Endoscopic Management of Primary Obstructive Megaureter: A Systematic Review

Abstract

Background:

The gold standard treatment for primary obstructive megaureter (POM) with declining renal function, worsening obstruction, or recurrent infections is ureteral reimplantation with or without tapering. In infants, open surgery can be technically demanding and associated with significant morbidity. We conducted a systematic review of the literature with special interest in endoscopic management of POM and its outcomes.

Materials and Methods:

A search was conducted of the MEDLINE/Ovid, PubMed, Embase, and Web of Science databases. Only full-text articles written in the English language and involving greater than one reported pediatric case per publication were included. Two authors independently extracted data and assessed strength of evidence for each study.

Results:

We found 11 retrospective and 1 prospective, single institution case series that met selection criteria, describing 222 patients with 237 obstructed renal units. Mean age at time of surgery was 24.6 months. The most common endoscopic approaches were cystoscopy+high-pressure balloon dilation+Double-J ureteral stent placement (49.5%), cystoscopy+incisional ureterotomy+Double-J ureteral stent placement (27.8%), and cystoscopy+Double-J ureteral stent placement (18.9%). For all approaches and age groups, anatomic and functional success rates were 79.3% (146/184) and 76.7% (132/172), respectively. Anatomic success rates were highest in children ≥12 months of age (82.3%, 117/142). Endoscopic retreatment was performed in 15.1% of cases with a 36.7% overall surgical reintervention rate. Forty-one ureters progressed to ureteral reimplantation. Complications were generally mild (Clavien–Dindo Grades I–II), but 12 ureters did develop vesicoureteral reflux. Mean follow-up period was 3.2 years.

Conclusions:

Endoscopic management for persistent or progressive POM in children ≥12 months of age is a minimally invasive alternative to ureteral reimplantation with modest success rates. In infants, it may best be utilized as a temporizing procedure. Approximately one-third of patients require surgical reintervention.

Introduction

Primary obstructive megaureter (POM) is a congenital, intrinsic obstruction of the distal ureteral segment at the ureterovesical junction. The gold standard treatment for persistent or progressive POM with declining renal function, worsening obstruction, or recurrent infections is ureteral reimplantation with or without tapering.^1,2 However, open surgery is invasive. In infants, surgery can be technically demanding and associated with significant morbidity.^3,4 Endoscopic management has emerged as a minimally invasive alternative. Currently, there is no defined role for endoscopic intervention in the management of progressive POM or consensus on its effectiveness. Herein, we present a systematic review of the literature regarding the endoscopic management of POM, focusing on the clinical demographics, surgical approaches, and postoperative outcomes.

Materials and Methods

After receiving exemption status by the IRB at our institution, an extensive electronic database search was conducted of MEDLINE/Ovid, PubMed, Embase, and Web of Science for citations related to endoscopic management of POM. The literature review was completed on September 26, 2016. Keyword phrases searched were “primary obstructive megaureter, endoscopic management” and “primary non-refluxing megaureter, endoscopic management.” Inclusion criteria were established before the search. Only articles written in the English language and involving greater than one reported pediatric case per publication were included. There were no date restrictions. Two reviewers then independently extracted data from each study using a standardized form that focused on baseline patient demographics, surgical approach, secondary procedures, and associated complications. Inconsistencies between the reviewers' data were resolved by a third independent author.

An effective endoscopic intervention was categorized on an anatomic and functional basis. Anatomic success was defined as complete or significant improvement in hydroureteronephrosis (HUN) on follow-up renal bladder ultrasound (RBUS). Functional success was defined as preserved or improved differential renal function (DRF) and/or no evidence of obstruction on functional assessment (Mag-3 renogram or magnetic resonance urography). Complications were analyzed from the perspective of intention to treat endoscopically.

