542515

The Journal of Urology
Volume 178, Issue 4, Supplement 1, Pages.1563-1827 (October 2007)
1.
This Issue in Pediatric Urology

Pages 1563-1565
Jack S. Elder
Kidney I
2.
Determining the Proper Stent Length to Use in Children: Age Plus 10

Pages 1566-1569
Jeffrey S. Palmer and Lane S. Palmer
3.
Prenatal Magnetic Resonance Imaging Demonstrates Linear Growth of the Human

Fetal Kidneys During Gestation
Pages 1570-1574
Mieke Cannie, Valerie Neirynck, Frederik De Keyzer, Steven Dymarkowski and
Guy A. Bogaert
4.
Minimally Invasive Open Renal Surgery

Pages 1575-1578
Job K. Chacko, Martin A. Koyle, Gerald C. Mingin and Peter D. Furness III
5.
Editorial Comment
Pages 1577-1578
Richard N. Schlussel
6.
Transperitoneal Laparoscopic Pyeloplasty for Primary Repair of Ureteropelvic

Junction Obstruction in Infants and Children: Comparison With Open Surgery
Pages 1579-1583
Lisandro A. Piaggio, Julie Franc-Guimond, Paul H. Noh, Mark Wehry, T. Ernesto
Figueroa, Julia Barthold and Ricardo Gonzlez
Clinical Prize Finalists

7.
Relationship of Varicocele Grade and Testicular Hypotrophy to Semen Parameters

in Adolescents
Pages 1584-1588
David A. Diamond, David Zurakowski, Stuart B. Bauer, Joseph G. Borer, Craig A.
Peters, Bartley G. Cilento Jr., Harriet J. Paltiel, Ilina Rosoklija and Alan B. Retik
8.
Surgical Treatment of Unilaterally Undescended Testes: Testicular Growth After

Randomization to Orchiopexy at Age 9 Months or 3 Years
Pages 1589-1593
Claude Kollin, Bengt Karpe, Ulf Hesser, Tina Granholm and E. Martin Ritzn
9.
Editorial Comment
Page 1593
Thomas F. Kolon
10.
Endoscopic Treatment of Vesicoureteral Reflux Associated With Ureterocele

Pages 1594-1597
Boris Chertin, Nochiparambil Mohanan, Amicur Farkas and Prem Puri
11.
Nerve Sparing Ventral Clitoroplasty: Analysis of Clitoral Sensitivity and Viability

Pages 1598-1601
Jennifer Yang, Diane Felsen and Dix P. Poppas
Guest Lecture
12.
Challenges for the Pediatric Urologic Surgeon in the 21st Century

Pages 1602-1605
William J. Cromie
Outcomes Research
13.
Esthetic Outcomes of Genitoplasty in Males Born With Bladder Exstrophy and

Epispadias
Pages 1606-1610
Brian A. VanderBrink, Jeffrey A. Stock and Moneer K. Hanna
14.
Editorial Comment
Page 1610
Terry W. Hensle
15.
Editorial Comment
Page 1610
Yegappan Lakshmanan
16.
Sexual Function in Adult Patients With Spina Bifida and its Impact on Quality of
Life
Pages 1611-1614
Jenny Lassmann, Francisco Garibay Gonzalez, Jeanne B. Melchionni, Patrick S.
Pasquariello Jr. and Howard M. Snyder III
Bladder I
17.
Urachal Anomalies: A Longitudinal Study of Urachal Remnants in Children and

Adults
Pages 1615-1618
Richard A. Ashley, Brant A. Inman, Jonathan C. Routh, Audrey L. Rohlinger,
Douglas A. Husmann and Stephen A. Kramer
18.
Results of Bladder Neck Reconstruction After Newborn Complete Primary Repair

of Exstrophy
Pages 1619-1622
John P. Gearhart, Andrew Baird and Caleb P. Nelson
19.
Editorial Comment
Page 1622
Richard Grady
20.
A Comparison of the Monti and Casale (Spiral Monti) Procedures

Pages 1623-1627
Jeffrey A. Leslie, Mark P. Cain, Martin Kaefer, Kirstan K. Meldrum, Andrew M.
Dussinger, Richard C. Rink and Anthony J. Casale
21.
Editorial Comment
Page 1627
Antonio Macedo
22.
Endoscopic Injection of Bulking Agents in Children With Incontinence:

Dextranomer/Hyaluronic Acid Copolymer Versus Polytetrafluoroethylene
Pages 1628-1631
Lori Dyer, Israel Franco, Casimir F. Firlit, Edward F. Reda, Selwyn B. Levitt and
Lane S. Palmer
23.
First Stage Approximation of the Exstrophic Bladder in Patients With Cloacal

ExstrophyShould This be the Initial Surgical Approach in all Patients?
Pages 1632-1636
J.C. Thomas, R.T. DeMarco, J.C. Pope IV, M.C. Adams and J.W. Brock III
24.
Editorial Comment
Page 1635
John P. Gearhart
25.
Editorial Comment
Pages 1635-1636
Moneer K. Hanna
26.
Reply by Authors
Page 1636
Basic Science Research Finalists

27.
In Utero Exposure to Benzophenone-2 Causes Hypospadias Through an Estrogen

Receptor Dependent Mechanism
Pages 1637-1642
Michael H. Hsieh, Erin C. Grantham, Benchun Liu, Reginald Macapagal, Emily
Willingham and Laurence S. Baskin
28.
Urothelial Inhibition of Transforming Growth Factor- in a Bladder Tissue

Recombination Model
Pages 1643-1649
Siam Oottamasathien, Karin Williams, Omar E. Franco, Marcia L. Wills, John C.
Thomas, Ali-Reza Sharif-Afshar, Romano T. DeMarco, John W. Brock III, Neil A.
Bhowmick, Simon W. Hayward, et al.
29.
Association of Transforming Growth Factor-1 Gene Polymorphism With

Familial Vesicoureteral Reflux
Pages 1650-1653
Seika Kuroda, Valeria Solari and Prem Puri
Pediatric Urology Medal

30.
A. Barry Belman
Pages 1654-1655
H. Gil Rushton Jr.
Hypospadias/Urethra
31.
Traumatic Obliterative Urethral Strictures in Pediatric Patients: Failure of the Cut

to Light Technique at Long-Term Followup
Pages 1656-1658
Matthew K. Tollefson, Richard A. Ashley, Jonathan C. Routh and Douglas A.
Husmann
32.
Editorial Comment
Page 1658
Richard A. Santucci
33.
Long-Term Patient Reported Outcome of Urinary Symptoms After Hypospadias

Surgery: Norm Related Study In Adolescents
Pages 1659-1662
Kimihiko Moriya, Hidehiro Kakizaki, Hiroshi Tanaka, Takahiko Mitsui, Tsuyoshi
Furuno, Takeya Kitta, Hiroshi Higashiyama, Hiroshi Sano and Katsuya Nonomura
34.
Editorial Comment
Page 1662
Terry W. Hensle
35.
Reply by Authors
Page 1662
36.
Split Prepuce In Situ Onlay Hypospadias Repair: 17 Years of Experience

