Abstract
Purpose:
To describe our experience with robot-assisted laparoscopic dismembered extravesical cross-trigonal ureteral reimplantation (RADECUR)—a novel minimally invasive surgical technique for treatment of obstructed megaureter.
Materials and Methods:
A retrospective review of our prospectively collected data, of all pediatric patients who underwent unilateral RADECUR for the treatment of primary or secondary obstructed megaureter at two medical centers during a 3-year period. Patient demographics, perioperative data, surgical technique, complications, and results are described.
Results:
Thirty-five pediatric patients underwent RADECUR between January 2016 and May 2019. Median age and weight at surgery were 28 months (interquartile range [IQR]: 20–58) and 20 kg (IQR: 13–27), respectively. There were no conversions to an open procedure. Median console time was 100 minutes (IQR: 90–125). Ten patients underwent intracorporeal excisional tapering of the obstructed ureter. Mean pre- and postoperative ureteral diameters were 14.5 mm (IQR: 12–18) and 7 mm (IQR: 0–10), respectively (p < 0.0001). All but one patient demonstrated a reduction in hydronephrosis and ureteral diameter on follow-up ultrasonography. Grade 1–2 Clavien–Dindo complications occurred in five patients (14%) and another patient developed a grade 3 complication. Three patients developed postoperative febrile urinary tract infection, one of them was found to have high-grade reflux and subsequently underwent open ureteral reimplantation.
Conclusions:
Unilateral robotic extravesical cross-trigonal ureteral reimplantation for treatment of obstructed megaureter in the pediatric population is safe and effective. Further larger scale comparative studies with other surgical techniques are needed to determine the role of RADECUR in the armamentarium of treatment for distal ureteral obstruction.
Introduction
Distal ureteral obstruction results in hindered urine evacuation and leads to dilation of the ureter and renal collecting system. 1 This condition was coined by Caulk in 1923 as megaureter. 2 Obstruction at the ureterovesical junction can be congenital (potentially with concomitant reflux) or acquired. The overall incidence of primary obstructed megaureter (POM) is estimated to be 1:1500–2000 live births. 3 Most will resolve over time, nevertheless, ∼20% will have significant renal function deterioration requiring surgery.
Secondary (acquired) obstructed megaureter (SOM) in the pediatric population is associated with utilization of bulking agents for correction of vesicoureteral reflux (VUR). Incidence of distal ureteral obstruction after bulking agent treatment of VUR is estimated to be 0.5%–1.3%. High volume of injected material, recurrent injections, high-grade reflux, inflamed bladder mucosa, and injection in cases of overlooked partial obstruction are all risk factors for developing secondary obstruction. 4 Many surgical approaches for the correction of obstructed megaureter such as laparoscopic extravesical reimplantation and the vesicoscopic approach have been described. 5,6 Various factors such as age, gender, etiology of the obstruction, concomitant anomalies, and surgeon's preference all play a role in the choice of surgical technique.
The da Vinci robotic surgical system has been shown to improve surgeon dexterity, ergonomics, and observation, providing a minimally invasive option for complex reconstructive procedures in children. 7 –9 The use of robotic surgery in the pediatric urologic population has been studied for the treatment of VUR. Over the past decade, robot-assisted laparoscopic ureteral reimplantation (RALUR) has become a viable minimally invasive surgical option for high-grade VUR. However, higher than expected complication rates and suboptimal reflux resolution rates precluded consensus on the role of RALUR for the surgical management of VUR.
We have hypothesized that further extension and modification of RALUR to patients with obstructed megaureters may improve patient outcomes with regard to all benefits of minimally invasive surgery. We implemented a novel technique for treatment of obstructed megaureter, robot-assisted laparoscopic dismembered extravesical cross-trigonal ureteral reimplantation (RADECUR). We reviewed our prospectively collected data and assessed the short-term safety and feasibility of the procedure.