The quality of evidence of each citation was rated using the Methodological Index for Non-randomized Studies (MINORS). MINORS is a validated 12-item checklist used for evaluating nonrandomized studies, especially those used in a systematic review or meta-analysis.⁵ In this study, 4 of the 12 items were omitted, because none of the articles were comparative studies. Individual scores per each item ranged from 0 to 2, depending on whether the criterion was not reported (0), reported but inadequate (1), or reported and adequate (2). Consequently, eight items were included with a total score ranging from 0 (low quality) to 16 (excellent quality). This systematic review was conducted in accordance with the Preferred Reporting Items for Systematic Reviews and Meta-analyses (PRISMA).⁶ Reporting of the background, search strategy, and methods was also conducted in accordance with the Meta-Analysis of Observational Studies in Epidemiology (MOOSE) guidelines.⁷ Statistical calculations were completed using Microsoft Excel^®.

Results

The literature search yielded 88 unique citations. After applying exclusion criteria, 76 citations were eliminated (Fig. 1). A total of 12 studies were included (Table 1). All were single institution case series. All but one were retrospective.

FIG. 1.

Study selection.

Table 1.

Baseline Demographics of Identified Citations

Study	N	Boys	Girls	Unilateral POM (left/right)	Bilateral POM	Mean age at surgery (months)	Mean follow-up (years)
Garcia-Aparicio, 2015	20	15	5	18 (12/6)	2	14.2	4.1
Capozza, 2015	12	8	4	12 (8/4)	0	8.0	1.8
Bujons, 2015	19	16	3	18 (10/8)	1	17.0	6.9
Shirazi, 2014	10	7	3	10 (9/1)	0	7.1	1.5
Romero, 2014	29	18	11	26 (16/10)	3	4.0^a	3.9^a
Torino, 2012	5	4	1	5 (2/3)	0	8.0	2.0
Christman, 2012	17	10	7	17 (13/4)	0	84.0	3.2
Farrugia, 2010	15	—	—	12 (—/—)	3	6.0^a	3^a
Carroll, 2010	31	23	8	31 (—/—)	0	37^a	3.3^a
Kajbafzadeh, 2007	47	30	17	42 (29/13)	5	44.4	3.3
Angerri, 2007	7	6	1	7 (6/1)	0	14.0	2.6
Castagnetti, 2006	10	6	4	9 (6/3)	1	3^a	1.6^a
Total	222	143	64	207 (111/53)	15	24.6	3.2

Median.

POM = primary obstructive megaureter.

In 11 of the studies, POM was diagnosed based on preoperative RBUS, voiding cystourethrography (VCUG), and Mag-3 renogram. One author diagnosed POM based on magnetic resonance urography alone.⁸ Most studies involved an initial period of conservative management. Indications for proceeding to endoscopic management were relatively consistent between authors and included obstructive drainage pattern on functional assessment in addition to DRF <40%, worsening or persistent HUN, recurrent febrile urinary tract infection (UTI) despite antibiotic prophylaxis, symptoms such as flank pain, or urolithiasis. There was no standardized definition of HUN that diagnosed POM or mandated intervention. Several authors utilized a distal ureteral diameter >10 to 15 mm on ultrasonography as a prognostic indicator suggestive of renal function deterioration.

We identified a total of 222 patients, 69.1% boys and 30.9% girls. There were a total of 237 obstructed renal units, of which 87.3% (207/237) were unilateral with a 67.7% (111/164) left-sided predominance. Mean age at time of surgery was 24.6 months (range 3–84). Initial attempts at endoscopic management were not effective in 10.5% (25/237), either because of an inability to pass the stent endoscopically (22) or difficulty with advancement of the pediatric cystoscope through the urethra (3). Of these patients, 19 were managed with cystotomy and stent placement, whereas 6 subsequently underwent ureteral reimplantation. Effectively completed endoscopic approaches were cystoscopy+high-pressure balloon dilation (HPBD)+Double-J ureteral stent placement in 49.5% (105/212), cystoscopy+incisional ureterotomy+Double-J ureteral stent placement in 27.8% (59/212), cystoscopy+Double-J ureteral stent placement in 18.9% (40/212), and cystoscopy+HPBD+incisional ureterotomy+Double-J ureteral stent placement in 3.7% (8/212) of operations. Mean stent duration was 2.3 months (range 0.25–6).