Pages 1663-1667
S. Sedberry-Ross, B.C. Stisser, C.G. Henderson, H.G. Rushton and A.B. Belman
Trauma/Imaging
37.
The Posterior Sagittal Approach for Recurrent Genitourinary Pathology

Pages 1668-1671
Shumyle Alam, Marc A. Levitt, Curtis A. Sheldon and Alberto Pea
38.
Editorial Comment
Page 1671
Ricardo Gonzlez
39.
Reply by Authors
Page 1671
40.
Dirt Bikes and All Terrain Vehicles: The Real Threat to Pediatric Kidneys
Pages 1672-1674
Hsi-Yang Wu and Barbara A. Gaines
41.
3-Dimensional Magnetic Resonance Genitography: A Different Look at Cloacal

Malformations
Pages 1675-1679
Steven M. Baughman, Randy R. Richardson, Daniel J. Podberesky, Neal C.
Dalrymple and Elizabeth B. Yerkes
42.
Editorial Comment
Pages 1678-1679
Howard M. Snyder III
Vesicoureteral Reflux I
43.
Intravesical Electrotherapy for Neurogenic Bladder Dysfunction: A 22-Year

Experience
Pages 1680-1683
Jennifer A. Hagerty, Ingrid Richards and William E. Kaplan
44.
Editorial Comment
Page 1683
Stuart B. Bauer
45.
Reply by Authors
Page 1683
46.
Predictive Factors of Early Spontaneous Resolution in Children With Primary

Vesicoureteral Reflux
Pages 1684-1688
Matthew J. Knudson, J. Christopher Austin, Zeb M. McMillan, Charles E.
Hawtrey and Christopher S. Cooper
47.
Ipsilateral Ureteroureterostomy in the Surgical Management of the Severely

Dilated Ureter in Ureteral Duplication
Pages 1689-1692
Job K. Chacko, Martin A. Koyle, Gerald C. Mingin and Peter D. Furness III
Penile Curvature
48.
Outcome Analysis of Severe Chordee Correction Using Tunica Vaginalis as a Flap

in Boys With Proximal Hypospadias
Pages 1693-1697
Luis H.P. Braga, Joao L. Pippi Salle, Sumit Dave, Darius J. Bagli, Armando J.
Lorenzo and Antoine E. Khoury
49.
Editorial Comment
Page 1697
A. Barry Belman
50.
Reply by Authors
Page 169
51.
Small Intestinal Submucosa for Corporeal Body Grafting in Severe Hypospadias

Requiring Division of the Urethral Plate
Pages 1698-1701
James M. Elmore, Andrew J. Kirsch, Hal C. Scherz and Edwin A. Smith
52.
Editorial Comment
Page 1701
Peter D. Furness III
Vesicoureteral Reflux II
53.
Multivariate Comparison of the Efficacy of Intraureteral Versus Subtrigonal

Techniques of Dextranomer/Hyaluronic Acid Injection
Pages 1702-1706
Jonathan C. Routh, Yuri Reinberg, Richard A. Ashley, Brant A. Inman, James J.
Wolpert, David R. Vandersteen, Douglas A. Husmann and Stephen A. Kramer
54.
Editorial Comment
Pages 1705-1706
Anthony A. Caldamone
55.
Editorial Comment
Page 1706
William F. Tarry
56.
Histopathological Changes Associated With Dextranomer/Hyaluronic Acid

Injection for Pediatric Vesicoureteral Reflux
Pages 1707-1710
Jonathan C. Routh, Richard A. Ashley, Thomas J. Sebo, David R. Vandersteen,
Jeffrey Slezak and Yuri Reinberg
57.
New Contralateral Vesicoureteral Reflux After Endoscopic Correction of

Unilateral RefluxIs Routine Contralateral Injection Indicated at Initial
Treatment?
Pages 1711-1713
Maria Menezes, Nochiparambil Mohanan, Juliana Haroun, Eric Colhoun and Prem
Puri
58.
Endoscopic Treatment of Moderate and High Grade Vesicoureteral Reflux in

Infants Using Dextranomer/Hyaluronic Acid
Pages 1714-1717
Prem Puri, Nochiparambil Mohanan, Maria Menezes and Eric Colhoun
59.
Editorial Comment
Page 1717
M. David Bomalaski
60.
Editorial Comment
Page 1717
Richard S. Hurwitz
Testis
61.
Scrotal Exploration for Unilateral Nonpalpable Testis

Pages 1718-1721
Warren T. Snodgrass, Selcuk Yucel and Ali Ziada
62.
Incidence of Testicular Ascent in Boys With Retractile Testes

Pages 1722-1725
Andrew A. Stec, John C. Thomas, Romano T. DeMarco, John C. Pope IV, John
W. Brock III and Mark C. Adams
63.
Editorial Comment
Pages 1724-1725
Mark F. Bellinger
64.
Further Evidence for Spontaneous Descent of Acquired Undescended Testes

Pages 1726-1729
S.W. Eijsbouts, S.M.P.F. de Muinck Keizer-Schrama and F.W.J. Hazebroek
65.
Ultrasonography Underestimates the Volume of Normal Parenchyma in Benign

Testicular Masses
Pages 1730-1732
Ashay S. Patel, Brian D. Coley and Venkata R. Jayanthi
Guest Lecture
66.
Doping in Sports: The Pediatric Perspective

Pages 1733-1737
James M. Betts
Voiding Dysfunction
67.
Clinical Urological Outcomes Following Primary Tethered Cord Release in

Children Younger Than 3 Years
Pages 1738-1743
Amanda M. Macejko, Earl Y. Cheng, Elizabeth B. Yerkes, Theresa Meyer, Robin
M. Bowman and William E. Kaplan
68.
Editorial Comment
Pages 1742-1743
Saul P. Greenfield
69.
Editorial Comment
Page 1743
Julian Wan
70.
Treatment of Nocturnal Enuresis in Children With Attention Deficit Hyperactivity

Disorder
Pages 1744-1747
Boris Chertin, Dmitry Koulikov, Wael Abu-Arafeh, Yoram Mor, Ofer Z. Shenfeld
and Amicur Farkas
71.
Spinal Cord Magnetic Resonance Imaging for Investigation of Nonneurogenic

Lower Urinary Tract DysfunctionCan the Yield be Improved?
Pages 1748-1751
K. Afshar, T. Blake, S. Jaffari, A.E. MacNeily, K. Poskitt and M. Sargent
72.
Editorial Comment
Pages 1750-1751
David A. Diamond
73.
Development of an Objective Score to Quantify the Pediatric Cystometrogram

Pages 1752-1757
A.L. MacNeily, M.P. Leonard, P.D. Metcalfe, A. Casale, L. Guerra, P. Steinbok
and H. Garton
74.
Editorial Comments
Pages 1756-1757
David B. Joseph
75.
Editorial Comments
Page 1757
Julian Wan
76.
Office Management of Pediatric Primary Nocturnal Enuresis: A Comparison of

