corbett2015

Accepted Manuscript
Outcomes of endopyelotomy for pelviureteric junction obstruction in the paediatric
population: a systematic review
Harriet J. Corbett, Dhanya Mullassery
PII: S1477-5131(15)00358-7
DOI: 10.1016/j.jpurol.2015.08.014
Reference: JPUROL 2045
To appear in: Journal of Pediatric Urology
Received Date: 20 April 2015
Accepted Date: 11 August 2015
Please cite this article as: Corbett HJ, Mullassery D, Outcomes of endopyelotomy for pelviureteric
junction obstruction in the paediatric population: a systematic review, Journal of Pediatric Urology
(2015), doi: 10.1016/j.jpurol.2015.08.014.
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to
our customers we are providing this early version of the manuscript. The manuscript will undergo
copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please
note that during the production process errors may be discovered which could affect the content, and all
legal disclaimers that apply to the journal pertain.

MANUS CRIP T ACCEP TED
ACCEPTED MANUSCRIPT
1
REVIEW
Outcomes of endopyelotomy for pelviureteric junction obstruction in the paediatric population: a
systematic review
Harriet J Corbett * and Dhanya Mullassery 1
Department of Surgery, Alder Hey Children’s Hospital, Liverpool, UK
* Corresponding author. Department of Surgery, Alder Hey Children’s Hospital, Eaton Road, Liverpool
L12 2AP, UK.
E-mail address: harriet.corbet[email protected]
1 Present address. Department of Paediatric Surgery, Addenbrooke’s Hospital NHS Trust, Cambridge
CB2 0QQ, UK.
Summary Introduction: Dismembered pyeloplasty is the gold standard treatment for pelviureteric ob-
struction (PUJO), although endourological techniques also are now used. The outcomes and success rates
of paediatric endopyelotomy are variably reported.
Objective: To systematically analyse published literature to give an overall success rate for en-
dopyelotomy in children.
Study design: Medline and Embase databases were searched to identify reports of paediatric en-
dopyelotomy. Literature reviews, case reports or series of fewer than three children, and adult studies (age
>20 years) were excluded. Data were extracted independently by two authors. Primary and secondary
procedures were considered separately. The procedure was considered successful if (a) the author report-
ed success AND (b) there was no alternate or subsequent surgical procedure to the PUJ 3 or more weeks
after endopyelotomy caused by technical failure.
Results: One-hundred and fourteen studies were assessed for eligibility; 15 were included in the final re-
view. Overall, 220 endopyelotomies were performed in 216 patients; 128 had a primary PUJO, 92 under-
went secondary endopyelotomy. Median success rate was 71% (range 46–100) in the primary group and
75% (25–100) in the secondary group. Previously undetected crossing vessels were found at open pyelo-
plasty in 12 cases (11 primary = 31% of all failed primary endopyelotomies). Complications were report-
ed in 14.8% of primary procedures and 14.1% of secondary procedures.
Conclusions: This study is limited by the data given in the individual series: varied criteria used for pa-
tient selection and outcome as well as inconsistent pre- and postoperative imaging data precluded a meta-
analysis. It is notable that designating procedures as failures if alternate/subsequent procedures were re-
quired, lowered the author-given success rate for some studies. The success rates for endopyelotomy do