All study subjects received either an anatomic or functional study; some received both. After a single intervention for all age groups, there was a 69.6% (128/184) anatomic and a 68.0% (117/172) functional success rate. After a second endoscopic intervention, success rates increased to 79.3% (146/184) and 76.7% (132/172) for anatomic and functional endpoints, respectively. For infants (<12 months of age), anatomic and functional success rates after a single intervention were 61.9% (26/42) and 64.8% (57/88), respectively. A second intervention resulted in an anatomic success rate of 69.0% (29/42) and a functional success rate of 73.9% (65/88). For children ≥12 months of age, a single intervention yielded an anatomic success rate of 71.8% (102/142) and a functional success rate of 71.4% (60/84). A second intervention increased success rates to 82.4% (117/142) and 79.8% (67/84) for anatomic and functional endpoints, respectively. Endoscopic retreatment was performed in 15.1% (32/212) of cases.

Complications were generally mild (Clavien–Dindo Grades I–II) and included transient hematuria, UTI (9.7%, 23/237), failure to pass ureteral stent endoscopically (9.2%, 22/237), vesicoureteral reflux (VUR) (5.1%, 12/237), stone formation (2.1%, 5/237), stent migration (1.7%, 4/237), failure to advance pediatric cystoscope through the urethra (1.3%, 3/237), and ureteral perforation (0.8%, 2/237). Four authors routinely screened for VUR. Forty-one (17.3%, 41/237) renal units progressed to ureteral reimplantation. Overall, surgical reintervention rate was 36.7% (87/237, Table 2). Mean follow-up was 3.2 years (range 1.5–6.9).

Table 2.

Overall Reintervention Rates

Secondary procedure	Renal units
Ureteral reimplantation	41
Endoscopic retreatment	32
Subureteral bulking therapy	8
Nephrectomy	2
Ureteroscopy+incision+lithotripsy	2
Ureterostomy	1
Percutaneous nephrostomy	1
Total	87 (36.7%)

According to the MINORS criteria, quality assessment scores ranged from 6 to 12 with a median value of 9, indicative of an average quality of evidence for the included studies (Table 3).

Table 3.

Quality of Evidence of Each Study Rated Based on Methodological Index for Nonrandomized Studies Criteria

Study	Clearly stated aim	Consecutive patients	Prospective data collection	Appropriate endpoints	Unbiased assessment of study endpoint	Appropriate follow-up	Loss to follow-up <5%	Total
Garcia-Aparicio, 2015	2	2	0	2	2	2	2	12
Capozza, 2015	1	2	0	1	1	1	2	8
Bujons, 2015	2	2	0	2	2	2	1	11
Shirazi, 2014	2	2	0	2	1	1	2	10
Romero, 2014	2	2	0	1	1	2	2	10
Torino, 2012	2	2	0	1	1	1	2	9
Christman, 2012	2	1	2	2	2	1	2	12
Farrugia, 2010	2	1	0	1	1	1	2	8
Carroll, 2010	1	1	0	1	0	1	2	6
Kajbafzadeh, 2007	1	2	0	2	1	1	2	9
Angerri, 2007	1	1	0	1	1	1	2	7
Castagnetti, 2006	1	1	0	1	1	1	2	7

Criteria not reported = 0. Criteria reported but inadequate = 1. Criteria reported and adequate = 2.

Discussion

The management of asymptomatic POM has evolved over the years. Initially, a surgery-first approach was advocated.⁴ Today, ∼80% of cases are managed conservatively.^9

–12 Ureteral reimplantation with or without tapering is the gold standard treatment for progressive or persistent POM. It has stood the test of time with a 90% to 95% success rate.¹³ However, open surgery with reimplantation of a dilated ureter into a small bladder can be technically challenging with morbidity ranging from 4% to 25%.^3,4,14,15 The most significant complications are ureteral stricture and VUR.¹ Several authors have also reported transient postoperative voiding dysfunction.^16,17

Endoscopic management offers a less invasive option for treatment. Advantages of the endoscopic approach include no surgical incision, no violation of the bladder, no manipulation of the distal ureteral blood supply, and no need for prolonged catheterization. Furthermore, if endoscopic management should fail, ureteral reimplantation can still be performed. However, the potential need for a second anesthetic for stent removal must be factored into discussions regarding the risks of the endoscopic approach.

