Using Intraoperative Portable CT Scan to Minimize Reintervention Rates in Percutaneous Nephrolithotomy: A Prospective Trial

Abstract

Background and Purpose:

More than 40% of patients undergoing percutaneous nephrolithotomy (PCNL) are left with residual stone fragments and often require secondary procedures. Portable CT (PCT) technology allows surgeons to obtain intraoperative cross-sectional imaging, identify and extract residual stones immediately, and thereby reduce the need for subsequent procedures. This prospective trial evaluates how incorporation of PCT during PCNL affects perioperative outcomes.

Patients and Methods:

We prospectively enrolled eligible patients undergoing initial PCNL for this trial (n = 60), which entailed a single intraoperative CT abdomen and ipsilateral antegrade ureteroscopy when the surgeon felt stone treatment was visually complete. If residual fragments were identified, the surgeon continued nephroscopy to find and remove them; if not, the procedure was concluded. These patients were compared with a retrospective cohort (n = 174) who underwent initial PCNL with postoperative imaging performed the following day.

Results:

The two cohorts had similar demographic properties and stone characteristics, and location of percutaneous access. In the prospective arm, 50% of intraoperative PCT scans identified residual fragments, prompting continuation of surgery to remove them. This cohort had significantly higher stone-free rate (82% vs 36%, p < 0.01), lower rate of planned reintervention (7% vs 32%, p < 0.01), lower rate of urgent presentation with ureteral obstruction (0% vs 7%, p = 0.04), lower total CT-based effective radiation dose (8.4 mSv vs 14.6 mSv, p < 0.01), and shorter length of stay (2.3 days vs 3.5 days, p < 0.01) when compared with the retrospective cohort that did not use intraoperative PCT.

Conclusions:

Obtaining an intraoperative PCT scan during PCNL can substantially improve perioperative outcomes. Further evaluation of this modality through a randomized controlled trial is warranted.

Clinical Trial Registration Number: NCT04556396.

Introduction

Percutaneous nephrolithotomy (PCNL) is considered a first-line management option for large kidney stones.¹ Due to significant stone burdens, more than 40% of these patients are left with residual stone fragments after their initial PCNL, requiring secondary procedures.^2,3 Additionally, nearly 20% of patients considered stone free intraoperatively have been demonstrated to have significant residual stone burden.⁴ Residual fragments are discovered either on postoperative imaging or when patients return symptomatically. Regardless, both scenarios create the possibility of a subsequent procedure to address the residual stone burden.

Cone beam-based portable CT (PCT) is a well-established portable imaging technique that allows cross-sectional imaging to be obtained intraoperatively, rather than postoperatively, and its feasibility in PCNL has been demonstrated.^5
–7 Incorporating this modality during PCNL allows the surgeon to determine whether the procedure should be continued—to remove residual fragments—or if it can be safely concluded. If successfully adopted, PCT could obviate the need for dedicated postoperative CT scans and reduce the need for subsequent procedures and utilization of health care resources.

Our aim is to determine how intraoperative PCT during PCNL can assist in surgical decision making, and how this technology affects patient-related outcomes such as stone-free rate, need for subsequent procedures, and length of stay following PCNL.

Materials and Methods

Study design

This trial, including a prospective arm and a retrospective arm, was approved by the Institutional Review Board and registered with ClinicalTrials.gov (NCT04556396). For the prospective arm, all adult patients (≥18 years old) undergoing initial PCNL were recruited before the surgery (June 2020 to March 2021). Patients were excluded if they had only ureteral stones or had undergone PCNL in the same renal unit within the previous 6 months. This cohort was then compared with a retrospective cohort, using the same criteria, of all patients undergoing initial PCNL at our institution (January 2018 to December 2019). We studied their perioperative imaging, intraoperative findings, and postoperative clinical courses. In patients undergoing bilateral PCNL, each renal unit was analyzed separately.

Demographic properties (age, sex, race, ethnicity, body mass index, and laterality) were noted for both cohorts. We studied preoperative imaging to register specifics regarding stone size, location, and details about percutaneous access. Total operative time, perioperative radiation exposure, intraoperative and postoperative imaging findings, and presence of residual stone fragments were noted. The postoperative course for each patient was evaluated, with attention to complications, inpatient length of stay, and need for planned or unplanned reintervention.