Physician Advised and Parent Chosen Alternative Treatment Outcomes
Pages 1758-1762
Dawn Diaz Saldano, Antonio H. Chaviano, Max Maizels, Elizabeth B. Yerkes,
Earl Y. Cheng, Jennifer Losavio, Sima P. Porten, Christine Sullivan, Kerry F.
Zebold, Jennifer Hagerty, et al.
77.
Editorial Comment
Page 1762
Yves Homsy
78.
Incontinence Symptom Index-Pediatric: Development and Initial Validation of a

Urinary Incontinence Instrument for the Older Pediatric Population
Pages 1763-1767
Caleb P. Nelson, John M. Park, David A. Bloom, Julian Wan, Rodney L. Dunn
and John T. Wei
79.
Editorial Comment
Page 1767
Kourosh Afshar
Bladder II
80.
The Durability of Intravesical Oxybutynin Solutions Over Time

Pages 1768-1770
Julian Wan and Cody Rickman
81.
Tumor in Bladder Reservoir After Gastrocystoplasty

Pages 1771-1774
Miguel Castellan, Rafael Gosalbez, Marcos Perez-Brayfield, Patrick Healey, Ruth
McDonald, Andrew Labbie and Thomas Lendvay
82.
Editorial Comment
Page 1774
Dominic Frimberger
83.
The Use of Botulinum Toxin A Injection for the Management of External

Sphincter Dyssynergia in Neurologically Normal Children
Pages 1775-1780
Israel Franco, Lori Landau-Dyer, Ginger Isom-Batz, Therese Collett and Edward
F. Reda
84.
Editorial Comment
Page 1779
Howard B. Goldman
85.
Editorial Comment
Page 1780
Zoran Radojicic
86.
Reply by Authors
Page 1780
Kidney II
87.
Incidence Based Fetal Urological Counseling Using the Virtual Pediatric Urology
Registry: Importance of Insignificant Fetal Pyelectasis (Sonographically Evident
Renal Pelvis)
Pages 1781-1785
Antonio H. Chaviano, Max Maizels, Elizabeth B. Yerkes, Earl Y. Cheng, Jennifer
Hagerty, Theresa Meyer and William E. Kaplan
88.
Renal Function Outcomes in Patients Treated With Nephron Sparing Surgery for
Bilateral Wilms Tumor
Pages 1786-1790
Dana W. Giel, Mark A. Williams, Deborah P. Jones, Andrew M. Davidoff and
Jeffrey S. Dome
89.
Editorial Comment
Pages 1789-1790
Fernando Ferrer
90.
Which is BetterRetroperitoneoscopic or Laparoscopic Dismembered Pyeloplasty

in Children?
Pages 1791-1795
Stephen J. Canon, Venkata R. Jayanthi and Gregory J. Lowe
91.
Editorial Comment
Page 1795
Pasquale Casale
Miscellaneous
92.
Corporeal Sparing Dismembered Clitoroplasty: An Alternative Technique for

Feminizing Genitoplasty
Pages 1796-1801
Joo L. Pippi Salle, Luis P. Braga, Nicanor Macedo, Nicolino Rosito and Darius
Bagli
93.
Editorial Comment
Page 1801
Howard M. Snyder III
Basic Science Research

94.
Nerve Sparing Ventral Clitoroplasty Preserves Dorsal Nerves in Congenital

Adrenal Hyperplasia
Pages 1802-1806
Dix P. Poppas, Ariella A. Hochsztein, Rebecca N. Baergen, Emily Loyd, Jie Chen
and Diane Felsen
95.
Editorial Comment
Page 1806
Selcuk Yucel
96.
The Effect of Oxybutynin on Structural Changes of the Obstructed Guinea Pig

Bladder
Pages 1807-1812
Jeroen R. Scheepe, Bas W.D. de Jong, Katja P. Wolffenbuttel, Marlous E.
Arentshorst, Petra Lodder and Dirk J. Kok
Clinical Research
97.
Artificial Urinary Sphincters Placed After Posterior Urethral Distraction Injuries in

Children are at Risk for Erosion
Pages 1813-1815
Richard A. Ashley and Douglas A. Husmann
98.
Long-Term Continence Outcomes After Immediate Repair of Pediatric Bladder

Neck Lacerations Extending Into the Urethra
Pages 1816-1818
Jonathan C. Routh and Douglas A. Husmann
99.
Editorial Comment
Page 1818
Richard A. Santucci
100.
The Common Ileal Ureter: A New Technique for Compliant

Ureterocystoplasty
Pages 1819-1823
D. Frimberger, J. Klein and B.P. Kropp
101.
Editorial Comment
Page 1823
Lane S. Palmer
102.
Computational Model for Predicting the Chance of Early Resolution in Children

With Vesicoureteral Reflux
Pages 1824-1827
Matthew J. Knudson, J. Christopher Austin, Moshe Wald, Antoine A. Makhlouf,
Craig S. Niederberger and Christopher S. Cooper
Dx/HA is stable and safe as long as 22 months after

subureteral injection.
PREDICTIVE FACTORS OF EARLY

SPONTANEOUS RESOLUTION OF
REFLUX
TUMOR IN BLADDER RESERVOIR

AFTER GASTROCYSTOPLASTY
An important component of the medical management of

vesicoureteral reflux is the expectation that the reflux
eventually will resolve spontaneously. Knudson et al
(page 1684) from Iowa City, Iowa retrospectively reviewed the medical records and cystograms of 324 children with primary reflux. Significant factors predictive of
resolution within 2 years included bladder volume at
onset of reflux on the initial cystogram of greater than
50% of predicted bladder capacity, age younger than 2
years at diagnosis and history of prenatal hydronephrosis. These factors were particularly true for children with
grades II and III reflux.
Tumor formation years after augmentation cystoplasty

using small or large bowel is well documented but to date
the risk following gastrocystoplasty has been unknown.
Castellan et al (page 1771) from Florida and Washington
retrospectively evaluated the records of 119 patients who
had undergone the procedure between 1989 and 2002. A
malignancy developed in 4 patients 11 to 14 years after
gastrocystoplasty. All 4 patients were boys with a neuropathic bladder. Of the tumors 3 were gastric carcinoma,
1 was a poorly differentiated transitional cell carcinoma
and 3 were at the gastrovesical anastomosis. The tumors
were diagnosed following new onset hydronephrosis (2)
and macroscopic hematuria (2). This study demonstrates
that patients undergoing gastrocystoplasty are at risk for
malignancy, particularly 10 years following the procedure.
EFFICACY OF INTRAURETERAL
VERSUS SUBTRIGONAL INJECTION
DEXTRANOMER/HYALURONIC ACID
TECHNIQUES
A recent technical modification of the STING procedure
is the intraureteral injection of dextranomer/hyaluronic acid (Dx/HA). Routh et al (page 1702) from Minneapolis, Minnesota retrospectively reviewed their experience with 301 children who underwent the STING
procedure and compared the results using the intraureteral versus the standard subureteral injection technique. Multiple surgeons participated in the study.
Success rate with the subureteral injection was 69% vs
81% for the intraureteral injection technique. On multivariate analysis reflux grade and surgeon experience
were the most significant predictors of success. Although the trend indicated that the intraureteral injection technique had a better success rate the results
were not statistically significant.
SURGICAL TREATMENT OF
UNDESCENDED TESTES
Orchiopexy for the undescended testis generally is recommended by age 1 year but evidence supporting that approach is lacking. Kollin et al (page 1589) from Stockholm,
Sweden performed a randomized prospective trial in 155
boys with an undescended testis, of whom half underwent orchiopexy at age 9 months and half at age 3
years. The main outcome was testicular size as measured by sonography. In boys undergoing early orchiopexy the undescended testis-to-descended testis size
ratio increased from 0.68 at 6 months to 0.81 at 4 years.
In contrast, in the boys undergoing late orchiopexy the
size ratio decreased from 0.68 to 0.56 at 5 years despite
orchiopexy. Assuming testicular size is an indirect indication of spermatogenic activity in the prepubertal
testis, this study provides supportive evidence for early
orchiopexy.
HISTOPATHOLOGICAL CHANGES
ASSOCIATED WITH Dx/HA INJECTION
FOR REFLUX
RELATIONSHIP OF VARICOCELE
GRADE AND TESTICULAR
HYPOTROPHY TO SEMEN
PARAMETERS
In children with reflux subureteral Dx/HA injection is