The specific endoscopic technique varied for each author. In general, after the induction of general anesthesia and administration of prophylactic antibiotics, cystoscopy was performed using a pediatric rigid cystoscope (8F–10.5F) and a flexible guidewire was placed up into the renal pelvis. For HPBD, a balloon catheter (usually 3F–5F, 2–4 cm balloon length, 3–7 mm inflated balloon diameter) was used to dilate the ureterovesical junction. Distal ureteroscopy was performed to ensure ureteral patency before Double-J ureteral stent placement. For incisional ureterotomy, both Kajbafzadeh et al. and Shirazi and coworkers performed incisions at the 6 o'clock position.^18,19 One group performed an incision at the 12 o'clock position followed by HPBD for stenotic segments 2 to 3 cm (HPBD only for segments <2 cm).⁸ Capozza et al. utilized a cutting balloon if the area of stenosis was unresponsive to HPBD.²⁰ All four authors reported no episodes of iatrogenic VUR. In one series, surgeons placed two indwelling ureteral stents into the treated ureter.⁸ Stent placement was consistently reported to be difficult because of tortuosity of the ureter.

Our review showed that endoscopic treatment of POM is an effective strategy with success rates approaching 70% after initial intervention and 75% to 80% after reintervention. Success rates were highest in children ≥12 months of age. For infants, success rates were slightly lower in the 60% to 75% range. In children <12 months of age, open surgery may be associated with higher morbidity and lower technical success when compared with that in older children.²¹ In this age cohort, endoscopic management may have a more important role as a temporizing rather than definitive procedure. Classically, temporization has been achieved by cutaneous ureterostomy, nephrostomy tubes, or refluxing ureteral reimplantation. Although temporizing ureterostomy accomplishes the goal of urinary diversion, it can be associated with a high rate of stoma stenosis and febrile UTI.^22,23 In addition, the stoma may be cosmetically undesirable to some parents. Refluxing reimplantation trades obstruction for the more manageable VUR. However, ∼80% of patients still require ureteral tapering at the time of definitive repair.²⁴ In 2006, Castagnetti and coworkers reported on 10 infants (median age of 3 months) with 11 obstructed renal units, who underwent stent placement. Five of the 11 renal units progressed to require ureteral reimplantation. Endoscopic treatment postponed open surgery an additional 3 months after stent removal, reaching a median age at time of surgery of 14 months. Of the five ureteral reimplants, none required ureteral tapering.²⁵ Overall, the rates of ureteral tapering in this series were poorly reported, so we were unable to compare the various methods of temporization.

A significant disadvantage to endoscopy in infants is the risk of not being able to pass the stent into the ureter. In our review, failure to pass the ureteral stent occurred in 22 obstructed renal units. Nineteen of the 22 involved infants. One author reported two cases (three obstructed renal units) of failure to pass the pediatric cystoscope into the urethra.²⁵

In patients undergoing ureteral reimplantation, 12% may require repeat surgery.⁴ The surgical reintervention rate presented in this review was significantly higher at 36.7%, with approximately half because of progression to ureteral reimplantation. These results can partially be explained by the possible use of endoscopic treatment as a temporizing procedure in infants. Of the 41 obstructed renal units that required ureteral reimplantation, 20 were in children under the age of 12 months at the time of endoscopy. In addition, this difference in reintervention rate may simply be reflective of the superior efficacy of ureteral reimplantation. However, to put things in perspective, one might consider the example of endoscopic treatment of VUR. The success rates after a single intervention using dextranomer/hyaluronic acid—often used as a first line procedure for all grades of reflux—are not dissimilar to those for endoscopic treatment in the management of POM already described.^26,27

There is concern that manipulation of the ureteral orifice can result in iatrogenic VUR. In our series, the incidence of de novo VUR was 5.1% (12/237). Of these 12 cases, 8 required subureteral endoscopic bulking therapy and 4 progressed to ureteral reimplantation. Four authors routinely screened for de novo VUR postoperatively with VCUG. In 2015, Garcia-Aparicio et al. evaluated VUR 6 months postdilatation of the ureterovesical junction.²⁸ The reported incidence of postoperative VUR was 27% (6/22). Repeat dilatation did not increase the risk of VUR. Of these six ureters, two proceeded to ureteral reimplantation because of the parents wanting a definitive solution. One patient with bilateral “high-grade” VUR was treated with dextranomer/hyaluronic acid and the remaining two ureters had resolution of VUR on follow-up VCUG. Garcia-Aparicio and coworkers concluded that VUR after HPBD was likely a transient process. Overall, the risk of clinically significant VUR appears low and needs only be ruled out in cases of febrile UTI.

