Journal of Pediatric Surgery 50 (2015) 167–170

Contents lists available at ScienceDirect

Journal of Pediatric Surgery

journal homepage: www.elsevier.com/locate/jpedsurg

Pediatric surgical complications of major genitourinary reconstruction in

the exstrophy–epispadias complex
Dylan Stewart a, Brian M. Inouye b, Seth D. Goldstein a, Bhavik B. Shah b, Eric Z. Massanyi b, Heather DiCarlo b,
Adam J. Kern b, Ali Tourchi b, Nima Baradaran c, John P. Gearhart b,⁎
a
The Johns Hopkins University School of Medicine, Division of Pediatric Surgery, Charlotte Bloomberg Children’s Hospital, 1800 Orleans St., Baltimore, MD 21287
b
The Johns Hopkins University School of Medicine, James Buchanan Brady Urological Institute, Division of Pediatric Urology, Charlotte Bloomberg Children’s Hospital, 1800 Orleans St., Suite 7203,
Baltimore, MD 21287
c
Medical University of South Carolina, Department of Urology, 171 Ashley Ave, Charleston, SC 29425

a r t i c l e i n f o a b s t r a c t

Article history: Purpose: Urinary continence is the goal of exstrophy–epispadias complex (EEC) reconstruction. Patients may re-
Received 5 October 2014 quire a continent urinary diversion (CUD) if they are a poor candidate for bladder neck reconstruction or are re-
Accepted 6 October 2014 ceiving an augmentation cystoplasty (AC) or neobladder (NB). This study was designed to identify the incidence
of surgical complications among various bowel segments typically used for CUD.
Key words: Methods: A prospectively kept database of 1078 patients with EEC at a tertiary referral center from 1980 to 2012
Exstrophy–epispadias complex
was reviewed for major genitourinary reconstruction. Patient demographics, surgical indications, perioperative
Continent urinary diversion
Augmentation cystoplasty
complications, and outcomes were recorded.
Neobladder Results: Among reviewed EEC patients, 134 underwent CUD (81 male, 53 female). Concomitant AC was per-
Surgical complications formed in 106 patients and NB in 11. Median follow up time after initial diversion was 5 years. The most common
CUD bowel segments were appendix and ileum. The most common surgical complications after CUD were small
bowel obstruction, post-operative ileus, and intraabdominal abscess. There was a significantly increased risk in
the occurrence of pelvic or abdominal abscess when colon was used as a conduit compared to all other bowel seg-
ments (OR = 16.7, 95% CI: 1.16–239) and following NB creation compared to AC (OR = 39.4, 95% CI: 3.66–423).
At postoperative follow-up, 98% of patients were continent of urine via their stoma.
Conclusion: We report the largest series to date examining CUD in the EEC population. The increased risk of ab-
dominal and pelvic abscesses in patients who receive a colon CUD and undergo NB compared to AC indicates
that while surgical complications following major genitourinary reconstruction are rare, they do occur. Practi-
tioners must be wary of potential complications that are best managed by a multi-disciplinary team approach.
© 2015 Elsevier Inc. All rights reserved.

The ultimate goal of surgical reconstruction in patients with the Augmentation cystoplasty (AC) can be concomitantly performed for
exstrophy–epispadias complex (EEC) is urinary continence. Urethral con- insufficient bladder capacity (less than expected for patient’s age) or a
tinence is often difficult to achieve in patients born with more severe non-compliant bladder [4]. If AC cannot be achieved because the native
forms of EEC (i.e. cloacal exstrophy) because of poor size and bladder tem- bladder template is deemed unsalvageable for being too small (typically
plate quality. However, series following classic bladder exstrophy closure less than 50 cc) or severely thickened, a separate segment of bowel can
have demonstrated stomal continence rates from 67% to 83% [1,2]. While be isolated to form a reservoir, called a neobladder (NB). In this patient
continence rates are lower in patients with cloacal exstrophy, dryness is population, the NB creation is accompanied by construction of a conti-
often attainable by implementation of continent urinary diversion nent stoma [5]. There are many reported variations in the surgical tech-
(CUD) with or without bladder augmentation or neobladder creation [3]. niques and types of bowel segment used for CUD, AC, and NB; however,
A catheterizable CUD is typically required when an EEC patient has the associated complications are not well understood or well described
persistent upper urinary tract changes or is not a candidate for bladder in long-term follow-up [6,7]. This study was designed to report a large-
neck reconstruction. During this procedure, a 2–3 cm segment of bowel volume experience with major genitourinary reconstructive operative
is fashioned into a tube of approximately 12 French internal diameter. strategies in an effort to find any association between different types
The proximal end is a submucosal tunnel into the bladder while the dis- of bowel segment used and subsequent surgical complications.
tal end is matured into a catheterizable stoma at the skin. Continence is
achieved when the bladder fills and collapses the intramural tunnel
preventing leakage from the stoma. 1. Methods