often performed. In another study Routh et al (page
1707) retrospectively reviewed the records of 16 patients who underwent ureteroneocystostomy following
failed Dx/HA injection. Median time from injection to
implant excision was 8 months. Giant cell reaction was
present in 94% of patients and generally was located in
the serosa. Few histological or immunophenotypic features correlated with duration of implantation. The
periureteral inflammatory reaction increased with time
but there was no increase in nuclear turnover or fibrosis in these failed cases. The authors conclude that
0022-5347/07/1784-1563/0
THE JOURNAL OF UROLOGY
Copyright 2007 by AMERICAN UROLOGICAL ASSOCIATION
Indications for surgical correction of varicocele in adolescents have included testicular hypotrophy, pain and large
varicocele. However, these findings have not been correlated with semen analysis. Diamond et al (page 1584)
from Boston, Massachusetts obtained a semen analysis in
57 adolescent males 14 to 20 years old (Tanner V) who
1563
Vol. 178, 1563-1565, October 2007

Printed in U.S.A.
DOI: 10.1016/j.juro.2007.07.068
1564
THIS ISSUE IN PEDIATRIC UROLOGY
had a varicocele. Testicular size was assessed by sonography. Boys with a testicular size differential of greater
than 10% had a significantly lower sperm concentration
and total motile sperm count compared to those with
similar size testes. This difference was even more significant in boys with a testicular size differential of greater
than 20%. No significant differences were observed with
regard to percent normal morphology or varicocele grade.
The authors conclude that a semen analysis should be
performed and monitored closely in adolescent boys who
are Tanner V, and who have a varicocele and testicular
size differential greater than 10%.
sion of its epithelium and fixed to the pubic attachments. The labia minora are reconstructed with preputial Byars flaps. At followup the cosmetic results were
quite satisfactory. The single teenage patient had
maintained clitoral sensation. The authors conclude
that this technique should be considered in girls with
clitoral hypertrophy.
TRANSPERITONEAL LAPAROSCOPIC
PYELOPLASTY FOR PRIMARY
REPAIR OF URETEROPELVIC
JUNCTION OBSTRUCTION
Following correction of ventral penile curvature and graft

placement, there is a risk of subsequent recurrence of
penile curvature. Braga et al (page 1693) from Toronto,
Canada retrospectively reviewed the records of 35 boys
with penoscrotal or perineal hypospadias and ventral
penile curvature of more than 45 degrees who underwent
surgical repair between 1995 and 2004. Ventral penile
grafting was performed most commonly with tunica vaginalis or lyophilized dura. At followup ranging from 1 to 11
years ventral curvature recurred in 5 of 35 patients. Of
these 5 patients 1 had undergone a tunica vaginalis graft
and 4 a dural graft. The authors conclude that penile
grafting with tunica vaginalis is associated with a high
short-term success rate.
Laparoscopic pyeloplasty is often used to treat children

with ureteropelvic junction (UPJ) obstruction. Piaggio et
al (page 1579) from Wilmington, Delaware retrospectively compared 37 children treated with laparoscopic
pyeloplasty to 41 treated with open pyeloplasty. Mean
operative time was longer in the laparoscopic group but
the operative time decreased significantly after the first
15 cases. The success rate and cosmesis in children older
than 14 months were higher in the laparoscopic group.
The authors conclude that laparoscopic pyeloplasty may
provide a higher likelihood of success compared with open
pyeloplsty in children older than 14 months.
RETROPERITONEOSCOPIC VERSUS
LAPAROSCOPIC DISMEMBERED
PYELOPLASTY
Laparoscopic pyeloplasty can be performed through either a
retroperitoneal or transperitoneal approach. Canon et al
(page 1791) from Columbus, Ohio retrospectively compared
their results with these 2 approaches in 49 consecutive
patients. There was no difference in patient age, success
rate, analgesic requirement or hospital stay. The retroperitoneoscopic approach took an average of 55 minutes
longer. The authors prefer the transperitoneal laparoscopic approach because of the larger working space for
suturing, ease of antegrade stent placement and subjective improvement in cosmesis.
CORPORAL SPARING DISMEMBERED

CLITOROPLASTY
The management of clitoral enlargement is controversial and there are numerous techniques. Pippi Salle et
al (page 1796) from Toronto, Canada retrospectively
reviewed their results with a newly described technique for clitoral hypertrophy, which involves dissecting the glans and its neurovascular bundle from the
corpora. The isolated corpus is divided in the midline,
starting at its bifurcation. Each hemi-corpus is rotated
inferiorly and laterally, and placed inside the labioscrotal folds. The glans is reduced by superficial exci-
SEVERE CHORDEE CORRECTION

USING TUNICA VAGINALIS
AS A FLAP
REPORTED OUTCOMES OF URINARY

SYMPTOMS AFTER HYPOSPADIAS
SURGERY
There have been few long-term studies of men who
underwent hypospadias repair. Moriya et al (page
1659) from Sapporo, Japan sent a detailed questionnaire to 33 men 18 to 26 years old who had undergone
hypospadias repair before age 6 years and to 50 age
matched controls. Responses were received from 67% of
the patients and 76% of the control group. A straight
urinary stream was reported by approximately 75% in
both groups. Only 5% to 10% of the subjects in both
groups reported bothersome urinary symptoms. However, 32% of those who had undergone hypospadias
repair had urinary dribbling after voiding compared
with 3% of controls. The dribbling was most severe in
men born with proximal hypospadias.
TRAUMATIC OBLITERATIVE
URETHRAL STRICTURES
In patients with a traumatic posterior urethral distraction injury antegrade or retrograde transurethral incision of the stricture often is used. Tollefson et al (page
1656) from Rochester, Minnesota retrospectively reviewed their experience with the cut to the light technique in 22 boys 3 to 16 years old (median age 11). All
boys had a traumatic posterior urethral stricture less
than 1 cm long. Half of the cases were incised with a cold
knife and half were treated with a KTP laser. An indwelling catheter was left in place for 3 weeks. Stricture recurred in all patients a median of 3 months (range 5 days
to 1 year) after treatment. These boys were then placed

on a regimen of clean intermittent catheterization, which
has been effective in only 2 boys with followup of 3 and 6
years. The remaining patients underwent open end-toend urethroplasty. The authors conclude that in boys
with a short posterior urethral stricture the cut to the
light procedure has a low likelihood of success.
SPINAL CORD MAGNETIC