To provide the most comprehensive review of the subject, we were purposefully broad in our literature search by including both POM and primary nonrefluxing megaureter. However, we may have introduced bias because of language restriction and by excluding single case reports. Furthermore, the quality of evidence from the included studies was average with retrospective case series design, small sample sizes, significant heterogeneity in primary endpoints, and short follow-up being common. Selection bias is another limitation as only 7 out of 12 studies clearly stated the use of consecutive patients. It is possible that patients selected for endoscopic intervention had radiographic findings that were more likely to result in improvement or resolution of POM with conservative management alone (i.e., more acceptable DRF or stable HUN). Defining success on an anatomic and functional basis was purposefully inclusive, but can also be viewed as a limitation. Ultrasound does introduce intra- and interobserver variability. It is unclear whether the outcomes would be different if success was defined solely on postoperative functional assessment. A subgroup analysis to determine which endoscopic intervention (dilatation, incision, and stenting) is superior was not performed because of concerns of inadequate power and heterogeneity of the included studies.

Conclusions

In conclusion, endoscopic management for persistent or progressive POM in children ≥12 months of age is a minimally invasive alternative to ureteral reimplantation with modest success rates. The procedure cannot be completed in 10% of cases and approximately one-third of patients require surgical reintervention. For infants, endoscopic management may best be utilized and viewed as a temporizing procedure. Prospective, multi-institutional studies with longer follow-up are needed to validate these findings. Studies comparing ureteral reimplantation and endoscopic management are needed to evaluate the role of endoscopy as a first line alternative.

Disclosure

The views expressed in this article are those of the authors and do not reflect the official policy or position of the Department of the Navy, Department of Defense, or the U.S. Government.

Footnotes

Author Disclosure Statement

No competing financial interests exist.

Abbreviations Used

References

Stehr

, Metzger

, Schuster

, et al. Management of the primary obstructed megaureter (POM) and indication for operative treatment. Eur J Pediatr Surg, 2002; 12:32–37.

Hendren

. Operative repair of megaureter in children. J Urol, 1969; 101:491–507.

Hendren

. Complications of megaureter repair in children. J Urol, 1975; 113:238–254.

Peters

, Mandell

, Lebowitz

, et al. Congenital obstructed megaureters in early infancy—Diagnosis and treatment. J Urol, 1989; 142:641–645.

Slim

, Nini

, Forestier

, et al. Methodological index for non-randomized studies (MINORS): Development and validation of a new instrument. ANZ J Surg, 2003; 73:712–716.

Moher

, Liberati

, Tetzlaff

, et al. Preferred reporting items for systematic reviews and meta-analyses: The PRISMA statement. Int J Surg, 2010; 8:336–341.

Stroup

, Berlin

, Morton

, et al. Meta-analysis of observational studies in epidemiology—A proposal for reporting. JAMA, 2000; 283:2008–2012.

Christman

, Kasturi

, Lambert

, et al. Endoscopic management and the role of double stenting for primary obstructive megaureters. J Urol, 2012; 187:1018–1022.

Liu

HYA

, Dhillon

, Yeung

, et al. Clinical outcome and management of prenatally diagnosed primary megaureters. J Urol, 1994; 152:614–617.

10.

Baskin

, Zderic

, Snyder

, et al. Primary dilated megaureter—Long-term follow-up. J Urol, 1994; 152:618–621.

11.

Arena

, Baldari

, Proietto

, et al. Conservative treatment in primary neonatal megaureter. Eur J Pediatr Surg, 1998; 8:347–351.

12.

McLellan

, Retik

, Bauer

, et al. Rate and predictors of spontaneous resolution of prenatally diagnosed primary nonrefluxing megaureter. J Urol, 2002; 168:2177–2180.

13.

Perdzynski

, Kalicinski

. Long-term results after megaureter folding in children. J Pediatr Surg, 1996; 31:1211–1217.

14.