Materials and Methods
After institutional ethical committee approval, data from all children who underwent unilateral RADECUR at two academic medical centers during a 3-year period were reviewed. All cases were performed by two surgeons, A.N. and B.C. Data gathered included demographic information, perioperative data, short-term perioperative complications, and follow-up. Hydronephrosis and ureteral diameter were assessed using standard ultrasound (US) measures recorded preoperatively and during last postoperative follow-up. Data were prospectively collected (Table 1) and retrospectively reviewed. Surgery success was defined if follow-up ultrasonography showed an improvement in the degree of hydronephrosis and reduction of ureteral diameter or an MAG-3 renal scan showed a nonobstructed drainage pattern. Surgical failure was defined as worsening of hydronephrosis and an obstructed pattern on diuretic renogram or need for an additional surgical intervention. Primary outcomes for the study were short-term surgical success and surgical complications graded by the Clavien–Dindo classification. 10
Demographic and Clinical Data
IQR = interquartile range; VUR = vesicoureteral reflux.
Study setting and population
Male and female children with either primary or acquired unilateral obstructed megaureter were included. Indications for surgical intervention in patients with POM were previously published. 11 In brief, patients with Society of Fetal Ultrasound (SFU) III and IV hydronephrosis and >1.2 cm ureteral dilation associated with renal function deterioration were allocated for surgical correction. Patients with SOM included in this study were cases following endoscopic correction of VUR by bulking agents (Deflux or Vantris). Patients with clinical manifestations, such as flank pain or infections, new onset or worsening of hydronephrosis on ultrasonography, and evidence of obstruction defined by diuretic renogram were referred for reimplantation. Pre- and postoperative clinical parameters were compared in the study group and in a further subgroup analysis comparing POM vs SOM and a tapered vs nontapered surgery.
Surgical technique
All patients were operated using the daVinci Xi® or Si® surgical systems. After induction of general anesthesia or combined general and caudal anesthesia with morphine, a urethral catheter was inserted after preparing the patient to facilitate bladder filling during surgery.
The patient was placed in the low lithotomy position, and the procedure was performed with central docking. Pneumoperitoneum was achieved through the umbilicus using the Hasson technique. A camera port was inserted into the umbilicus and two 8 mm robotic ports were inserted 5–6 cm diagonally on the right and left lower abdomen. An additional 5 or 10 mm assistant port was inserted, typically into the upper abdominal quadrant, contralateral to the reimplantation side. Each port site was infiltrated with 0.25% bupivacaine just above the fascia and under vision, upon trocar insertion.
After dissection and exposure of the bladder and ureter, the bladder was filled with saline to better observe and plan the detrusorrhaphy and anastomosis. The distal ureter was ligated and transected as distally as possible and proximal to the obstructed segment. In cases of obstruction after endoscopic therapy, the distal segment of the ureter was excised.
A transcutaneous holding suture to pull up the bladder was utilized per surgeon preference. A 4–5 cm long transverse incision in the posterior–inferior aspect of the bladder and through the detrusor was made. The incision began at the ipsilateral side and was extended to the contralateral side (Fig. 1). Care was taken not to disturb the integrity of the bladder mucosa or injure the healthy contralateral ureter. An opening in the bladder mucosa at the contralateral point was made using scissors without cautery to minimize mucosal damage.

Transverse subdetrusor incision in the posterior–inferior aspect of the bladder with preservation of the bladder mucosa.
If ureteral tailoring was performed, a 12 cm long 8F ureteral catheter was inserted through one of the trocars. The tube was introduced into the distal ureter that served both as a precautionary measure while trimming the ureter and as a mold while tailoring the ureter. A distal suture was placed to align the ureter with the tube, and a running suture was used starting at the proximal part of the ureter. A 4-0 or 5-0 polydioxanone (PDS®) or Maxon was used intracorporeally in accordance with the size of the ureter. At completion of tailoring, the tube was removed, and a guidewire was placed through one of the ports. A double pigtail ureteral stent was introduced over the wire, and the guidewire was removed. The distal coil protruding from the ureter was then manipulated into the bladder through a new ureteral orifice that was opened in the bladder mucosa. For completion of the anastomosis between the ureter and bladder, an anchoring suture was placed between the bladder and ureter at the most lateral point incorporating the detrusor. The anastomosis was then carried out with interrupted 5-0 PDS and incorporating partial thickness of the detrusor when appropriate. The detrusorrhaphy was closed over the ureter with a running 5-0 PDS or 3-0 V-Loc™ suture (Fig. 2). Ureteral adventitia was incorporated at the most contralateral and most ipsilateral points to prevent the ureter from sliding. No external drains were inserted. The urethral catheter was removed 24–48 hours postoperatively. Postoperative pain management was consistent with the surgeon's practice pattern for other robotic procedures. The ureteral stent was removed under short general anesthesia 2–6 weeks after surgery.