RESONANCE IMAGING FOR
NONNEUROGENIC LOWER URINARY
TRACT DYSFUNCTION
Magnetic resonance imaging (MRI) has been used to diagnose occult neuropathy in children with nonneurogenic
lower urinary tract dysfunction. Afshar et al (page 1748)
from Vancouver, Canada performed a retrospective review of 80 cases of spinal MRI between 1995 and 2004.
Although none of the patients had a known neurological
abnormality before the MRI the study revealed an abnormality in 6 (7.5%), including intradural arachnoid cyst,
sacral dysgenesis, syrinx and tethered cord. An abnormal
1565
lumbar cutaneous finding was the only variable associated with an abnormal MRI. The authors conclude that
children with severe voiding dysfunction and a normal
lumbosacral spine on physical examination are unlikely
to have an underlying neurological abnormality recognized by MRI.
INCONTINENCE SYMPTOM
INDEX-PEDIATRIC
There are few validated instruments for assessing urinary incontinence symptoms and bother in older children.
Nelson et al (page 1763) from Boston, Massachusetts
developed the incontinence symptom index-pediatric,
which is an 11-item instrument consisting of a domain for
impairment and a domain for symptom severity. The
survey was administered to children 11 to 17 years old
with urinary incontinence and to a control group 2 weeks
apart. This preliminary study provided sufficient consistency to provide validation of the questionnaire.
Jack S. Elder
Supplement Editor
THIS ISSUE IN PEDIATRIC UROLOGY
Kidney I
Determining the Proper Stent Length to Use in Children: Age Plus 10
Jeffrey S. Palmer* and Lane S. Palmer
From the Divisions of Pediatric Urology, Glickman Urological Institute, Cleveland Clinic, Cleveland, Ohio, and Schneider Childrens
Hospital of the North Shore-Long Island Jewish Health System (LSP), New Hyde Park, New York
Purpose: Double-J ureteral stents are used in children following various surgeries. The length of Double-J stent to use is
often determined empirically. An accurate and easy to recall mathematical relationship between stent length and another
parameter would be helpful to a urologist treating children. We determined whether such a relationship exists between
appropriate Double-J stent length and the age of the child.
Materials and Methods: We reviewed the ages of and Double-J stent length used in all children in the last 5 years. Proper
stent length was defined as a gentle curve of the distal coil in the bladder on direct vision and of the proximal coil in the renal
pelvis on fluoroscopy or ultrasound. Data were stratified according to laterality and gender, and then regression analyses
were performed between patient age and stent length.
Results: There were 153 patients who were stented during this period. The surgeries for which a stent was used included
ureteral reimplantation in 37 cases, renal surgery in 57 and stone disease in 59. Patient age was 2 months to 17.9 years.
Regression analyses demonstrated consistent and reliable (r2) relationships between patient age and Double-J stent
length irrespective of gender or laterality. The general relationship is defined by the formula, stent length patient age
in years 10.
Conclusions: To our knowledge this is the first study evaluating the proper length of Double-J stent to use in children. We
determined that simply adding 10 to patient age served as a reliable and accurate rule of thumb to determine Double-J stent
length irrespective of gender or laterality.
Key Words: ureter, stents, endoscopy
ouble-J ureteral stents were first described in 1978

by Finney.1 When compared to the available diversion options of that era, their benefits included management ease, decreased risk of infection and earlier hospital discharge. The use of these stents in children lagged
behind their use in adults because of limited manufacturing
of stents of appropriate diameter and length. However, today JJ stent use is commonplace and it has replaced the use
of externally draining feeding tubes or nephrostomy tubes
by the majority of pediatric urologists. Their primary use in
children is to facilitate ureteral access2 and maintain ureteral patency after ureteral instrumentation (ureteroscopy)
or following repair of the ureteropelvic junction or ureteroneocystostomy, typically with tapering.
Use of a properly sized stent decreases the possibility of
irritative symptoms and stent migration.3,4 However, how
does one determine the accurate stent length to use? In
adults patient height has been used as a predictor of appropriate stent length.5 Other investigators have advocated
direct measurement using a guidewire6 or endocatheter
ruler,5 or measurement of ureteral length from IVP.7 There
is disagreement as to the best method of accurately assessing proper stent length.3,8 10
For the child who requires stenting there are no guidelines to determine proper length. Several physiological parameters have previously been related to patient age, in-
* Correspondence: 1999 Marcus Ave., Lake Success, New York

11042.
0022-5347/07/1784-1566/0
cluding bladder capacity,1113 renal length14,15 and meatal

caliber.16,17 In children age and height are related according
to well established growth curves. We determined whether
the proper stent length in children could be related to patient age. If such a relationship were to exist, this would
potentially provide a simple formula that would be easy to
recall.
MATERIALS AND METHODS
We reviewed the ages of and JJ stent length used in all
children in the last 5 years who were treated by the 2 of us
(JSP and LSP). Indications for placing a ureteral stent were
ureteroscopy, or following ureteral reimplantation or pyeloplasty in select children. Stents were placed cystoscopically
following ureteroscopy or before laparoscopic pyeloplasty.
Stents were placed directly in retrograde fashion during
ureteral reimplantation and in antegrade fashion during
open pyeloplasty. Children with congenitally abnormal ureteral length, ie duplicated collecting system, renal ectopia,
or reimplantation using a Boari flap or psoas hitch, were
excluded.
Proper stent length was defined as a gentle curve of the
distal coil in the bladder on direct vision and of the proximal
coil in the renal pelvis, as determined by renal ultrasound or
fluoroscopy (fig. 1). Stents were exchanged until the proper
length stent was inserted in cases where its length was
deemed to be too short or too long. The length of the final
stent was the one recorded in the data. Data were stratified
according to laterality and gender. Regression analyses were
1566
Vol. 178, 1566-1569, October 2007

Printed in U.S.A.
DOI:10.1016/j.juro.2007.03.191
DETERMINING PROPER STENT LENGTH IN CHILDREN
1567
TABLE 1. Patient age vs stent length