The primary outcome was the percentage of renal units requiring subsequent reintervention to remove residual stones (i.e., a “second-look”) within 90 days of the index procedure. Secondary outcomes included average length of inpatient stay, incidence of complications, the percentage of renal units totally cleared of stones (i.e., the “stone-free” rate), the incidence of acute obstructive presentations within 90 days, and the effective radiation dose received by each patient.

Power calculation

To determine the appropriate size of our prospective cohort, we performed a statistical power calculation to detect a difference between the reintervention rate of our retrospective cohort and an upper-bound projection of the reintervention rate in our PCT cohort. Our retrospective cohort of 174 patients yielded a reintervention rate of 32%, and our upper-bound projection of the second-look rate in our PCT cohort was 15%. Using an α = 0.05 and 1 − β = 0.80, we calculated that a prospective sample size of 60 renal units would allow for adequate statistical power.

Statistical analyses

Variables were compared between the retrospective and prospective cohorts using Student's t-test for continuous variables and χ ² tests or Fisher's exact tests for categorical variables, as appropriate. Renal units were categorized based on original cohort designation (i.e., intention-to-treat), not based on the imaging modality ultimately used (i.e., per-protocol). A p-value of 0.05 was used to denote statistical significance. All statistical analyses were performed using SPSS version 26.0 (2019; IBM Corp., Armonk, NY).

Perioperative workflow

The patients at our tertiary referral center were managed by one of four urologists with standardized techniques. Percutaneous access was obtained by the Interventional Radiology team before each scheduled PCNL and all surgeries were performed in the prone position. A balloon dilator and a 30F sheath were used to establish renal access. At the discretion of the operating surgeon, stones were treated using a combination of pneumatic/ultrasonic fragmentation, laser lithotripsy, and/or basket extraction.

In the prospective cohort, after both nephroscopy and ipsilateral antegrade ureteroscopy were performed and the surgeon felt that all stones had been removed, an intraoperative portable cone beam CT scan (O-arm^®; Medtronic, Minneapolis, MN) of the abdomen was performed (Supplementary Fig. S1). If residual fragments were identified on the intraoperative PCT (Supplementary Fig. S2), the surgeon would continue nephroscopy to find and remove them. Patients were only considered stone free if all residual stone fragments identified on PCT were retrieved. Otherwise, they would undergo a dedicated CT scan the next day for further planning. Patients could undergo standard postoperative imaging at the discretion of the surgeon. At the conclusion of each procedure, a 20F council-tip nephrostomy catheter was placed over a 5F Double-J open-ended ureteral catheter.

In the retrospective cohort, our standard practice was to perform follow-up imaging on postoperative day 1; if any residual ureteral fragments or >3 mm renal fragments were identified, a second look PCNL procedure would be scheduled during the same admission or an outpatient ureteroscopy with lithotripsy would be arranged depending on surgeon discretion. Otherwise, all patients who were stone free underwent nephrostomy tube removal and were discharged from the hospital when appropriate.

Radiation dose

PCT generates images using a cone-shaped X-ray beam and two-dimensional detectors, whereas standard CT use a fan-shaped X-ray beam and one-dimensional detectors, thus the radiation outputs cannot be directly compared. Converting the outputs to effective radiation doses, however, permits meaningful comparison of the radiation dose associated with each modality. To accomplish this, we multiplied the dose length product of each study by a published conversion factor to estimate the effective radiation dose.^8
–10

Results

Demographic properties, physical characteristics, laterality, stone size and location, and location of percutaneous access did not differ between cohorts (Table 1).

Table 1.

