.6
−0.9
−0.3
< .001
2. Understanding stone size and shape before and after the VR model
−0.7
−1.0
−0.4
< .001
3. Understanding why PCNL was recommended before and after the VR model
−0.2
−0.4
0.10
.03
4. Concerned about surgery before and after the VR model
0.1
−0.4
0.5
.4
5. Worried about the anesthetic before and after the VR model
0.0
−0.153
0.2
.5
6. Anesthetic on my mind before and after the VR model
−0.1
−0.260
0.01
.1
7. Would want to know as much as possible about anesthetic before and after the VR
model
0.0
−0.207
0.2
.4
8. Worried about procedure before and after the VR model
0.1
−0.07
0.3
.1
9. Procedure on my mind continuously before and after the VR model
0.05
−0.3
0.2
.3
10. Want to know as much as possible about procedure before and after the VR model
Securing safe and effective access to the renal collecting system is among the most
challenging steps of PCNL because placement of the nephrostomy tract affects surgical
outcomes.14,15 Conventional imaging modalities such as ultrasound or fluoroscopy offer only a limited,
2-dimensional anatomical visualization.16 To overcome this difficulty, some surgeons have created CT-curated, patient-specific
physical models of the stone-bearing kidney. While 3D printing enables the creation
of these models that one can hold and view in all dimensions, this method has some
serious drawbacks.16-20 First, it is both time-consuming and costly (ie, $500/model).16 Second, the model is static; thus, the user cannot interact with the model’s components
or gain an understanding of the perirenal anatomy.
Virtual reality bridges the gap between standard and novel imaging modalities for
PCNL.6,21-24 It has 3 different categories: augmented reality (AR), mixed reality (MR), and iVR;
each of these formats has its own principles, applications, techniques, and necessary
equipment.6,21-23
AR uses computer-generated models overlaid on tangible objects and thus seeks to enhance
a user’s real-world environment.21,22 Rassweiler-Seyfried et al21,22 effectively implemented AR during PCNL, using a tablet in conjunction with anatomical
markers to enable live projection of renal anatomy during percutaneous access. Overall,
despite enhancing the surgeon’s understanding of the anatomy, the radiation exposure
failed to be reduced by using AR.21,22
Similar to AR, MR enables surgeons to intraoperatively project iVR models onto a patient’s
skin, creating a holographic image of underlying anatomy. However, as opposed to AR,
these models are not just mere static projections imposed on the real world but rather
interactive models which the user can manipulate at the time of puncture.24 By doing so, one could hypothesize improved puncture time and reduced radiation exposure.
Indeed, the pivotal work by Porpiglia et al23 in MR use for PCNL revealed a 50% decrease in radiation exposure; however, surprisingly,
this reduction did not correspond with a decreased puncture time, likely because of
the learning curve required to integrate MR into clinical practice and the small sample
size (N = 10).
In contrast to AR or MR, iVR completely immerses the viewer in a digital world of
synthetic, interactive, anatomically correct elements. iVR allows the surgeon to visualize
and manipulate the relevant renal and perirenal anatomical landmarks as well as view
an intrarenal stone through the planned nephrostomy tract.6 This interaction is both visual and interactive as the operator’s virtual hand can
remove the renal parenchyma from the collecting system to better view the stone’s
relationship to the various calyces and infundibulae.6 Moreover, components of the segmentation can be “ghosted” and allow the surgeon and
the patient to gain an optimal understanding of the stone’s position within the collecting
system given a transiently, transparent renal parenchyma.6
In this study, from a surgeon’s standpoint, the iVR experience resulted in a safer,
more effective PCNL. Overall, the rates of achieving stone-free status and the occurrence
of Clavien II to IIIa complications favored the iVR group. This was true irrespective
of the surgeon’s experience level, as demonstrated by our multivariate binary logistic
regression analysis. Of note, despite surgeons initially feeling confident in their
surgical plans based on CT scans alone, their approach was altered in 30% of cases
after reviewing the iVR models. This indicates the ability of iVR to further refine
surgical strategies based on an improved presentation of the anatomy and the attendant'
ability of the surgeon to view it in multiple orientations and to tightly focus on
the pathology at hand and its immediate surrounding structures.
Our findings align with the outcomes of the prospective randomized control study by
Cui et al16 examining the efficacy of preoperative visualization of patient-specific 3D-printed
models in guiding PCNL procedures. Through a 1:1 randomization scheme into either
a CT-only or 3D-printed group, Cui et al16 observed significant advantages favoring the latter. These benefits included reduced
complications, fewer and more accurate punctures, and improved stone-free rates.16
Not surprisingly, 3D printing, AR, and iVR have not exhibited advantages in terms
of radiation exposure over standard methods.6,16-22 While these technologically advanced modalities assist surgeons in selecting the
optimal access calyx for stone ablation and assessing the relationship between the
renal collecting system, adjacent vessels, and nearby organs, they do not accommodate
respiratory movements or organ deformation.6,16-23 This limitation could one day be overcome by MR.23
From a patient perspective, the educational significance of iVR technology is noteworthy.
