A Pilot Study to Determine the Role of Spatulating the Ureter During Pyeloplasty in Children for Ureteropelvic Junction Obstruction in the Robotic Era

Abstract

Purpose:

Spatulation during ureteropelvic junction obstruction repair was evaluated in children treated by robot-assisted retroperitoneal pyeloplasty anastomosis (RRPA).

Methods:

Intraoperative video recordings (IVRs) of RRPA (n = 22 ureters) performed at a single institute between 2018 and 2022 were reviewed blindly by 5 independent surgeons for perceived difficulty of suturing (DOS; 5 = impossible; 4 = difficult; 3 = tedious; 2 = slow; 1 = easy) and spatulation ranking as superior (+1), inferior (−1), or unnecessary (0). The retroperitoneal space was accessed in the lateral decubitus position using a closed technique under direct vision to avoid air leakage and subcutaneous emphysema. All subjects had a Double-J stent (4.7F) placed.

Results:

Subjects had similar demographics and preoperative ureter diameters. IVRs were RRPA with spatulation of the ureter on the lateral side (RRPA +SP) (n = 13) and RRPA without spatulation of the ureter (RRPA −SP) (n = 9). Overall DOS scores and coefficients of variation for time taken to place one suture were similar. Total anastomotic time was significantly shorter for RRPA −SP; 67.9 ± 8.4 minutes versus 57.9 ± 9.2 minutes, P = .01. Overall spatulation ranking was 0. Postoperative scanning showed improved drainage in 12 of 13 (92%) in RRPA +SP and 8 of 9 (88%) in RRPA −SP; differences were not significant. One anastomotic stricture in RRPA −SP required open repair.

Conclusions:

RRPA was quicker and more precise without spatulation. Outcomes of scanning 1 year after RRPA were similar for RRPA −SP and RRPA +SP.

Introduction

Anderson–Hynes dismembered pyeloplasty is the conventional treatment for ureteropelvic junction (UPJ) obstruction with a reported success rate of >90%.^1,2 Laparoscopic pyeloplasty is achieving acceptance for treating UPJ obstruction in the pediatric population as success rates are high, similar to open pyeloplasty.³ The most commonly used technique is a dismembered pyeloplasty with spatulation of the ureter. However, the most difficult, time-consuming, and critical maneuvers during laparoscopic pyeloplasty are ureteral spatulation, apical ureteral stitch placement, and the ureteropelvic anastomosis.

Several techniques for laparoscopic pyeloplasty have been described to overcome these difficult maneuvers.^4,5 Consequently, there is a rather steep learning curve for laparoscopic pyeloplasty because of technical challenges requiring considerable experience and advanced laparoscopic skills.

The introduction of robotic surgical systems (RSSs) has helped overcome the limitations of laparoscopic surgery. Results of robot-assisted laparoscopic pyeloplasty have consistently been reported to be similar to open pyeloplasty, even in children.^6,7 In a previous study on hepaticojejunostomy, suturing and knot tying with RSSs were more precise, meticulous, and stable compared with laparoscopy, even when lumens of <5 mm were anastomosed.⁸ Such precision was also considered likely to benefit suturing anastomoses during pediatric urological surgery such as pyeloplasty, but the necessity for spatulation has been questioned.

In this study, the role of spatulation in the RSS era, specifically, during robot-assisted retroperitoneal pyeloplasty anastomosis (RRPA) was evaluated by comparing surgical outcomes and functional results of pyeloplasty after resection of an obstructed UPJ and reconstruction with and without spatulation of the ureter.

Materials and Methods

All pediatric patients with UPJ obstruction who underwent RRPA between November 2018 and June 2022 at a single institute were reviewed. All RRPA were performed by a team of 2 board-certified pediatric surgeons (H.K. and A.Y.) with extensive technical experience. Only patients with unilateral single system hydronephrosis were included in this study. The diagnosis of UPJ obstruction was made using computerized tomography, ultrasonography, magnetic resonance imaging, diuretic renography, and diethylene-triamine-penta-acid (DTPA) renography.