Comparison of Patient Characteristics Between Retrospective Cohort and Prospective Cohort

Characteristic	Retrospective cohort (n = 174)	Prospective cohort (n = 60)	p
Age, years, mean (SD)	57 (15)	59 (15)	0.36
Female sex, n (%)	88 (51)	37 (62)	0.14
Race, n (%)
White race	145 (84)	46 (77)	0.12^a
Black race	12 (7)	5 (8)
Asian race	5 (3)	4 (7)
Other race	11 (6)	3 (5)
Unknown race	0 (0)	2 (3)
Ethnicity, n (%)
Non-Hispanic	151 (88)	54 (90)	0.65
Hispanic	21 (12)	6 (10)	0.65
Height, m, mean (SD)	1.68 (0.13)	1.66 (0.14)	0.40
Weight, kg, mean (SD)	93 (32)	93 (33)	0.96
BMI, mean (SD)	33 (10)	33 (10)	0.72
Renal unit laterality, n (%)
Left	86 (49)	26 (43)	0.42
Right	88 (51)	34 (57)	0.42
Part of bilateral procedure	43 (25)	9 (15)	0.12
Stone location, n (%)
Complete staghorn	18 (11)	1 (2)	0.06^a
Partial upper pole staghorn	8 (5)	3 (5)
Partial lower pole staghorn	24 (14)	9 (16)
Upper pole	16 (9)	3 (5)
Mid pole	13 (8)	0 (0)
Lower pole	57 (33)	24 (41)
Renal pelvis	28 (16)	15 (26)
Proximal ureter	8 (5)	3 (5)
Stone size, mm, mean (SD)	23.5 (13.8)	20.4 (9.7)	0.06
Percutaneous access location, n (%)
Upper pole	50 (29)	15 (25)	0.60^a
Mid pole	24 (14)	6 (10)
Lower pole	81 (47)	34 (58)
Multiple accesses	17 (10)	4 (7)

Denotes Fisher's Exact test used due to subgroup n < 5.

BMI = body mass index; SD = standard deviation.

Of the 60 renal units enrolled in the prospective arm, 52 underwent intraoperative PCT. Fifty percent of these (26/52) had no residual fragments; of the other 50% who were found to have residual stones, the surgeons were able to immediately continue the procedure and successfully cleared residual fragments from 88% of renal units (23/26). Reasons for PCT deferral were: en bloc stone extraction without lithotripsy or fragmentation (2), abortion of procedure before completion of lithotripsy (3), and surgeon preference for standard post-op CT (3). Reasons for standard postoperative imaging in addition to PCT were: low-quality PCT due to morbidly obese habitus (1), large amount of residual burden on PCT (2), and suspected false-positive findings on PCT due to residual contrast or intraparenchymal calcifications (4). Upon completion of the initial PCNL procedure, 82% (49/60) of all renal units in the prospective arm were rendered completely stone free; of the other 11, only 4 renal units (7%) had residual stone burden significant enough for reintervention (Supplementary Fig. S3).

Of the 174 renal units in the retrospective arm, 36% (63/174) were completely stone free after the initial PCNL, and 32% of renal units (56/174) underwent planned reintervention within 90 days. The other 32% (55/174) were not stone free but their stone burden was not significant enough to justify a reintervention (Supplementary Fig. S4). Comparing the two cohorts, the prospective arm had a significantly higher stone-free rate after initial PCNL (82% vs 36%, p < 0.01), significantly lower rate of planned reintervention within 90 days (7% vs 32%, p < 0.01), and significantly lower rate of urgent presentation with ureteral obstruction (0% vs 7%, p = 0.04) (Table 2).

Table 2.