⁎ Corresponding author. Tel.: +1 410 955 2960; fax: +1 410 502 5314. After institutional review board (IRB) approval, 1046 patients who
E-mail address: [email protected] (J.P. Gearhart). underwent a CUD between 1980 and 2012 at a single, tertiary, academic

http://dx.doi.org/10.1016/j.jpedsurg.2014.10.036
0022-3468/© 2015 Elsevier Inc. All rights reserved.
168 D. Stewart et al. / Journal of Pediatric Surgery 50 (2015) 167–170

hospital were identified with Current Procedural Terminology surgery Table 1

billing codes. These patients were cross-referenced with a 1078 patient CUD procedure details.

IRB-approved exstrophy database to identify 134 patients who both had CUD Associated Procedure
a CUD and diagnosis of complete epispadias, classic bladder exstrophy None 17
or exstrophy variant, cloacal exstrophy or cloacal exstrophy variant. Pa- Augmentation 106
tients were excluded if the original CUD tissue was unknown, their orig- Neobladder 11
inal CUD procedure date was unknown, or if they had less than CUD Indication
6 months of follow-up. Failed Primary Exstrophy Closure 36 (27%)
Records were reviewed for patient demographics, surgical history, Failed BNR 31 (23%)
length of follow-up, perioperative outcomes, and bowel segments Insufficient Bladder Capacity 20 (15%)
Neurogenic Bladder 3 (2%)
used for CUD, AC, and NB. Special attention was given to post- Unknown 44 (33%)
operative surgical complications that required surgery or hospitaliza- CUD Bowel Segment
tion and the length of hospital stay for each complication. Fisher’s
Appendix 103 (77%)
exact test for categorical data, Wilcoxon test for non-parametric data, Tapered Ileum 22 (16%)
Chi-squared test for parametric data, and odds ratios were used to de- Colon 3 (2%)
termine statistical significance between complications and types of Appendix & Other Segment 6 (5%)
bowel segments used for the genitourinary reconstruction. All statistical Median Length of Stay (days) (Range)

analysis was performed with Microsoft Excel® 2010 (Redmond, WA) CUD alone (n = 17) 7 (5–22)
and SPSS® 16 (Chicago, IL). CUD at different time of augmentation (n = 9) 7 (6–19)
CUD at same time as augmentation (n = 97) 9 (6–44)
CUD at same time as neobladder (n = 11) 10 (7–49)
2. Results