Detrusorraphy of the left distal ureter after anastomosis to the bladder mucosa.
Follow-up
Postoperative follow-up included serial ultrasonography at 1 and 3 months, and then every 6 months. MAG-3 renal scan was performed if dilation did not improve or worsened.
At the initiation of the study, we considered the first eight cases as a pilot study and performed voiding cystourethrography (VCUG) for all to verify that the surgical technique did not cause postoperative reflux. Thereafter, postoperative VCUG was performed as clinically indicated in cases of febrile urinary tract infection (UTI).
Statistical analysis
Continuous variables were described as medians and interquartile range (IQR). Owing to lack of normal distribution, study group and subgroup analyses were performed using the Mann–Whitney U test. Categorical variables were described as frequency and percentage. Categorical variables were compared using the chi square test of independence. All statistical analyses were performed on SPSS statistics© v.25 (IBM, Armonk, NY) and statistical significance was set at p < 0.05.
Results
Thirty-five patients (23 boys, 12 girls) underwent RADECUR during a 3-year period. Patients demographic and perioperative data are presented in Table 1. Indications for surgery were POM in 24 patients and obstruction after endoscopic treatment of high-grade VUR in 11 patients (SOM).
Median age and weight at surgery for the entire group were 28 months (IQR: 20–58) and 20 kg (IQR: 13–27), respectively. Median operative console time was 100 minutes (IQR: 90–125). There were no conversions to an open procedure. Median postoperative hospital stay was 1.5 days (IQR: 1–2.25). Median pre- and postoperative ureteral diameters were 14.5 mm (IQR: 12–18) and 7 mm (IQR: 0–10), respectively (p < 0.0001). All but one patient demonstrated a reduction in ureteral diameter and degree of hydronephrosis during follow-up. In this case, an MAG-3 renal scan was performed that demonstrated a nonobstructed renogram. Median patient follow-up was 10 months (IQR: 5–22).
A subgroup analysis comparing patients operated because of POM (n = 24) vs SOM (n = 11) demonstrated that there were no statistically significant differences in clinical parameters or outcomes (Table 2).
Comparison of Surgery Indications (Primary Obstructed Megaureter vs Secondary Obstruction After Reflux Treatment)
POM = primary obstructed megaureter; SOM = secondary obstructed megaureter.
A second subgroup analysis comparing patients who underwent intracorporeal excisional tapering of the obstructed ureter (n = 10) vs surgery without tapering (n = 25) was performed. Preoperative ureteral diameter was significantly more prominent in patients who underwent tapering (15.5 mm [IQR: 13–18] vs 12 mm [IQR: 10–16], p = 0.037). In addition, longer operative time (console time—137 minutes [IQR: 100–146] vs 90 minutes [IQR: 80–120], p = 0.001) and longer hospital stay (3 days [IQR: 2–3] vs 1 day [IQR: 1–2], p = 0.001) were demonstrated in the tapered than in the nontapered group. No other differences in clinical parameters or outcomes were shown (Table 3).
Comparison of Surgery with Ureteral Tapering vs No Tapering
Grade 1–2 Clavien–Dindo postoperative complications occurred in five patients (14%). Two patients developed urinary retention after removal of the catheter. A urethral catheter was reinserted and removed after additional 48 hours. The patients voided without residual urine.