Comparison of Outcomes Between the Retrospective Cohort and Prospective Cohort

Characteristic	Retrospective cohort (n = 174)	Prospective cohort (n = 60)	Odds ratio or mean difference (95% CI)	p
Perioperative outcomes
Procedure aborted before lithotripsy completion, n (%)	2 (1)	2 (3)	3.0 (0.4 to 21.5)	0.27^a
Stone-free on postoperative imaging, n (%)	63 (36)	49 (82)	7.8 (3.8 to 16.2)	<0.01^b
Fluoroscopy dose area product, Gy·cm², mean (SD)	23 (114)	22 (75)	0 (−38 to 38)	0.98
Fluoroscopy cumulative air kerma, mGy, mean (SD)	27 (33)	64 (196)	−37 (−90 to 15)	0.20
Length of procedure, minutes, mean (SD)	84 (45)	95 (40)	−11 (−24 to 2)	0.11
Estimated blood loss, mL, mean (SD)	72 (50)	50 (40)	22 (8 to 36)	<0.01^b
Clavien–Dindo grade III–V complications, n (%)	17 (10)	2 (3)	0.3 (0.1 to 1.4)	0.16^a
Any perioperative complication, n (%)	37 (21)	9 (15)	0.7 (0.3 to 1.4)	0.29
Total CT-based effective radiation dose, mSv, mean (SD)	14.6 (9.8)	8.4 (6.6)	6.2 (3.3 to 9.2)	<0.01^b
Inpatient length of stay, days, mean (SD)	3.5 (2.5)	2.3 (1.5)	1.1 (0.4 to 1.8)	<0.01^b
Primary stone composition, n (%)
Calcium oxalate	90 (52)	33 (57)	—	0.51^a
Calcium phosphate	33 (19)	8 (14)
Uric acid	12 (7)	4 (7)
Struvite	20 (12)	9 (16)
Cystine	8 (5)	0 (0)
Mixed/other	9 (5)	4 (7)
Postoperative outcomes (within 90 days)
Planned surgical reintervention (i.e., “second-look”), n (%)	56 (32)	4 (7)	0.2 (0.1 to 0.4)	<0.01^a,b
Planned intervention type, n (%)
Percutaneous nephrolithotomy	52 (93)	4 (100)	—	1.00^a
Ureteroscopy with holmium laser lithotripsy	4 (7)	0 (0)	—	1.00^a
Urgent presentation with ureteral obstruction, n (%)	12 (7)	0 (0)	—	0.04^a,b
Unplanned surgical reintervention, n (%)	5 (42)	—	—	—
Unplanned intervention type, n (%)
Ureteral stent placement	4 (80)	—	—	—
Percutaneous nephrostomy tube placement	1 (20)	—	—	—

Denotes Fisher's Exact test used due to subgroup n < 5.

Denotes statistical significance.

CI = confidence interval.

Performing intraoperative PCT added an average of 11 minutes to the length of procedure in the prospective cohort, but this was not statistically significant (84 minutes vs 95 minutes, p = 0.11). While there was no difference in the radiation exposure from intraoperative fluoroscopy between the two cohorts, the prospective cohort received a lower perioperative total CT-based effective radiation dose (8.4 mSv vs 14.6 mSv, p < 0.01) (Table 2).

There was no significant difference in adverse outcomes between the retrospective and prospective cohorts, examined either as severe (i.e., Clavien–Dindo IIIa or greater) complications (10% vs 3%, p = 0.16) or all complications (21% vs 15%, p = 0.29). Patients in the prospective arm had a shorter inpatient length of stay compared with those in the retrospective arm (2.3 days vs 3.5 days, p < 0.01) (Table 2).

Discussion

Patients undergoing PCNL frequently have residual stone fragments, often requiring the patient to undergo additional procedures.^2,3 PCT can be used to obtain intraoperative cross-sectional images,⁶ allowing surgeons to immediately extract residual stones and avoid subsequent procedures. In this prospective trial, we sought to evaluate how incorporation of PCT technology during initial PCNL affects outcomes such as stone-free rate, need for subsequent procedures, and length of stay following PCNL.

In the prospective arm, we found that 50% of renal units had residual stones identified on intraoperative PCT, which aligns with previously reported rates of residual fragments after initial PCNL.^2,3,11 Using intraoperative PCT, however, we were able to intervene immediately and address the residual stones before the surgery concluded. Fewer renal units in our prospective arm (50%) had residual stones on intraoperative PCT compared with postoperative standard CT in our retrospective arm (64%), possibly due to the difference in sample size or a Hawthorne effect among the surgeons.

Upon completion of the initial PCNL procedure, 82% of all renal units in the prospective arm were considered stone free, compared with a 36% stone-free rate in the retrospective cohort. This significant difference may represent an overestimation, as most patients in the prospective arm did not undergo standard postoperative CT after having the intraoperative PCT. In a feasibility evaluation of PCT in PCNL, Van den Broeck and colleagues reported similar findings: that 50% of patients undergoing standard postoperative imaging after initial PCNL had residual stones, whereas only 20% had residual stones after the use of intraoperative PCT; reintervention rates were not reported in either cohort.⁶

By incorporating intraoperative PCT during initial PCNL, we decreased our reintervention rate from 32% to 7%. With respect to unplanned postoperative hospital visits, 12 patients (7%) in the retrospective arm presented urgently with signs of ureteral obstruction within 90 days of the index procedure—with five requiring an unplanned reintervention. Comparatively, there were no such events among the patients in the prospective arm within 90 days of the index procedure (p = 0.04). This can be attributed to the number of patients (55/174) in the retrospective cohort who were not stone free on postoperative imaging, but their stone burden was not significant enough to justify additional procedures.