Inclusion criteria for this study were significant flank pain, infection, or decreased renal function. Patients with previous renal surgery were excluded. Perioperative and postoperative data including age and weight at surgery, body mass index (BMI), RRPA diameters, number of sutures required for RRPA, RRPA ratio (RRPA diameter divided by the number of sutures), total anastomotic time (TAT), TAT per suture, coefficient of variation for TAT per suture, total operative time, time taken to ambulate postoperatively, duration of drain insertion, and duration of hospitalization were analyzed.

A scoring system developed previously, proven to quantify subjective difficulty of suturing (DOS) in a published report,^8,9 was used again in this study. In brief, five levels of difficulty were chosen arbitrarily and defined to convey the concept of the level, simply. Thus, impossible (too challenging to progress smoothly) scored 5, difficult (challenging but smooth progression) scored 4, tedious (challenging but hindered by technical, physical, or visual issues) scored 3, slow (poor general progression without hindrances and without stress) scored 2, and easy (smooth progress without any hindrances/stress) scored 1.

A panel of 5 independent board-certified senior specialist pediatric surgeons experienced in open and laparoscopic urological procedures was tasked with scoring DOS during RRPA blindly from intraoperative video recordings (IVRs). They were also asked to rank the perceived usefulness of spatulation from IVRs blindly and asked to rank whether spatulation made RRPA superior (+1), inferior (−1), or made no difference (0).

Postoperative outcome was assessed using diagnostic imaging 12 months after RRPA surgery and was defined as resolution of obstruction on ultrasonography or DTPA renography. Preoperative and postoperative split renal function and glomerular filtration rate (GFR) were compared using DTPA renography.

Surgical technique

Common preoperative management

After induction of general endotracheal anesthesia, the patient was placed in the lateral decubitus position with the affected side up, slightly flexed with a small gel positioner under the contralateral iliac crest. All retroperitoneal procedures were performed in the lateral decubitus position. The operating surgeon, assistant, and nurse were positioned along the patient's back with the monitor on the opposite side of the operating table. A Foley catheter was inserted to decompress the bladder.

The retroperitoneal space was accessed according to a closed technique to prevent leakage during insufflation and subcutaneous emphysema, described previously elsewhere.¹⁰ In brief, the initial access point was the costovertebral angle at the lateral border of the sacrospinalis muscle. A 5 mm bladeless optical trocar (Xcel™; Ethicon Endo-Surgery, Cincinnati, OH) was used to allow the operator complete visual control of orientation while advancing directly into the retroperitoneal space under direct vision, through the lumbodorsal fascia, until fat, characteristic of the retroperitoneal space, was identified. Insufflation with CO₂ at 8–12 mmHg was commenced.

Blunt camera dissection was then used to dilate the retroperitoneum under endoscopic vision by horizontally swinging the laparoscope from posterior to anterior, taking great care not to injure the retroperitoneum. Under laparoscopic guidance, additional second and third 5 mm trocars were placed just above the iliac crest in the anterior axillary line and in the abdomen, lateral to the iliac fossa. The assistant's trocar should be positioned more posteriorly than usual to reduce the risk of conflicts with the arms of the RSSs.

Robot-assisted retroperitoneal pyeloplasty anastomosis

The da Vinci Surgical system (Intuitive Surgical, Sunnyvale, CA) was used for RRPA. Although the da Vinci RSSs has four arms, only three were used. Three laparoscopic 5 mm trocars were replaced with three da Vinci 8 mm trocars. The RSS platform was prepared and docked from the anterior aspect of the patient using a 0° telescope camera. The dilated renal pelvis was detected and isolated, together with the UPJ and the cranial portion of the ureter.

After the UPJ was fully exposed and transected, UPJ obstruction was treated by RRPA with spatulation of the ureter on the lateral side (RRPA +SP) or by resection of the obstructed segment of the UPJ without spatulation (RRPA −SP) according to the surgeon's preference. Ureter and anastomosis diameters were measured directly during RRPA using a pair of needle driver forceps with 1 mm calibrations.