Three patients developed postoperative UTI treated conservatively, all of them underwent VCUG. One patient demonstrated high-grade reflux on VCUG and subsequently underwent open reimplantation (one patient was considered as surgical failure).
Grade 3 Clavien–Dindo complications occurred in one patient who developed port site herniation of omentum. The herniation was reduced bedside with short sedation.
As part of a pilot study, the first eight patients underwent postoperative VCUG. One of these patients was found to have grade 1 VUR in the ipsilateral ureter and was managed conservatively. None of these patients demonstrated de novo reflux into the contralateral ureter.
Discussion
In this study, we describe a novel surgical approach for the treatment of obstructed megaureter in the pediatric population. Unilateral RADECUR was implemented for POM or SOM. Intracorporeal tailoring was performed as indicated to achieve a ratio of 1/5 in ureteral diameter/tunnel length.
Treatment of obstructive megaureter has evolved during the past decades. In their study from 1989, Peters et al. 12 reported on 42 patients treated with open surgical correction for moderate-to-severe obstructive megaureter. All infants showed functional and structural urographic improvement with a mean follow-up of 2.3 years. Reflux was seen postoperatively in eight patients that resulted in repeat reimplantation in three patients. During follow-up, two additional patients showed progressive obstruction, which required a second surgical repair. They concluded that repair of obstructed megaureters in early infancy improves renal drainage and offers the potential for preventing renal damage before the development of symptoms or infection.
Since then, several studies have shown that other approaches such as expectant management and endoscopic treatment can be applied in the majority of POM cases. 6,13,14
Surgical repair for treatment of obstructive megaureter has been reviewed in the British Association of Pediatric Urologists (BAPU) consensus 15 and is usually reserved for cases with initial differential renal function <40%, symptomatic patients, or failure of conservative management (breakthrough UTIs, worsening dilation, or deteriorating renal function).
Teklali and associates 6 retrospectively reviewed 35 children treated with endoscopic ureteral dilatation and Double-J stent placement for obstructive megaureter. Thirty-four of 35 children (97%) remained asymptomatic with preserved renal function during follow-up. Two children required open surgery for a failed endoscopic treatment procedure and one more patient required secondary reimplantation surgery. Twelve of 35 children had a complication, including 2 Clavien 3 complications. Comparison of pre- and postoperative ultrasonography showed a significant decrease in ureteral diameter in 31 cases. Postoperative scintigraphy showed a significant improvement in renal drainage in 20 children.
Lopez et al. have described laparoscopic extravesical ureteral reimplantation for the treatment of POM. 16 In this study, laparoscopic extravesical reimplantation with ureteral tapering was performed in seven patients between 2011 and 2014 using the Lich Gregoir technique. Postoperative follow-up with ultrasonography, VCUG, and MAG-3 renogram at 6 months showed a nonobstructive pattern in all patients. One case of VUR was found during VCUG.
When reviewing the literature and in comparison with open and laparoscopic reimplantation for obstructive megaureter, success rate, operating time, and complications are comparable when considering RADECUR. In comparison with open ureteral reimplantation in the Cohen technique for treatment of VUR, operating times are reported to be 110–180 minutes 17,18 with hospital stay of 2.8–5.6 days. Operative time for RADECUR in our series is well within the range of open reimplantation while providing shorter length of hospital stay.
In our study, failure was defined as the need for further surgical intervention, therefore, the overall success rate was 97%. The reported success rate for ureteral reimplantation without tapering in the literature is high (95%–100%), in contrast, when performing tapering, complication rate increases and success decreases. DeFoor and associates 19 reported on their experience with 65 megaureters undergoing open ureteral reimplantation and tapering. The definition for success was improvement of hydronephrosis and no VUR. Success rate was 76% for extravesical reimplantation and 86% for the intravesicle approach.
In our study when comparing the subgroup of patients who underwent tapering with those who did not, we see similar results.