It is generally accepted that small residual fragments can be monitored without reintervention after PCNL, however, there is a reported 16.9% reintervention on long-term follow-up of small residual fragments (≤4 mm).¹¹ In contrast to our standard practice, patients in our prospective cohort had the luxury of intraoperative PCT so we were able to intervene regardless of fragment size because the patients were still under anesthesia. Hence, we were able to render the patients truly stone free and minimize the need for multiple procedures as well as unplanned interventions.

The prospective cohort received a lower total CT-based effective radiation dose (8.7 mSv vs 14.6 mSv, p < 0.01). This number includes all CT-based images that a patient received in the process of evaluating and managing the stone in question; that is, it represents the combined exposure from preoperative, intraoperative, and postoperative CT scans. The noted reduction in radiation dosing with PCT is likely due to a lower-dose image produced by the cone beam technology (Supplementary Fig. S2) and condensed scanning bounds of the PCT. As the distal ureter could not be visualized on PCT, we performed ipsilateral antegrade ureteroscopy to rule out any residual fragments in the ureter. These radiation exposure numbers could still be improved upon by the adoption of low-dose or ultra-low-dose CT scanning technology.^12,13

In terms of perioperative complications, 21% of patients in our retrospective arm suffered from adverse events. Similarly, De la Rosette et al. reported an overall complication rate of 21.5%, with 4.1% being Clavien–Dindo IIIa or worse, in an analysis of over 5800 patients undergoing PCNL.¹⁴ There were no significant differences in complications between our two cohorts, regardless of whether the analysis compared all complications (21% vs 15%, p = 0.29) or it was stratified into severe (i.e., Clavien–Dindo IIIa or greater) complications only (10% vs 3%, p = 0.16). There is a trend toward fewer complications in the prospective cohort, which may be associated with the significantly lower percentage of patients undergoing secondary procedures.

Using PCT, the average length of stay for patients in the prospective arm was reduced by 1.2 days compared with the retrospective arm. Previously reported lengths of stay after PCNL have ranged widely from 1.7 to 11.4 days, as perioperative workflow for PCNL varies widely.¹⁵ The reduction in length of stay is most replicable at institutions where the secondary intervention is typically performed within the same admission, as avoiding reintervention thereby shortens the length of stay.

There are various notable limitations to this study. First, this trial was performed with prospective and retrospective arms rather than in a randomized controlled fashion. At our institution, it proved difficult to blind the surgeons when implementing intraoperative PCT due to the logistics of reserving and preparing the PCT machine before surgery. In this setting, utilizing a comparable retrospective cohort from the same institution seemed to offer an acceptable statistical comparator as a true randomized control arm as there were no other changes in clinical practice of these surgeons. Second, one could argue that a patient cannot be considered completely stone free without a confirmatory postoperative image even after the use of PCT, since patients in the retrospective arm were only considered stone free based on postoperative imaging. Given that our purpose with this project is to reduce overall morbidity for our patients, however, we believed it reasonable to consider a patient stone free if every stone identified on the PCT was subsequently extracted. This avoids unnecessary radiation and parallels the clinical practice in our retrospective cohort, as we did not obtain confirmatory imaging following second-look procedures.

Lastly, there is an inherent limitation in the lower resolution of the PCT, such that it may overestimate the stone-free rate compared with standard CT. Despite this, even if small residual fragments went undetected due to the lower resolution, we did not find any increase in adverse outcomes or unplanned interventions in our PCT cohort. We hope our findings will encourage other surgeons and researchers, especially at large centers who already utilize a PCT machine for neurosurgical or orthopedic procedures, to incorporate this modality into their PCNL workflow. Additional studies of this technique at different institutions or in a randomized manner could increase the external validity of the concept.