Microbipolar forceps and a needle driver were inserted on the left and right sides, respectively. RRPA was performed with interrupted 5/0 or 6/0 absorbable sutures with all knots tied intracorporeally. Once the posterior wall anastomosis was completed, a Double-J stent (4.7F) was placed using an antegrade approach transcutaneously with a guidewire or a retrograde approach with a cystoscope under fluoroscopic control to confirm correct positioning.

Common postoperative management

Perioperative cefcapene pivoxil hydrochloride hydrate (Flomox^®; 90 mg/kg per day) was administered intravenously from the day before surgery until postoperative day (POD) 3. A closed suction drain was inserted using the assistant trocar site for monitoring and managing any anastomotic leakage. The Foley catheter inserted for bladder decompression was removed on PODs 2–3. The retroperitoneal drain was removed once output decreased to <10 mL per day. The Double-J stent was removed 4–6 weeks postoperatively.

Statistics/ethics

Data are expressed as mean ± standard deviation. The Student's t test and chi-squared test were used for statistical analysis. A P value of <.05 was considered to be statistically significant. This study was approved by the Juntendo University School of Medicine Institutional Review Board (IRB No.: J18-024) and complies with the Helsinki Declaration of 1975 (revised 1983).

Results

All surgical procedures, RRPA +SP (n = 13) and RRPA −SP (n = 9), were completed successfully with no conversions to open repair. Subject demographic data are summarized in Table 1. There were no significant differences between the two groups for age at the time of surgery: 6.7 ± 4.3 years for RRPA +SP versus 8.1 ± 4.5 years for RRPA −SP (P = .46); weight at the time of surgery: 24.3 ± 13.0 kg for RRPA +SP versus 28.2 ± 13.8 kg for RRPA −SP (P = .50); and BMI: 17.2 ± 1.7 kg/m² for RRPA +SP versus 16.9 ± 2.6 kg/m² for RRPA −SP, (P = .74). Mean ureter size was similar between the two groups for RRPA +SP 4.7 ± 1.3 mm (range: 4–6 mm) versus for RRPA −SP 4.8 ± 0.7 mm (range: 5–9 mm) (P = .83).

Table 1.

Patient Demographics

	RRPA +SP (n = 13)	RRPA −SP (n = 9)	P
Mean age at surgery, years	6.7 ± 4.3	8.1 ± 4.5	NS
Mean weight at surgery, kg	24.3 ± 13.0	28.2 ± 13.8	NS
Mean BMI, kg/m²	17.2 ± 1.7	16.9 ± 2.6	NS
Mean ureter size (range), mm	4.7 ± 1.3 (4–6)	4.8 ± 0.7 (4–6)	NS
Mean RRPA diameter (range), mm	12.4 ± 4.6 (10–20)	6.4 ± 1.5 (5–9)	.001

BMI, body mass index; NS, not significant; PA, pyeloplasty anastomosis; RRPA +SP, robot-assisted retroperitoneal pyeloplasty anastomosis with spatulation of the ureter on the lateral side, RRPA −SP, robot-assisted retroperitoneal pyeloplasty anastomosis without spatulation of the ureter.

Spatulation resulted in mean RRPA diameter being significantly larger in RRPA +SP than in RRPA −SP; 12.4 ± 4.6 mm (range: 10–20 mm) versus 6.4 ± 1.5 mm (range: 5–9 mm), (P = .001). For the same reason, time taken to perform RRPA (TAT) and the mean number of sutures required were also significantly higher in RRPA +SP compared with RRPA −SP, (67.9 ± 8.4 minutes versus 57.9 ± 9.2 minutes, P = .01) and (11.2 ± 3.6 versus 8.1 ± 0.8, P = .02). Mean total operative time for RRPA +SP (314.2 ± 59.8 minutes) was longer than for RRPA −SP (292.3 ± 22.2 minutes) but the difference was not statistically significant, (P = .30). RRPA ratio (RRPA diameter divided by the number of sutures) was significantly higher in RRPA +SP than in RRPA −SP (1.1 ± 0.2 versus 0.8 ± 0.1, P = .001).