Sripathi

, King

, Thomson

, et al. Primary obstructive megaureter. J Pediatr Surg, 1991; 26:826–829.

15.

Ossandon

, Romanini

, Torre

. A modified technique of ureteroplasty for megaureter in children. J Urol, 2005; 174:1417–1420.

16.

Fung

LCT

, McLorie

, Jain

, et al. Voiding efficiency after ureteral reimplantation—A comparison of extravesical and intravesical techniques. J Urol, 1995; 153:1972–1975.

17.

Lipski

, Mitchell

, Burns

. Voiding dysfunction after bilateral extravesical ureteral reimplantation. J Urol, 1998; 159:1019–1021.

18.

Kajbafzadeh

, Payabvash

, Salmasi

, et al. Endoureterotomy for treatment of primary obstructive megaureter in children. J Endourol, 2007; 21:743–749.

19.

Shirazi

, Natami

, Hekmati

, et al. Result of endoureterotomy in the management of primary obstructive megaureter in the first year of life: Preliminary report. J Endourol, 2014; 28:79–83.

20.

Capozza

, Torino

, Nappo

, et al. Primary obstructive megaureter in infants: Our experience with endoscopic balloon dilation and cutting balloon ureterotomy. J Endourol, 2015; 29:1–5.

21.

Ottolenghi

, Campagnola

, Sulpasso

. Should primary megaureter be treated d'emblee in the neonate. Z Kinderchir, 1998; 43:154–155.

22.

MacGregor

, Kay

, Straffon

. Cutaneous ureterostomy in children—Long-term follow-up. J Urol, 1985; 134:518–520.

23.

Kitchens

, DeFoor

, Minevich

, et al. End cutaneous ureterostomy for the management of severe hydronephrosis. J Urol, 2007; 177:1501–1504.

24.

Kaefer

, Misseri

, Frank

, et al. Refluxing ureteral reimplantation: A logical method for managing neonatal UVJ obstruction. J Pediatr Urol, 2014; 10:824–830.

25.

Castagnetti

, Cimador

, Sergio

, et al. Double-J stent insertion across vesicoureteral junction—Is it a valuable initial approach in neonates and infants with severe primary nonrefluxing megaureter?. Urology, 2006; 68:870–875.

26.

Warchol

, Krzemien

, Szmigielska

, et al. Comparison of results of endoscopic correction of vesicoureteral reflux in children using two bulking substances: Dextranomer/hyaluronic acid copolymer (Deflux) versus polyacrylate-polyalcohol copolymer (Vantris). J Pediatr Urol, 2016; 12:256.e1–256.e4.

27.

Chertin

, Puri

. Endoscopic management of vesicoureteral reflux: Does it stand the test of time?. Eur Urol, 2002; 42:598–606.

28.

Garcia-Aparicio

, Blazquez-Gomez

, de Haro

, et al. Postoperative vesicoureteral reflux after high-pressure balloon dilation of the ureterovesical junction in primary obstructive megaureter. Incidence, management and predisposing factors. World J Urol, 2015; 33:2103–2106.

29.

Bujons

, Saldana

, Caffaratti

, et al. Can endoscopic balloon dilation for primary obstructive megaureter be effective in a long-term follow-up. J Pediatr Urol, 2015; 11:37.e1–37.e6.

30.

Romero

, Angulo

, Parente

. Primary obstructive megaureter: The role of high pressure balloon dilation. J Endourol, 2014; 28:517–523.

31.

Torino

, Collura

, Mele

, et al. Severe primary obstructive megaureter in the first year of life: Preliminary experience with endoscopic balloon dilation. J Endourol, 2012; 26:325–329.

32.

Farrugia

, Steinbrecher

, Malone

. The utilization of stents in the management of primary obstructive megaureters requiring intervention before 1 year of age. J Pediatr Urol, 2011; 7:198–202.

33.

Carroll

, Chandran

, Joshi

, et al. Endoscopic placement of double-J ureteric stents in children as treatment for primary obstructive megaureter. Urol Ann, 2010; 2:114–118.

34.

Angerri

, Caffaratti

, Garat

, et al. Primary obstructive megaureter: Initial experience with endoscopic dilation. J Endourol, 2007; 21:999–1004.