The authors identified 134 EEC patients (81 male, 53 female) with at
least six months of follow-up who underwent CUD. The median age for CUD compared to all other CUD tissue types (OR = 16.7, 95% CI:
diversion was 7 years (range: 2 years–25 years) and median follow up 1.16–238.53).
was 5 years (range: 6 months–20 years). Among these patients, 4 The subset of 117 CUD patients who also had AC or NB presented
were identified with complete female epispadias, 11 with complete with six events of small bowel obstruction, four events of post-
male epispadias, 96 with classic bladder exstrophy or bladder exstrophy operative ileus, one event of inguinal hernia, and three events of ab-
variant, and 23 with cloacal exstrophy or a cloacal variant. scess. However, there was no correlation between the rates of these
Seventeen patients with appropriate bladder size underwent only complications and the tissue used for reconstructed bladder/reservoir.
CUD, while all others received CUD with AC or NB creation. Indications There were no events of small bowel obstruction, hernia, or post-
for reconstruction included failed primary exstrophy closure, failed operative ileus in the patients who received an NB (Table 4). However,
bladder neck reconstruction resulting in cutaneous fistula formation 4 patients developed pelvic or abdominal abscesses. When comparing
or persistent urinary incontinence, insufficient bladder capacity (less this group to AC patients, there was a significant increase in the rate of
than expected for the patient’s age), and neurogenic bladder. Bowel seg- abscess formation in those with NB (OR = 39.4, 95% CI: 3.66–423.17).
ments used to create the CUD included ileum, colon, and appendix ei-
ther alone or combined with other bowel segments. Fifty-six percent
of the patients’ medical records documented the stoma location, with 3. Discussion
the overwhelming known majority of patients having their stoma at
the umbilicus. Table 1 provides further details regarding the major gen- While traditional EEC management relied on acceptance of simple
itourinary reconstructions. Median hospital length of stay was 7 days. survival, recent advancements have shifted the goal to improved quality
After CUD placement or stomal revision, 98% of the patients were conti- of life measures such as cosmesis and urethral or voided continence
nent via stomal catheterization at most recent follow-up. Two patients [8,9]. The latter is usually achieved with bladder neck reconstruction if
have persistent leakage from their stomas and one patient did not there is sufficient sphincteric tone and bladder capacity [6]. Unfortu-
follow-up after stomal revision. nately, reconstruction is precluded in patients with hydroureter, a
Tissues used for AC (Table 2) include: sigmoid colon, ileum, a combi- small bladder capacity, or unfavorable urodynamic parameters
nation of ileum and sigmoid colon, ureter, and stomach. The tissue type (i.e. poor compliance, elevated detrusor pressure, or poor voiding pres-
used in four patients is unknown. The one patient who underwent sures) [10,11]. Instead, most of these patients will undergo CUD with or
known gastrocystoplasty had the longest length of stay (45 days) without AC or NB creation.
while the mean lengths of stay for ileal and sigmoid cystoplasty were
equal (9 days). Eleven neobladders were constructed either with
colon or a combination of ileum and colon. Eight of these eleven patients
had a Mitrofanoff CUD with their neobladder while three patients had a
Table 2
Monti CUD. Augmentation cystoplasty and neobladder bowel types.
Table 3 shows the CUD post-operative surgical complications. The
Augmentation Cystoplasty Bowel Length of Stay (days)
most prevalent complication was small bowel obstruction, occurring
Type (n = 106)
in 5% of all CUD patients at a median of 21.5 days (range: 14–40)
post-operatively. The occurrence rate of this complication was 5% for Ileum 48 (45%) 9 (7–32)
Sigmoid 45 (42%) 9 (7–24)
both appendiceal and ileal patients. All six of the patients who had a Sigmoid & Ileum 6 (6%) 9 (8–12)
small bowel obstruction underwent an emergent exploratory laparoto- Ureter 1 (1%) 11
my and spent a median of an additional 9.5 days (range: 8–23) after re- Ileum & Ureter 1 (1%) 10
operation. Five CUD patients with prolonged post-operative ileus had a Stomach 1 (1%) 45
Unknown 4 (4%) N/A
longer median length of hospital stay of 21 days (range: 11–26 days)
Neobladder Bowel Type (n = 11) Length of Stay (days)
compared to 9 days (range: 5–49 days) in all other patients (P =
0.03). All five of these patients had a Mitrofanoff CUD. There was a sig- Colon 6 (55%) 9 (7–24)
Colon + Ileum 5 (45%) 11 (9–49)
nificant 16-fold increased risk of abdominal and pelvic abscess in colon
 D. Stewart et al. / Journal of Pediatric Surgery 50 (2015) 167–170 169

Table 3
Surgical complications by CUD type.