Parameter
No.
All pts
Males
Females
Rt side
Lt side
153
89
64
75
78
R2
Formula
Length
Length
Length
Length
Length
0.9
0.9
1.0
0.9
1.0
(age)
(age)
(age)
(age)
(age)
10.6
10.9
10.0
10.9
10.2
0.88
0.88
0.87
0.86
0.90
17.9 years old. There were 89 boys and 64 girls. Stents were
placed on the left side in 78 cases and on the right side in 75.
Surgeries for which a stent was used included ureteral reimplantation in 37 cases, renal surgery in 57 and ureteroscopy in 59.
Regression analyses demonstrated consistent and reliable (R2) relationships between patient age and JJ stent
length irrespective of gender or laterality (table 1). The
general relationship was shown by the equation, stent
length patient age in years 10.
DISCUSSION
FIG. 1. Plain x-ray shows proper stent length on right side with
gentle proximal curl in renal pelvis and gentle distal curl in bladder.
performed between patient age and stent length. R2 was

calculated as a measure of predictability of the equation of
the regression line. R2 closer to 1 indicates a more linear
relationship, predicted by the equation of the line.
RESULTS
There were 153 patients who required placement of a ureteral stent during this period. The children were 2 months to
To our knowledge this is the first published study establishing a formula to determine the appropriate length of ureteral stent to use in children. The choice of which parameter
to evaluate was height, as had been evaluated in adults, or
the age of the child. In children age and height are related
and age is an easier and more reliable parameter to use. The
standard growth curves used by pediatricians demonstrate a
linear relationship during childhood in males between approximately 1 and 15 years old and in females between
approximately 1 and 13 years old (fig. 2). In this study we
performed regression analysis including all patients and
noted an excellent correlation between the stent length used
and the age of the child (fig. 3). Using the age limits from the
growth curves would only decrease the number of outliers
and tighten this relationship.
The use of height as a parameter to determine proper
stent length in adults is controversial. Attempts have been
FIG. 2. Growth curves for boys (A) and girls (B) from ages 2 to 20 years. Linear relationship (tangent lines) between child age and height holds
for boys 1 to 13 years and girls 1 to 15 years old. Modified from Centers for Disease Control and Prevention: Pediatrics 2002; 109: 45.
1568
made to use ureteral length, as measured on IVP7 or measured endoscopically using a guidewire5 or endocatheter
ruler.6 However, the reliability of these methods has been
called into question (table 2). In effect the most common
method is empiricism based on surgeon experience. In children
age is a more attractive and reliable parameter to use than
height. The height of the child may not be based on a measurement made the day of surgery, but rather on a previous measurement or parental recollection of a previous measurement.
As long as the birthday is known, age is always accurately
determined. There are several examples in the literature relating patient age to various urological measurements. Berger
et al11 and Koff12 independently related expected bladder capacity in children to age plus 2. This value is decreased by 25%
in children with myelodysplasia.13 Ultrasound derived renal
length measurements can be related by age in a cross-section of
the pediatric population14 or in those with myelodysplasia.15
The size of the urethral meatus in boys16 and girls17 can also be
related to patient age.
There are several benefits to using this formula. It is quite
simple and easy to recall, and it carries a high degree of
statistical reliability. Once remembered, recall of this formula
allows the proper selection of a stent that decreases the risk of
ureteral trauma and complications from repeat cystoscopic
placement of wires and stents. Antegrade placement of stents
during open pyeloplasty is facilitated by securing an adequately long stent, thus, avoiding a stent that is short and ends
in the distal ureter or migrates proximal, making it difficult to
remove. This also avoids a stent that is too long and situated in
the bladder neck or urethra, causing significant irritative
symptoms and possible voiding difficulties. In this study children with congenitally short or long ureteral length, such as
those with renal ectopia or duplication, were excluded. This
formula serves as a starting point for stent selection for these
children or those who are tall or short. Start with the standard
age 10 and adjust according to clinical judgment.
There are some limitations with the formula and this study.
Every stent was considered satisfactorily placed and, therefore,
this may have led to bias. While this argument may have
some validity, there were children in whom stents needed to
be replaced because of improper length. In this retrospective
study we used the length of the final stent that was placed for
the purpose of analysis. For children who are of an odd numbered age (1, 3, 5 years old, etc) the formula leads to a stent
length that is not manufactured. We suggest choosing stent
length based on how close the child is to the birthday that
TABLE 2. Available literature on adult patients regarding

selection of proper length ureteral stents
References
Slaton and
Kropp3
Breau and
Norman4
Wills et al5
Herrera et al6
No. Pts/
Ureters
48/48
99/99
30/30
Not listed
Pollack and
Banner7
Pilcher and
Patel8
Not listed
Shah and
Kulkarni9
25/25
Paick et al10
35/41
203/406
Methodology
Length on IVP
vs ht
Length on IVP
vs ht
Length on IVP
Direct endocatheter
measurement
Direct guidewire
measurement
Direct guidewire
measurement
vs ht
Direct catheter
measurement
vs ht
Length on IVP
vs ht
Preference
Unclear
Length on IVP
Length on IVP
Direct measurement
Direct measurement
Ht
Direct measurement
Length on IVP
would lead to an even sized stent or considering whether the

child is of average, above average or below average height, and
then choosing accordingly. Finally, when considering the appropriate length of stent to be used in an infant younger than
1 year, the equation would indicate the use of a 10 cm stent.
However, one must be careful when applying this precept
equally to neonatal and 11-month-old patients. In many of the
youngest patients an 8 cm stent would suffice but they could be
well served with a 10 cm stent. Older infants may require a 10
or 12 cm stent, again depending on the physical stature of the
child at this or any age.
CONCLUSIONS
To our knowledge this is the first study evaluating the
proper length of JJ stent to use in children. We determined
that simply adding 10 to patient age served as a reliable and
accurate rule of thumb to determine JJ stent length irrespective of gender or laterality.
ACKNOWLEDGMENTS
Data were collected at Rainbow Babies and Childrens Hospital, Cleveland, Ohio.
Abbreviations and Acronyms

IVP excretory urogram
JJ Double-J
REFERENCES
1.
FIG. 3. Scatterplot of stent length used in all children vs patient

age.
Finney RP: Experience with new Double-J ureteral catheter

stent. J Urol 1978; 120: 378.
2. Hubert KC and Palmer JS: Passive dilation by ureteral stenting before ureteroscopy: eliminating the need for active
dilation. J Urol 2005; 174: 1079.
3. Slaton JW and Kropp KA: Proximal ureteral stent migration:
an avoidable complication? J Urol 1996; 155: 58.
4. Breau RH and Norman RW: Optimal prevention and management of proximal ureteral stent migration and remigration.
J Urol 2001; 166: 890.
5. Wills MI, Gilbert HW, Chadwick DI and Harrison SC: Which
ureteric stent length? Br J Urol 1991; 68: 440.