In the tapered group (10 cases), 1 case required further surgery caused by symptomatic high-grade VUR and 1 patient developed grade 1 reflux not requiring further intervention. All but one patient showed significant improvement in ureteral dilatation, this patient had a nonobstructed renogram on MAG-3 renal scan. There were no cases or recurrent obstruction requiring subsequent surgery during the follow-up period of this study.
As VCUG was not performed for all cases, one should keep in mind that reflux after RADECUR is potentially underdiagnosed in this study.
Three cases (8.6%) developed febrile UTI with one of them was found to have VUR. This patient initially underwent ureteral tailoring because of severely dilated ureters.
Postoperative urinary retention is a well-known complication of bilateral RALUR and was also reported in unilateral RALUR. 20 Urinary retention after ureteral reimplantation is thought to be related to traction or injury of the pelvic parasympathetic nerve plexus during distal ureteral dissection. This phenomenon is documented after bilateral as well as unilateral cases regardless of an intravesical or extravesical approach. 20 Although urinary retention is usually transient, in rare cases it can be long standing with the risk of significant morbidity. Two patients in our series developed urinary retention, which resolved after 48 hours.
Our described surgical technique has several potential advantages. By performing a horizontal detrusorrhaphy as opposed to a vertical detrusorrhaphy, we believe the anastomosis is more ergonomic to perform, rather than trying to extend the tunnel proximally on the ipsilateral side. This may be even more advantageous in a boy, in whom there is concern of iatrogenic injury to the Vas deferens, which is in proximity to the ureter. We believe the subdetrusor transverse incision with preservation of the bladder mucosa is essential for prevention of reflux from the reimplanted ureter.
The da Vinci robotic system improves the ability to perform suturing in various operations, nevertheless, intracorporeal tailoring is still challenging. Ureteral tailoring can be performed with a running suture or interrupted suture. We believe the tailored part of the ureter should be longer than the tunnel. If a part of the dilated ureter is incorporated in the detrusorrhaphy, kinking and obstruction of the ureter can potentially develop. Should the suture line of the tailored ureter face the bladder mucosa or face the overlying detrusor? This still needs to be clarified. A ureteral suture line facing the bladder mucosa can potentially result in formation of fistulas between the ureter and the bladder, leading to VUR, especially if the mucosa was violated at the time of dissection. In contrast, there is perhaps a greater risk of intraperitoneal urine leak if the suture line faces the detrusor suture lines. Another potential downside of a cross-trigonal approach as was implemented in our study is a difficulty with endoscopic ureteral intubation during follow-up if needed. This issue may not be a factor with the Politano-Leadbetter type of repair, which was utilized in previously described robotic techniques.
The limitations of our study include a small cohort and the descriptive nature of the report. VCUG was performed in only the first eight pilot cases to verify the durability of the surgical technique. Since reflux was not present in any of the cases, we recommended VCUG only as clinically indicated. Therefore, the real prevalence of postsurgical reflux was not fully evaluated. It is our belief that reserving VCUG for patients who develop postoperative febrile UTI during follow-up can reduce the number of unnecessary invasive imaging procedures. Median follow-up is relatively short; however, once postoperative imaging demonstrated improvement, it is unlikely to deteriorate at longer follow-up. 21
There were mild surgical differences in technique between the two surgeons participating in the study, nevertheless, results from both surgeons were similar.
Another possible limitation is the widespread application of this technique to children with more severe dilation. Since in our study group, median preoperative ureteral diameter was 14.5 mm, children with severe dilation might have worse outcomes in terms of operative times, postoperative VUR, short- and long-term complications, and rates of reoperation. Expansion of this technique to severely dilated high-risk patients may find that RADECURE is not a suitable surgical technique.
Conclusions
Unilateral robotic extravesical cross-trigonal ureteral reimplantation for treatment of obstructed megaureter in the pediatric population is safe and effective. Further larger scale comparative studies to other surgical techniques are needed to determine the role of RADECUR in the armamentarium of treatment for distal ureteral obstruction.
Footnotes
Author Disclosure Statement
No competing financial interests exist.
Funding Information
No funding was received for this article.