Tissue Type N Median Length of Hospital Stay (days) SBO (%) Post-Op Ileus (%) Hernia (%) Abscess (%)

Appendix 103 9 5 (5) 5 (5) 0 (0) 3 (3)

Ileum 22 10 1 (5) 0 1 (5) 0 (0)
Appendix & Other Bowel Segment 6 10.5 0 (0) 0 (0) 0 (0) 0 (0)
Colon 3 11 0 (0) 0 (0) 0 (0) 1 (33)
Total 134 6 (5) 5 (4) 1 (1) 4 (3)
P-Value⁎ 0.931 0.931 0.165 0.014
⁎ P-value calculated by ANOVA comparing each proportion of each complication per tissue type.

CUD has gained popularity over ureterosigmoidostomy. While combination of colon and ileum can be identified, again with the colon’s
ureterosigmoidostomy avoids a stoma and confers continence rates be- tinea serving as an area for ureteral reimplantation. Unfortunately, both
tween 95% and 100% [12,13], it also requires functioning bowel and anal may be unsuitable for AC if there is short mesentery, prior bowel resec-
sphincter, a patent anus, and also confers metabolic acidosis and an in- tion, lack of available colon or short gut syndrome (seen in cloacal
creased risk for malignancy [14]. CUD is also preferred over urethral exstrophy patients), or previous irradiation to the abdomen. In these
catheterization due to comfort and ease of catheterization, as patients cases, stomach or ureter can be used, though each confers different
who intermittently catheterize via stoma have reported an increased complications to the patient [7,28]. No difference was found between
quality of life compared to those who use their native urethra [15]. Con- the observed complications observed and any of these possible tissues
tinence rates after CUD in exstrophy patients after bladder neck transec- used for AC.
tion approaching 100% only reinforce its utility [16,17]. EEC patients may not have bladders amenable to AC and thus require
The authors prefer constructing CUD with appendix due to its anat- cystectomy and NB creation. The significant increase in the risk of pelvic
omy, vascularity that allows for safe manipulation, high continence and abdominal abscess in the NB population reported in this study cor-
rates, and minimization of tissue ischemia leading to stomal stenosis responds to a previous study that reports the higher risk of abscesses
[18,19]. Compared to other CUD, appendix has the longest functional following cystectomy and urinary diversion [29]. Both AC and NB may
profile length [20]. However, appendix may be unavailable or immobile confer increased risk of an abscess if there is an unrecognized anasto-
due to prior appendectomy, small size, short mesentery, or fibrosis. In motic bowel leak or tissue breakdown in the reservoir [30]. NB may
these cases, ileum or colon can be tubularized and used to construct a add additional risk due to an increased chance of leak resulting from
catheterizable channel [21,22]. the complexity of NB and requirement of multiple suture lines during
For the 134 CUD patients in this study, small bowel obstruction was its creation compared to AC. Our results may be complicated by the
the most common complication occurring in 5% of patients. Every ob- small number of NB cases in this study, and the amount of time spent
struction was secondary to adhesions and resolved with lysis. This is during each operation. More studies are needed to examine the forma-
comparable to previously reported rates of bowel obstruction after tion of abscess after NB creation and after colon versus small intestine
major abdominal surgery in children [23,24], though less than the ob- urologic surgeries.
served rates after intestinal stoma formation, which can approach 25% To the best of our knowledge this is the first report relating tissue