Most of the patients reviewing the iVR model reported an enhanced understanding of
stone characteristics, such as location, size, and shape of their stone as well as
the indication for the PCNL procedure. Moreover, we observed a trend toward decreased
worry regarding the procedure after iVR visualization (P = .092). Likewise, patients exposed to 3D-printed kidney characteristics demonstrated
improved comprehension of the procedure and expressed higher satisfaction with the
doctor-patient communication.16
There are limitations to our study. First, the quality of the iVR model heavily relies
on the quality of the initial CT imaging and the proficiency of personnel in creating
models using a 3D slicer. In addition, the learning curve associated with the creation
of the iVR models is steep; initial models take up to 10 hours to segment and render.7 In our experience, after 20 cases, the time to produce the iVR model falls to 90
minutes, still a clearly laborious endeavor. Given the mean hourly wage estimations
provided by the US Bureau of Labor Statistics in May 2023 for a radiology technician
or a radiologist, this would correspond to a cost of an iVR model ranging from $112.5
to $255. Although still more affordable than the $500 cost described by Cui et al
for creating a personalized 3D-printed model, we believe that full automation is needed
to widely disseminate the iVR preoperative PCNL planning approach. One potential solution
would be the future development of an artificial intelligence neural network capable
of accurately and efficiently identifying relevant anatomical landmarks, thereby streamlining
the creation of iVR models.25 Another limitation of the study is that we did not record the specific amount of
time each surgeon spent viewing the CT scan or the iVR model. Indeed, surgeons viewing
the iVR model had the benefit of spending time first viewing the CT scan and then
the iVR model providing them with a longer period studying the patient’s anatomy.
We are unable to determine whether having the surgeon in the CT-only group spend even
more time reviewing the NCCT scan would have altered the results of our study. Finally,
while surgeons and patients were unaware of the randomization status until the day
of the procedure, given the nature of the intervention group (ie, visualization of
an iVR model preoperatively), we realize that this is not blinding in the true scientific
sense.
CONCLUSIONS
Preoperative viewing of a patient-specific iVR model altered the surgeon’s plan for
calyceal access in 30% of cases. Moreover, preoperative visualization of iVR models
resulted in a safer and more effective PCNL procedure. The benefit of this novel technology
was independent of a surgeon’s overall experience. In addition, there was significant
patient benefit from viewing their iVR anatomy with regard to a better understanding
of their stone burden and the rationale for a PCNL procedure.
REFERENCES
1. . Surgical management of stones: American Urological Association/Endourological Society
guideline, part I. J Urol.2016; 196(4):1153-1160. doi: 10.1016/j.juro.2016.05.090Link, Google Scholar
6. . Pilot assessment of immersive virtual reality renal models as an educational and preoperative
planning tool for percutaneous nephrolithotomy. J Endourol.2019; 33(4):283-288. doi: 10.1089/end.2018.0626Crossref, Medline, Google Scholar
11. . A unifying approach for GFR estimation: recommendations of the NKF-ASN task force
on reassessing the inclusion of race in diagnosing kidney disease. Am J Kidney Dis.2022; 79(2):268-288.e1. doi: 10.1053/j.ajkd.2021.08.003Crossref, Medline, Google Scholar
14. . A prospective and randomised trial comparing fluoroscopic, total ultrasonographic,
and combined guidance for renal access in mini-percutaneous nephrolithotomy. BJU Int.2017; 119(4):612-618. doi: 10.1111/bju.13703Crossref, Medline, Google Scholar
15. . Access related complications during percutaneous nephrolithotomy: urology versus radiology
at a single academic institution. J Urol.2006; 176(1):142-145. doi: 10.1016/S0022-5347(06)00489-7Link, Google Scholar
17. . Impact of three-dimensional printed pelvicaliceal system models on residents’ understanding
of pelvicaliceal system anatomy before percutaneous nephrolithotripsy surgery: a pilot
study. J Endourol.2016; 30(10):1132-1137. doi: 10.1089/end.2016.0307Crossref, Medline, Google Scholar
18. . A new model with an anatomically accurate human renal collecting system for training
in fluoroscopy-guided percutaneous nephrolithotomy access. J Endourol.2014; 28(3):360-363. doi: 10.1089/end.2013.0616Crossref, Medline, Google Scholar
19. . MP20-02 Three-dimensional printed kidney models with extensive urolithiasis: a novel
resident educational tool for planning percutaneous nephrolithotomy. J Urol.2016; 195(4S):e212. doi: 10.1016/j.juro.2016.02.2771Link, Google Scholar
20. . Use 3D printing technology to enhance stone free rate in single tract percutaneous
nephrolithotomy for the treatment of staghorn stones. Urolithiasis.2020; 48(6):509-516. doi: 10.1007/s00240-019-01164-8Crossref, Medline, Google Scholar
25. . Efficient and accurate computed tomography-based stone volume determination: development
of an automated artificial intelligence algorithm. J Urol.2024; 211(2):256-265. doi: 10.1097/JU.0000000000003766Link, Google Scholar
Funding/Support: This research received no external funding from any funding agency in the public,
commercial, or not-for-profit domains.
Conflict of Interest Disclosures: The Authors have no conflicts of interest to disclose.
Ethics Statement: All human subjects provided written informed consent with guarantees of confidentiality.
Author Contributions:
Conception and design: Cumpanas, Landman, Jiang, Clayman, Patel, Ali.
Data analysis and interpretation: McCormac, Cumpanas, Gao, Tran, Altamirano-Villarroel, Landman, Lee, Hernandez, Ericson,
Jiang, Clayman, Ali, Tano.
Data acquisition: Cumpanas, Tran, Hernandez, Ali.
Drafting the manuscript: McCormac, Cumpanas, Tran, Altamirano-Villarroel, Landman, Lee, Hernandez.
Critical revision of the manuscript for scientific and factual content: McCormac, Gao, Tran, Altamirano-Villarroel, Landman, Hernandez, Ericson, Jiang, Clayman,
Patel, Ali, Tano.
![Figure 3.Trial summary (CONSORT [Consolidated Standards of Reporting Trials] diagram). iVR indicates immersive virtual reality; US, ultrasound.](/cms/asset/678aeb42-3522-474f-b7b2-c27efc41f7f9/ju.0000000000004284f3.gif)