6.
Herrera M, Brawerman S, Castaneda WR, Kotula F and Amplatz K: The endocatheter ruler: a useful new device. AJR
Am J Roentgenol 1982; 139: 828.
7. Pollack HM and Banner MP: Percutaneous nephrostomy and
related pyeloureteral manipulative techniques. Urol Radiol
1981; 2: 147.
8. Pilcher JM and Patel U: Choosing the correct length of ureteric
stent: a formula based on the patients height compared
with direct ureteric measurement. Clin Radiol 2002; 57: 59.
9. Shah J and Kulkarni RP: Height does note predict ureteric
length. Clin Radiol 2005; 60: 812.
10. Paick SH, Park HK, Byun SS, Oh SJ and Kim HH: Direct
ureteric length measurement from intravenous pyelography: does height represent ureteric length? Urol Res 2005;
33: 199.
11. Berger RM, Maizels M, Moran GC, Conway JJ and Firlit CF:
Bladder capacity (ounces) equals age (years) plus 2 predicts
12.
13.
14.
15.
16.
17.
1569
normal bladder capacity and aids in diagnosis of abnormal

voiding patterns. J Urol 1983; 129: 347.
Koff SA: Estimating bladder capacity in children. Urology
1983; 21: 248.
Palmer LS, Richards I and Kaplan WE: Age-related bladder
capacity and bladder capacity growth in children with myelomeningocele. J Urol 1997; 158: 1261.
Chen JJ, Pugach J, Patel M, Luisiri A and Steinhardt GF: The
renal length nomogram: multivariable approach. J Urol
2002; 168: 2149.
Sutherland RW, Wiener JS, Roth DR and Gonzales ET Jr: A
renal size nomogram for the patient with myelomeningocele. J Urol 1997; 158: 1265.
Immergut M, Culp D and Flocks RH: The urethral caliber in
normal female children. J Urol 1967; 97: 693.
Litvak AS, Morris JA Jr and McRoberts JW: Normal size of the
urethral meatus in boys. J Urol 1976; 115: 736.
Prenatal Magnetic Resonance Imaging Demonstrates

Linear Growth of the Human Fetal Kidneys During Gestation
Mieke Cannie, Valerie Neirynck, Frederik De Keyzer, Steven Dymarkowski and Guy A. Bogaert*
From the Departments of Urology/Paediatric Urology and Radiology (CM, FDK, SD), Universitaire Ziekenhuizen Leuven, Leuven, Belgium
Purpose: Although fetal kidneys have only 10% of their postnatal blood flow, fetal renal development is essential for normal
postnatal function. In addition, to our knowledge it is unknown whether human fetal kidney development follows a linear or
exponential evolution. We created normative magnetic resonance imaging curves for the kidney prenatal growth pattern in
relation to gestational age.
Materials and Methods: A total of 142 human fetuses at between 20 and 36 weeks of gestation underwent prenatal
magnetic resonance imaging for nonurological pathology (pulmonary, cerebral or maternal abnormalities). Twins, urinary
tract abnormalities, oligohydramnios or anhydramnios were excluded. T2-weighted images (single shot turbo spin-echo) were
used to measure kidney variables. Bipolar and anteroposterior diameters of the 2 kidneys were measured and correlated with
gestational age using linear regression analysis.
Results: A linear relationship of the bipolar and anteroposterior diameters was found according to gestational age in weeks
(p 0.0001). No substantial differences between growth of the left and right fetal kidneys was identified. Bipolar length of the
human fetal kidney increases at an average rate of 1.24 mm per week of gestation, while anteroposterior diameter increases
at 0.57 mm per week of gestation between weeks 20 and 36. This indicates that bipolar diameter increases twice as rapidly
as anteroposterior diameter. Regression analysis curves showed a good fit to the measured data points (anteroposterior and
bipolar measurements R2 0.36 and 0.58, respectively). The CIs of function variables were small, enabling their use as
normative curves.
Conclusions: This prenatal magnetic resonance imaging study demonstrates linear growth of the bipolar and anteroposterior diameters of human fetal kidneys in relation to gestational age. In addition, it is clear that bipolar diameter grows twice
as rapidly as anteroposterior diameter and there is less variability for bipolar diameter. These normative curves may allow
us to detect abnormal fetal evolution.
Key Words: kidney, fetus, reference values, magnetic resonance imaging, abnormalities
rinary tract abnormalities account for 30% to 50% of

all structural abnormalities found at birth1 4 and
they are responsible for approximately 40% of end
stage renal failures in childhood.5,6 Today US is the most
accessible modality and, therefore, the modality of choice for
detecting urinary tract abnormalities on fetal screening during gestation. It has a number of inherent advantages. It
does not use ionizing radiation, can be repeated safely, does
not require sedation and is easy to perform. It also provides
an opportunity for in vivo blood flow examinations of the
kidney.
Although in most cases US is able to detect urinary tract
disorders, in some cases the fetal anatomy may be difficult to
analyze because of maternal obesity, interfering bone structures or oligohydramnios, which is frequently associated
with urinary tract malformations.1,7 In addition, US remains a subjective examination, and images and interpretation depend on examiner experience. For these limitations
Study received approval from the ethics committee on clinical

studies at Universitaire Ziekenhuizen Leuven and Katolieke Universiteit Leuven, Belgium.
* Correspondence: Department of Urology-Pediatric Urology, Universitaire Ziekenhuizen Leuven, Herestraat 49, B-3000 Leuven,
Belgium (telephone: 3216346930; FAX: 3216346931; e-mail:
guy.bogaert@uzleuven.be).
0022-5347/07/1784-1570/0
the recent development of rapid sequence MRI has been

validated as an efficient complementary technique in cases
of inconclusive sonographic findings.1
Similar to US, MRI does not require ionizing radiation
and, therefore, it carries no risk of examination induced
damage. Advantages of prenatal MRI compared to US are
stronger tissue differentiation characteristics and the possibility of visualizing the entire fetus on a single imaging
sequence. In addition, MRI is also feasible in the situation of
oligohydramnios and it has fewer problems with air or bone
interfering with images. MRI images are easier to discuss in
a multidisciplinary meeting compared to US images, which
are more difficult to interpret by others than the examiner.
However, MRI requires sedation and the costs are higher.
Today prenatal MRI can act as a second confirmation of
pathological conditions and, therefore, it can have an important role in the decision to intervene during pregnancy or
prepare for delivery.1 When trying to detect urinary tract
abnormalities, kidney size and growth during gestation are
important parameters.
Thus, it is important to have normative curves of renal
growth pattern of normal fetuses on MRI to allow the diagnosis of renal pathologies as early in gestation as possible.
US size charts for fetal kidneys were made in several previous studies.8 10 However, as mentioned, accurate mea-
1570
Vol. 178, 1570-1574, October 2007

Printed in U.S.A.
DOI:10.1016/j.juro.2007.03.178
PRENATAL MAGNETIC RESONANCE IMAGING OF HUMAN KIDNEYS

surements on US can be limited by concomitant factors. We
created normative curves for MRI of normal prenatal kidney
growth by measuring the kidneys and we assessed the relationship with GA. This may contribute to the early diagnosis
of fetal renal pathology and it could help make a good decision concerning prenatal management of fetuses and postnatal treatment of children with urinary tract abnormalities.
This single institution study was approved by the ethics
committee on clinical studies at our hospital and Katholieke
Universiteit Leuven, Belgium.
Study Subjects and Design
We performed a retrospective chart review of all fetuses that
underwent MRI between 2003 and 2005 because of inconclusive US results. Patients referred for prenatal MRI carried a strong suspicion of anomaly on US or US was indecisive due to imaging difficulties caused by maternal obesity,
air containing structures or bones.
However, our aim was to create normative curves of the
growth pattern of fetal kidneys. Therefore, all MRI examinations of fetuses with renal or urology related pathologies,
including anhydramnios/oligohydramnios and multiple
pregnancies, were excluded. In addition, MRI with incomplete data were excluded. Most fetuses in the study presented with isolated cerebral or pulmonary pathology. Any
suspicion of a syndrome or genetic anomaly was also an
exclusion criterion (see table). Of the total of 284 MRI examinations of 195 fetuses 142 (104 fetuses) were selected.
GA was 20 to 36 weeks. In the study population 17 fetuses
1571
were scanned twice, 5 were scanned 3 times, 1 was scanned