There was no significant difference between the two groups for time taken for one suture during RRPA (TAT divided by the number of sutures); 6.2 ± 1.4 minutes for RRPA +SP versus 7.1 ± 1.6 minutes for RRPA −SP (P = .17). The coefficient of variation for the time taken to place one suture was similar for RRPA +SP and RRPA −SP (0.23 versus 0.22). DOS scores for RRPA +SP were also similar for RRPA −SP (1.2 ± 0.5 versus 1.3 ± 0.7, P = .69). Spatulation ranking was 0, unanimously (Table 2).

Table 2.

RRPA +SP Versus RRPA −SP

	RRPA +SP (n = 13)	RRPA −SP (n = 9)	P
Mean total operative time (minutes)	314.2 ± 59.8	292.3 ± 22.2	NS
Mean total anastomotic time for RRPA (minutes)	67.9 ± 8.4	57.9 ± 9.2	<.05
Mean number of sutures	11.2 ± 3.6	8.1 ± 0.8	<.05
Mean RRPA ratio	1.1 ± 0.2	0.8 ± 0.1	<.01
Mean total anastomotic time/suture (minutes)	6.2 ± 1.4	7.1 ± 1.6	NS
Coefficient of variation: total anastomotic time/suture	0.23	0.22
Mean difficulty of suturing during pyeloplasty anastomosis	1.2 ± 0.5	1.3 ± 0.7	NS
Spatulation ranking (n = 5)	All 0 (unnecessary)

NS, not significant; RRPA, robot-assisted retroperitoneal pyeloplasty anastomosis; RRPA +SP, robot-assisted retroperitoneal pyeloplasty anastomosis with spatulation of the ureter on the lateral side, RRPA −SP, robot-assisted retroperitoneal pyeloplasty anastomosis without spatulation of the ureter.

Mean duration of drain tube insertion was similar; 3.3 ± 0.9 days for RRPA +SP versus 3.4 ± 1.1 days for RRPA −SP, (P = .81); mean time taken to ambulate postoperatively: 1.4 ± 0.5 days for RRPA +SP versus 1.7 ± 0.4 days for RRPA −SP (P = .15); and mean hospitalization: 4.6 ± 1.1 days for RRPA +SP versus 4.6 ± 1.0 days for RRPA −SP (P = .99). Mean length of follow-up after surgery was 28.2 ± 14.2 months for RRPA +SP and 26.1 ± 16.6 months for RRPA −SP, (P = .76). There was no significant difference in the overall number of successful outcomes as assessed by postoperative diagnostic imaging: 92.3% (12/13) for RRPA +SP versus 88.9% (8/9) for RRPA −SP (P = .99).

Preoperative and postoperative mean split renal function on DTPA renography were similar; for RRPA +SP: 40.4% ± 12.1% (pre) versus 43.2% ± 11.5% (post), for RRPA −SP: 42.0% ± 7.8% (pre) versus 42.9% ± 7.9% (post), (P = .82). Preoperative and postoperative mean GFRs were also similar; for RRPA +SP: 39.6 ± 14.1 mL/min (pre) versus 44.7 ± 13.9 mL/min (post) and for RRPA −SP: 37.8 ± 9.5 mL/min (pre) versus 38.1 ± 8.8 mL/min (post), (P = .53) (Fig. 1). There was one anastomotic stricture in RRPA −SP (11.1%) and no complications in RRPA +SP. The stricture was repaired by conventional open repyeloplasty (Table 3).

FIG. 1.

Comparison of pre- and postoperative split renal function and GFR on DTPA renography with respect to spatulation. No specific change in function related to spatulation. DTPA, diethylene-triamine-penta-acid; GFR, glomerular filtration rate; RRPA +SP, robot-assisted retroperitoneal pyeloplasty anastomosis with spatulation of the ureter on the lateral side; RRPA −SP, robot-assisted retroperitoneal pyeloplasty anastomosis without spatulation of the ureter.

Table 3.