MANUS CRIP T ACCEP TED
ACCEPTED MANUSCRIPT
2
not compare favourably with pyeloplasty. Crossing vessels are a significant concern in primary PUJO and
should be excluded before considering the procedure.
KEYWORDS
Introduction
The pelviureteric junction (PUJ) is a common site of dysfunctional drainage within the urinary tract. The
condition is often labelled PUJ obstruction (PUJO) even though there is rarely true obstruction. The gold
standard treatment is dismembered pyeloplasty, a procedure that can be performed laparoscopically or as
an open procedure, and the success rate of this time-honoured operation typically exceeds 94% [1–4].
Minimally invasive ‘endourological’ options that avoid excision of the PUJ include endopyelotomy and
balloon dilatation of the PUJ [4,5]. These procedures have been reported since the 1980s yet success rates
in children are difficult to assess because of small study size, variable operative techniques, and variable
outcome criteria [4]. There are few published studies regarding balloon dilatation of the PUJ in paediatric
patients making a systematic review of this procedure impractical. However, there are sufficient papers
reporting outcomes for endopyelotomy so a systematic review of the English literature has been undertak-
en to evaluate the success rate of endopyelotomy in children.
Methods
Medline and Embase databases were searched using relevant key search terms (pelviureteric junction,
ureteropelvic junction, obstruction, paediatric, pediatric, child$, balloon, dilatation, endopyelotomy, open,
laparoscopic, pyeloplasty). No publication date limits were applied and the final search was performed on
29 September 2014. All retrieved studies were evaluated by reading the abstracts, and full papers were
retrieved for articles that could not be rejected based on abstract alone. References from the retrieved arti-
cles were also searched manually for additional relevant studies. Literature reviews, individual case re-
ports, exclusively adult studies (age >20 years), and small case series of fewer than three paediatric pa-
tients were excluded. Paediatric data from combined adult and paediatric studies were included only if it
was possible to select out the paediatric data. The inclusion of publications was decided by both the study
authors and data were extracted independently by both study authors. The data were collected on a pre-
designed Microsoft excel proforma. The procedure was considered successful if (a) the author reported
success and/or (b) there was no alternate or subsequent surgical procedure (open or endourological)
caused by technical failure 3 or more weeks after the initial procedure. Authors were not contacted for
additional data because of the wide time range of the included studies.
Results
The initial search retrieved 242 titles. Once duplicate records had been excluded (n=128), 114 titles were
assessed for eligibility (Figure 1). Following application of exclusion criteria and removal of duplicate

MANUS CRIP T ACCEP TED
ACCEPTED MANUSCRIPT
3
data series, 15 studies were eligible to be included in the review [6–20], four reporting the results of pri-
mary endopyelotomy [6,10,12,13], seven reporting secondary endopyelotomy [15–20], and five reporting
results for both primary and secondary procedures [7–9,11,14]. Detailed in these studies were 220 en-
dopyelotomies, either performed or attempted in 217 children (2 bilateral, 1 redo); 128 endopyelotomies
were performed for a primary PUJO, and 92 for ‘recurrent’ or secondary PUJO, typically following a
primary pyeloplasty. Data for primary procedures are given in Table 1, and data for secondary procedures
are given in Table 2. Patient age at the time of endopyelotomy ranged from 6.5 weeks to 20 years; from
those studies with available data, the mean age for primary endopyelotomy was 10.9 years and for sec-
ondary endopyelotomy, 7.0 years.
A range of techniques was described for the procedure. Most studies described retrograde place-
ment of a guide-wire across the PUJ then an antegrade approach to the PUJ via a percutaneously placed
sheath with a diathermy hook, cold knife, or laser, aiming for full thickness division followed by deploy-
ment of a stent across the PUJ. More recent series reported use of the Acucise device by both the ante and
retrograde approach. Balloon dilatation following endopyelotomy was also described in some series. Typ-
ically stents were left in-situ for 4–6 weeks. The median success rate was 71% (range 46–100) in the pri-
mary group and 75% (range 25–100) in the secondary group. Complications included urinary tract infec-
tion, blood transfusion, ileus, further procedures to reposition stents or drains, urinary ascites and retro-
peritoneal haematoma, or urinoma causing prolonged admission. Complications were reported in 32 pa-
tients, giving a complication rate of 14.8% for primary procedures and 14.1% for secondary procedures.
Endopyelotomy was unsuccessful in 62 patients (36 primary, 26 secondary) and most underwent
alternate procedures, either immediately or subsequently (Tables 1 and 2). From the studies with data re-
garding management of failed endopyelotomy, technical failure led to an immediate open pyeloplasty in
seven patients (5 primary, 2 secondary), two required exploration for postoperative bleeding that culmi-
nated in open pyeloplasty, 24 underwent open pyeloplasty at a later date (13 primary, 11 secondary), six
had redo endopyelotomy, one failure was managed by ureterocalicostomy, two patients underwent ne-
phrectomy, and four had no further procedure. Previously undetected crossing vessels were found at open
pyeloplasty in 12 patients, 11 following primary endopyelotomy, which equates to 31% of all failed pri-
mary procedures. Mean duration of follow-up was 23 months (range 8.5–50) following primary en-
dopyelotomy and 31 months (range 8.5–61) following secondary procedures (Tables 1 and 2).
Discussion
Dismembered pyeloplasty is a time honoured procedure for primary PUJO for which the risks, benefits,
and outcomes are well described, both in adults and children [1,2,4]. Modern endourological techniques
have allowed application of new approaches to PUJO such as division of the PUJ, or ‘endopyelotomy’
[4,5]. Paediatric endopyelotomy is less well established than dismembered pyeloplasty and most series