[25]. It is normal for adhesions to form in these patients who have un- types used in major genitourinary reconstruction to general surgical
dergone substantial abdominal surgery, but maneuvers including gentle complications. Without any significant difference between appendiceal
handling of organs and use of latex-free gloves can mitigate the natural and ileal CUD and surgical complication, either of these commonly used
history of post-operative healing in the body [26]. techniques is recommended. Furthermore, the similar incidence of the
This study found a significant increase in the occurrence of abdomi- observed surgical complications following ileal and sigmoid colon AC
nal and pelvic abscesses in patients with colon CUD compared to all also permits the use of either tissue. The finding that an NB confers an
other CUD tissue types. Regardless of the tissue being used for CUD, all increased risk of abdominal and pelvic abscess compared to AC may
patients had mechanical and antibiotic bowel prep. The change in be a necessary risk in patients with unsalvageable bladders. While sur-
prep over time from Nichols bowel preparation to one day of Golytely® gical complications are rare following major genitourinary reconstruc-
and only intraoperative intravenous antibiotics may introduce bias, but tion, they exist in both CUD and AC for all tissue types, and therefore
both are known to be sufficient. Still, large intestine is known to be more require management by not only a pediatric urologist, but a multidisci-
heavily colonized with bacteria and could be the cause for the increased plinary team of pediatric surgeons and other consultants.
occurrence of abscesses in the colon CUD patients [27]. More research While this series analyzes CUD and AC outcomes in a non-
must be done with a larger sample size of colonic CUD to strengthen randomized group of selected patients in a retrospective manner,
this finding, but this is another reason why appendix is the preferred tis- there can be selection and analytical biases. Furthermore, the rarity of
sue for CUD. the disease requires many years of referrals and procedures which are
Redundant colon usually lends itself for use in AC. The tinea of the subject to universal advances in perioperative and general medical
bowel is especially useful with ureteral reimplantation. If the patient care. Future research should include the use of a multi-institutional, ran-
has a short colon, which occurs more commonly in cloacal patients, a domized study of CUD patients that also analyzes objective evaluations
of bladder function (such as bladder capacity and urodynamics) and
quality of life.
Table 4
This is the largest study to examine CUD in the EEC population.
Neobladder complications compared to augmentation cystoplasty complications.
The increased risk of abdominal and pelvic abscesses in patients
Complication Neobladder (%) Partial Aug (%) P-Value⁎ who receive a colon CUD compared to ileal CUD and who undergo
n = 11 n = 106 NB instead of AC indicates that while surgical complications follow-
Small Bowel Obstruction 0 (0) 6 (6) 0.422
ing major genitourinary reconstruction are rare, they are possible.
Post-Op Ileus 0 (0) 4 (4) 0.516 Therefore, patients who develop surgical complications following
Hernia 0 (0) 1 (1) 0.906 major genitourinary reconstruction should be managed by a multi-
Abscess 3 (27) 1 (1) 0.002 disciplinary team approach involving pediatric surgeons, pediatric
⁎ Calculated with Chi-Squared Test. urologists, and other consultants.
170 D. Stewart et al. / Journal of Pediatric Surgery 50 (2015) 167–170