Postoperative Results

	RRPA +SP (n = 13)	RRPA −SP (n = 9)	P
Mean postoperative drain removal (days)	3.3 ± 0.9	3.4 ± 1.1	NS
Mean time to ambulate(days)	1.4 ± 0.5	1.7 ± 0.4	NS
Mean length of hospital stay (days)	4.6 ± 1.1	4.6 ± 10	NS
Mean length of follow-up (months)	28.2 ± 14.2	26.1 ± 16.6	NS
Success rate	92.3% (12/13)	88.9% (8/9)	NS

NS, not significant; RRPA +SP, robot-assisted retroperitoneal pyeloplasty anastomosis with spatulation of the ureter on the lateral side; RRPA −SP, robot-assisted retroperitoneal pyeloplasty anastomosis without spatulation of the ureter.

Discussion

The application of RSSs to pediatric surgery has enhanced the safety and stability of suturing, reducing morbidity associated with surgical intervention in children.^11,12 In fact, robot-assisted laparoscopic pyeloplasty is the most common RSS procedure performed in pediatric surgery around the world because it allows a complex procedure such as dismembered pyeloplasty to be performed under precise control. The functions of RSSs most attributed to precise meticulous surgery are tremor filtering and motion scaling; large movements made by the operator are converted to minute tremor-free movements of instruments, increasing the operator's dexterity immensely, resulting in much easier suturing and knot tying during an anastomosis because the operator sits comfortably at a console, with no ergonomic issues to hinder performance.

Thus, RSSs allow complicated and delicate reconstructive procedures in challenging circumstances to be more manageable and are likely to be preferred for complicated suturing as the learning curve for robot-assisted procedures is better than for endoscopic suturing alone. Suturing is definitely more precise and easier, definitely outperforming purely endoscopic suturing.

RRPA can be compared with conventional laparoscopic surgery using a selection of calculated markers such as the coefficient of variation. Despite differences in pyeloplasty diameters after spatulation, and the number of sutures required, coefficients of variation for RRPA −SP and RRPA +SP were similar. A lower coefficient of variation infers there is better precision and reproducibility and this reflects the ease to which the operator can suture while sitting comfortably at a console without juggling instruments directly.

DOS scores for IVRs assessed blindly by experienced skilled surgeons also suggest that performance of both RRPA +SP and RRPA −SP resulted in the same stable and precise outcome without stressing the operator, which is advantageous for reducing complications such as leakage and stenosis. Although DOS is based on subjective interpretation of visual images, the basis and rationale for its reproducibility and reliability have been published,^8,9 showing that DOS correlates with the quality of anastomosis as assessed by objective criteria such as TAT, TAT per suture, and the coefficient of variation for TAT per suture.

In this study, the focus was on the role of spatulation and its impact on outcome. To this end, both groups had similar demographics with similar preoperative ureter and anastomosis diameters as well as similar robot-assisted procedures except for spatulation, and there were no statistically significant differences in postoperative recovery and progress such as time taken to ambulate, duration of drain tube insertion, and hospitalization.

In this study, DTPA renography performed preoperatively and 12 months postoperatively did not show any significant differences between RRPA +SP and RRPA −SP with regard to split renal function or GFR, although success rates were similar. In other words, there was no improvement in differential function postoperatively in both groups, suggesting that spatulation of the ureter during RRPA did not appear to make any specific impact on postoperative outcome or contribute to improving renal function assessed preoperatively.

To date, there has been no consensus reached regarding the technique of choice for the anastomosis, although various techniques for ureteropelvic suturing have been described. One factor known to be correlated with success is eliminating trauma to the ureter caused by grasping. All instruments will cause some degree of trauma at the time of suturing that can contribute to postoperative anastomotic complications such as leakage and stricture, even if minimal.