Conclusion

While PCNL has demonstrated superior efficacy in the management of large kidney stones, intraoperative PCT may be able to improve outcomes by allowing surgeons to visualize residual fragments and intervene within the same anesthetic encounter. We found that employing PCT during PCNL can afford higher stone-free rate, lower reintervention rate, lower rate of urgent presentation with ureteral obstruction, reduced radiation dose to the patient, and shorter length of stay. Further evaluation of this modality through a randomized controlled trial is warranted.

Footnotes

Authors' Contributions

The authors of this study have participated fully in the conduct of the research, analysis of the data, development, and approval of the article.

Acknowledgment

This work was supported internally by the Department of Urology at Loyola University Medical Center.

Author Disclosure Statement

No competing financial interests exist.

Funding Information

No funding was received for this article.

Supplementary Material

Supplementary Figure S1

Supplementary Figure S2

Supplementary Figure S3

Supplementary Figure S4

Abbreviations Used

References

Assimos

, Krambeck

, Miller

, et al. Surgical management of stones: AUA. Available at: www.auanet.org/guidelines/surgical-management-of-stones-(aua/endourological-society-guideline-2016) Accessed December 1, 2021 .

el-Nahas

, Eraky

, Shokeir

, et al. Factors affecting stone-free rate and complications of percutaneous nephrolithotomy for treatment of staghorn stone. Urology, 2012; 79:1236–1241.

Desai

, De Lisa

, Turna

, et al. The clinical research office of the endourological society percutaneous nephrolithotomy global study: Staghorn versus nonstaghorn stones. J Endourol, 2011; 25:1263–1268.

Nevo

, Holland

, Schreter

, et al. How reliable is the intraoperative assessment of residual fragments during percutaneous nephrolithotomy? A prospective study. J Endourol, 2018; 32:471–475.

Roy

, Angle

, Jenkins

, et al. Cone beam computed tomography for percutaneous nephrolithotomy: Initial evaluation of a new technology. J Endourol, 2012; 26:814–818.

Van den Broeck

, Zhu

, Kusters

, Futterer

, et al. Percutaneous nephrolithotomy with intraoperative computed tomography scanning improves stone-free rates. J Endourol, 2021; 35:267–273.

Vicentini

, Botelho

LAA

, Braz

JLM

, et al. Use of the Uro Dyna-CT in endourology—The new frontier. Int Braz J Urol, 2017; 43:762–765.

AAPM

Reports—The measurement

, reporting, and management of radiation dose in CT [Internet]. [cited 2021 May 1]. Available at: https://www.aapm.org/pubs/reports/detail.asp?docid=97 Accessed December 1, 2021 .

Medtronic O-arm Imaging System Dosimetry Report. Available at: https://www.medtronic.com/us-en/healthcare-professionals/products/neurological/surgical-imaging-systems/o-arm/dose-considerations.html Accessed December 1, 2021 .

10.

GE Revolution CT Technical Reference Manual. November 2015. Available at: https://manualzz.com/doc/38864775/ge-revolution-ct-user-manual Accessed December 1, 2021 .

11.

Emmott

, Brotherhood

, Paterson

, et al. Complications, re-intervention rates, and natural history of residual stone fragments after percutaneous nephrolithotomy. J Endourol, 2018; 32:28–32.

12.

Rob

, Bryant

, Wilson

, Somani

. Ultra-low-dose, low-dose, and standard-dose CT of the kidney, ureters, and bladder: Is there a difference? Results from a systematic review of the literature. Clin Radiol, 2017; 72:11–15.

13.

Cheng

, Shkolyar

, Chang

, et al. Ultra-low-dose CT: An effective follow-up imaging modality for ureterolithiasis. J Endourol, 2020; 34:139–144.

14.

de la Rosette

, Assimos

, Desai

, et al. The Clinical Research Office of the Endourological Society Percutaneous Nephrolithotomy Global Study: Indications, complications, and outcomes in 5803 patients. J Endourol, 2011; 25:11–17.

15.

, Autorino

, Kim

, et al. Percutaneous nephrolithotomy versus retrograde intrarenal surgery: A systematic review and meta-analysis. Eur Urol, 2015; 67:125–137.

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.

0.00 MB

3.77 MB

0.33 MB

0.16 MB

0.11 MB