MANUS CRIP T ACCEP TED
ACCEPTED MANUSCRIPT
4
are low volume such that comparison with dismembered pyeloplasty is difficult. The purpose of this re-
view was to combine data from published series of endopyelotomy in children to give an overview of
outcomes for this approach to PUJO. Variability in selection criteria, technique, and outcome measures
precluded a formal meta-analysis but the data were suitable for systematic analysis. Although some stud-
ies combined data for treatment of primary PUJO and secondary PUJO, these should be considered as
different entities, thus data for primary and secondary procedures were analysed separately.
Fifteen studies published between 1987 and 2012 are included in the review [6–20]. As distribu-
tion of the data is not normal, the success rates are presented as median and range: in the primary group
median success was 71%, in the secondary group success was 75%. One of the main difficulties with this
review was the varied criteria given by authors for both patient selection and judging successful outcome.
Acceptance of each author’s report of success or failure was considered to be the most practical way to
calculate success rates. However, for two reasons the success versus failure figures used in this review
may be different to those reported in several of the studies. Firstly, although some authors reported those
who required immediate conversion to an open pyeloplasty as failed procedures, not all authors did so.
For the purposes of this review, we treated all such immediately converted procedures as failures on an
‘intention to treat’ basis. Secondly, some authors included any case requiring further intervention as a
failure but, again, not all authors did so; thus for this analysis all cases where a redo endopyelotomy or
another procedure (open or endourological) was reported 3 or more weeks from the initial procedure were
considered failures. This time-scale was used to exclude procedures where technical issues were encoun-
tered but resolved, but to capture those in whom the procedure did not give lasting benefit. It is notable
that in the series by Kim et al., for which the follow-up period is one of the longest reported, the success
of primary endopyelotomy is relatively low and they report a late failure rate which is mostly seen in old-
er patients [14]. This is supported by the results of Nicholls et al. who report a similarly long follow-up
period yet a low success rate [12]. However, further analysis of the influence of age was not possible as
not all studies reported outcomes in this way.
It would be ideal to compare successful and unsuccessful procedures by numerical values such as
t-half drainage times, grade of hydronephrosis, or split function. Unfortunately this was not possible, pri-
marily because comparable data were not given in many of the reports or the data were purely descriptive.
As the studies come from a wide time range and it was noted that investigations have evolved over this
time period, this was another factor precluding meaningful numerical comparison. The problem of com-
paring outcomes applies to any procedure utilised to treat PUJO, where success may be variably described
as symptomatic improvement, improved drainage on renography, improvement of dilatation on USS,
preservation of function, or a combination of these criteria [21]. Ramsden et al.’s paper demonstrates the
difficulties of analysing results for pyeloplasty against various definitions for a successful procedure [21].