Long-term results are better if grasping can be minimized. A retroperitoneal approach improves access to the UPJ. This approach also allows the kidney to be maneuvered medially or anteriorly as required to maintain exposure of the UPJ without any stay sutures as RSSs can be docked from the anterior aspect. In the RRPA presented in this study, the UPJ was hardly touched before anastomosis. The first sutures between the edges of the ureter and the renal pelvis can be placed meticulously and precisely using RSSs, helping to prevent torsion of the ureter that may otherwise be unrecognized. When spatulation is not performed, only the obstructed segment of UPJ is resected, allowing a reasonable anastomosis to be performed without tension between the ureter and pelvis.

In conclusion, although there are limitations to this study because of its retrospective nature and small number of cases, this would appear to be the first report comparing RRPA +SP with RRPA −SP to assess the role of spatulation in a series of pediatric UPJ obstruction cases in the robotic era. RRPA −SP was performed successfully without difficulty, and outcome of RRPA −SP was comparable with that of RRPA +SP. From this study, RRPA for pyeloplasty in children was safe and feasible provided that robotic surgery was indicated and has potential to reduce anastomosis-related complications. Postoperative function at 12 months would suggest that spatulation of the ureter may not contribute to the overall success of RRPA.

Footnotes

Authors' Contributions

H.K., and A.Y. designed the study. H.K., M.T., T.O., S.S., Y.M., M.T., R.A., Y.Y., and A.Y. were involved in clinical treatment. H.K., T.O., and S.S. collected and analyzed data. H.K. prepared the article. G.J.L. revised the article as a native English speaker. All authors have read and approved the final article.

Disclosure Statement

No competing financial interests exist.

Funding Information

No funding was received for this article.

References

Chan

, Zauberman

, Chan

, et al. Outcomes after implantable collamer lens surgery in a Canadian cohort. Can J Ophthalmol, 2017; 52(2):150–154.

Varda

, Wang

, Chung

, et al. Has the robot caught up? National trends in utilization, perioperative outcomes, and cost for open, laparoscopic, and robotic pediatric pyeloplasty in the United States from 2003 to 2015. J Pediatr Urol, 2018; 14(4):336 e1–e8.

Ravish

, Nerli

, Reddy

, et al. Laparoscopic pyeloplasty compared with open pyeloplasty in children. J Endourol, 2007; 21(8):897–902.

Koga

, Okawada

, Miyano

, et al. Modified retroperitoneal laparoscopic dismembered pyeloplasty for children. J Pediatr Endosc Surg, 2019; 1(2):59–63.

Yang

, Yao

, Li

, et al. A modified suture technique for transperitoneal laparoscopic dismembered pyeloplasty of pelviureteric junction obstruction. Urology, 2015; 85(1):263–267.

Schwentner

, Pelzer

, Neururer

, et al. Robotic Anderson-Hynes pyeloplasty: 5-Year experience of one centre. BJU Int, 2007; 100(4):880–885.

Taktak

, Llewellyn

, Aboelsoud

, et al. Robot-assisted laparoscopic pyeloplasty versus laparoscopic pyeloplasty for pelvi-ureteric junction obstruction in the paediatric population: a systematic review and meta-analysis. Ther Adv Urol, 2019; 11:1756287219835704.

Koga

, Murakami

, Ochi

, et al. Comparison of robotic versus laparoscopic hepaticojejunostomy for choledochal cyst in children: A first report. Pediatr Surg Int, 2019; 35(12):1421–1425.

Koga

, Murakami

, Seo

, et al. Retroperitoneoscopic pyeloplasty for ureteropelvic junction obstruction in children: Value of robotic assistance. J Pediatr Surg, 2023; 58(7):1291–1295.

10.

Koga

, Okawada

, Doi

, et al. Comparison between closed and open techniques for creating a retroperitoneal space for retroperitoneoscopic renal surgery in children. Pediatr Surg Int, 2014; 30(9):933–936.

11.

Esposito

, Masieri

, Castagnetti

, et al. Robot-assisted vs laparoscopic pyeloplasty in children with uretero-pelvic junction obstruction (UPJO): Technical considerations and results. J Pediatr Urol, 2019; 15(6):667 e1–e8.

12.

Spinoit

, Nguyen

, Subramaniam

Role of robotics in children: A brave New World!. Eur Urol Focus, 2017; 3(2–3):172–180.