References [15] Vajda P, Kispal Z, Lenart I, et al. Quality of life: urinary bladder augmentation or
substitution in children. Pediatr Surg Int 2009;25(2):195–201.
[1] Lottmann H, Melin Y, Lombrail P, et al. Reconstruction of bladder exstrophy: retro- [16] Novak TE, Salmasi AH, Lakshmanan Y, et al. Bladder neck transection for intractable
spective study of 57 patients with evaluation of criteria in favor of the acquisition pediatric urinary incontinence. J Urol 2009;181(1):310–4 [discussion 314].
of continence. Ann Urol (Paris) 1998;32(4):233–9. [17] Capolicchio G, McLorie GA, Farhat W, et al. A population based analysis of continence
[2] Surer I, Baker LA, Jeffs RD, et al. Modified Young–Dees–Leadbetter bladder neck outcomes and bladder exstrophy. J Urol 2001;165(6 Pt 2):2418–21.
reconstruction in patients with successful primary bladder closure elsewhere: a [18] Lorenzo JL, Castillo A, Serrano EA, et al. Urodynamically based modification of
single institution experience. J Urol 2001;165(6 Pt 2):2438–40. Mitrofanoff procedure. J Endourol 1997;11(1):77–81.
[3] Mathews R, Jeffs RD, Reiner WG, et al. Cloacal exstrophy—improving the quality of [19] Duckett JW, Lotfi AH. Appendicovesicostomy (and variations) in bladder reconstruc-
life: the Johns Hopkins experience. J Urol 1998;160(6 Pt 2):2452–6. tion. J Urol 1993;149(3):567–9.
[4] Woodhouse CR, North AC, Gearhart JP. Standing the test of time: long-term outcome [20] Watson HS, Bauer SB, Peters CA, et al. Comparative urodynamics of appendiceal and
of reconstruction of the exstrophy bladder. World J Urol 2006;24(3):244–9. ureteral Mitrofanoff conduits in children. J Urol 1995;154(2 Pt 2):878–82.
[5] Gearhart JP, Peppas DS, Jeffs RD. The application of continent urinary stomas to [21] Ahyai SA, Sayedahmed K, Engel O, et al. Urinary diversion and colon: transverse
bladder augmentation or replacement in the failed exstrophy reconstruction. Br J conduit and transverse pouch. Urologe A 2012;51(7):937–46.
Urol 1995;75(1):87–90. [22] Monti PR, Lara RC, Dutra MA, et al. New techniques for construction of efferent con-
[6] Surer I, Ferrer FA, Baker LA, et al. Continent urinary diversion and the exstrophy– duits based on the Mitrofanoff principle. Urology 1997;49(1):112–5.
epispadias complex. J Urol 2003;169(3):1102–5. [23] Young JY, Kim DS, Muratore CS, et al. High incidence of postoperative bowel obstruc-
[7] Chalmers D, Ferrer F. Continent urinary diversion in the epispadias–exstrophy complex. tion in newborns and infants. J Pediatr Surg 2007;42(6):962–5 [discussion 965].
Semin Pediatr Surg 2011;20(2):102–8. [24] Lautz TB, Raval MV, Reynolds M, et al. Adhesive small bowel obstruction in children
[8] Tourchi A, Hoebeke P. Long-term outcome of male genital reconstruction in child- and adolescents: operative utilization and factors associated with bowel loss. J Am
hood. J Pediatr Urol 2013 Dec;9(6 Pt B):980–9. Coll Surg 2011;212(5):855–61.
[9] Inouye BM, Massanyi EZ, Di Carlo H, et al. Modern management of bladder [25] Grant HW, Parker MC, Wilson MS, et al. Adhesions after abdominal surgery in children.
exstrophy repair. Curr Urol Rep 2013 Aug;14(4):359–65. J Pediatr Surg 2008;43(1):152–6 [discussion 156–7].
[10] Gearhart JP, Canning DA, Jeffs RD. Failed bladder neck reconstruction: options for [26] Brüggmann D, Tchartchian G, Wallwiener M, et al. Intra-abdominal adhesions:
management. J Urol 1991;146(4):1082–4. definition, origin, significance in surgical practice, and treatment options. Dtsch
[11] Hollowell JG, Hill PD, Duffy PG, et al. Evaluation and treatment of incontinence after Arztebl Int 2010;107(44):769–75.
bladder neck reconstruction in exstrophy and epispadias. Br J Urol 1993;71(6): [27] Walter J, Ley R. The human gut microbiome: ecology and recent evolutionary changes.
743–9. Annu Rev Microbiol 2011;65:411–29.
[12] Pahernik S, Beetz R, Schede J, et al. Rectosigmoid pouch (Mainz Pouch II) in children. [28] Youssif M, Badawy H, Saad A, et al. Augmentation ureterocystoplasty in boys with
J Urol 2006;175(1):284–7. valve bladder syndrome. J Pediatr Urol 2007;3(6):433–7.
[13] Miles-Thomas J, Gearhart JP, Gearhart SL. An initial evaluation of pelvic floor function [29] Large MC, Kiriluk KJ, DeCastro GJ, et al. The impact of mechanical bowel preparation
and quality of life of bladder exstrophy patients after ureterosigmoidostomy. on postoperative complications for patients undergoing cystectomy and urinary di-
J Gastrointest Surg 2006;10(4):473–7. version. J Urol 2012;188(5):1801–5.
[14] Smeulders N, Woodhouse CR. Neoplasia in adult exstrophy patients. BJU Int 2001;87 [30] Farnham SB, Cookson MS. Surgical complications of urinary diversion. World J Urol
(7):623–8. 2004;22(3):157–67.