4 times and 2 were scanned 5 times. Interobserver and
intra-observer variation was included to evaluate the reproducibility of measurements.
MRI
MRI was performed on a clinical 1.5 Tesla whole body SonataVision magnetic resonance unit (Siemens Medical Systems, Erlangen, Germany) with gradient switching capabilities of 25 mT/m in 300 microseconds. Maternal sedation
with 0.5 mg flunitrazepam was administered orally 30 minutes before MRI to decrease fetal movements and related
motion artifacts. Patients were positioned in the left lateral
position to prevent supine hypotension syndrome (aortocaval compression) with a combination of a 2-channel phase
array body coil and a spine coil positioned over the lower
pelvic area.
The MRI protocol consisted of only T2-weighted HASTE
(half-Fourier acquired single shot turbo spin-echo) sequences. The geometric parameters of the T2-weighted images consisted of 38 adjacent slides with a 4 mm slice thickness without an intersection gap, a field of view of 380 380
mm, a matrix of 173 256, a response/echo time of
1,000/88 milliseconds, a partial Fourier factor of 5/8, a
resulting pixel resolution of 1.8 1.5 6.0 mm and a
bandwidth of 475 Hz per pixel. T2-weighted images were
acquired in the transverse, coronal and sagittal planes
according to fetal orientation. No breath holding was requested of the patient. The radiologist (MC), who had 4
years of experience, adjusted the field of view, number of
sections and image orientation for each fetus as required
for optimal acquisition of images and measurements. Sequences that were degraded by fetal motion were repeated
with the same parameters. Mean SD examination time
for the prospective study was 20 4 minutes per patient.
Pathological conditions in study fetuses

Pathological Site (specified diagnosis)
Lung:
Congenital diaphragmatic hernia
Congenital cystic adenomatoid malformation
Diaphragm hypoplasia
Agenesis of 1 lung
Cerebral:
Cytomegalovirus
Hydrocephaly
Corpus callosum agenesis
Intravenous bleeding
Cortical necrosis
Myelomeningocele
Aquaeductus stenosis
Colpocephaly
Periventricular leukomalacia
Schizencephalic porencephaly
Lat ventricle dilatation
Chiari II
Sacral teratoma
Trauma
Parietal cleft
Maternal:
Premature pre-term membrane rupture
Placenta accreta
Placenta percreta
Pelvimetry
Urogenital-anal:
Ovarian cyst
Anal atresia
Duplication cyst
Structural abnormality (multicystic cheek structure)
Normal (doubt on US but normal on MRI)
Extra (in utero cardiosurgery)
No. MRIs
60
5
2
1
17
6
6
5
3
3
2
2
1
1
1
1
1
1
1
3
1
1
1
2
1
1
1
11
1
Image Analysis
All images were examined on an Impax picture archiving
and communication system for measuring kidney diameters.
AP and bipolar diameters of the 2 kidneys were measured.
All measurements were performed by a single trained investigator (VN).
AP diameter, defined as the largest AP distance where
the renal pelvis was best visualized, was drawn on the
transverse HASTE images (fig. 1, B). Delineation of the
bipolar diameter, which is the maximum length between
the 2 renal poles, was performed on sagittal or coronal
images (fig. 1, A).
Data and Statistical Analysis
Linear regression analyses were performed to examine the
change in AP and bipolar diameters with GA. Statistical
calculations were performed using Microsoft Excel 9.0
and the Analyze-it (Analyze-it Software Ltd, Leeds, United
Kingdom) software package with p 0.05 considered statistically significant.
RESULTS
Of the 142 MRI examinations in the study 138 bipolar and
103 AP diameters of the right kidney, and 135 bipolar and
101 AP diameters of the left kidney could be delineated. The
1572
FIG. 1. Sample delineations for bipolar (A) and AP (B) views
FIG. 3. Evolution of AP diameters of left (A) and right (B) fetal

kidney with GA.
main reason for the different number of measurements was

the lack of the correct plane for several desired measurements. A Gaussian distribution of residuals indicated the
appropriateness of the linear regression analysis.
The results of interobserver and intra-observer variation
were intraclass bipolar left 0.94 and right 0.95, and AP left
0.92 and right 0.88, interclass bipolar left 0.83 and right
0.74, and AP left 0.89 and right 0.73.
Statistical analysis showed that the best fit between kidney diameter in mm and GA in weeks was a linear relationship. For bipolar diameter highly significant relations with
GA were proved (p 0,0001), including left kidney
diameter 1.15 GA 2.38 (r2 0.58, fig. 2, A), right
kidney diameter 1.33 GA 0.78 (r2 0.61, fig. 2, B) and
AP diameter growth as a function of GA expressed by a
linear relationship (p 0.0001), that is left kidney diameter 0.56 GA 5.50 (r2 0.34, fig. 3, A) and right kidney
diameter 0.58 GA 4.98 (r2 0.39, fig. 3, B).
When comparing the normative curves of the left and
right kidneys, no significant difference was found for AP and
bipolar diameters (fig. 4). Bipolar kidney growth per week
was about twice that of AP kidney growth for the left and
right kidneys (1.15 and 1.33 mm vs 0.56 and 0.58, respectively, fig. 5).
US and they may help with counseling for prenatal or perinatal treatment.5
To detect and compare renal or urological abnormalities
on MRI there is a need for exact and normative curves of
fetal renal growth. To date only semisubjective US normative curves have been available in the literature.8 US remains and will remain the state of the art examination for
detecting prenatal urinary tract abnormalities. However, it
remains a subjective examination that requires a high level
of experience during the examination and in some circumstances the diagnosis is difficult due to interfering factors. In
DISCUSSION
Prenatal MRI of the kidneys of 104 fetuses on a total of 142
MRI examinations resulted in exact and accurate normative
curves of the kidney growth pattern during weeks 20 and 36
of pregnancy. These normative growth curves can be used as
a reference for obtaining additional information on prenatal
FIG. 2. Evolution of bipolar diameters of left (A) and right (B) fetal
kidney with GA.
FIG. 4. Left and right of AP (A) and bipolar (B) diameters
FIG. 5. Bipolar kidney growth per week is about twice that of AP

growth (1.15 and 1.33 mm vs 0.56 and 0.58 for left and right
kidneys, respectively).
addition, US is the most accessible examination, it is noninvasive and it can be repeated easily. However, MRI has
been accepted as a valuable complementary, safe examination for prenatal diagnostics.11
We found a linear relationship between the bipolar and
AP diameters of the fetal kidneys in relation to GA. Figure 3
shows that the best fit between kidney diameter and GA is
linear. This was additionally corroborated by the Gaussian
distribution of the data points around the regression line,
indicating the appropriateness of the linear regression analysis.
We also noted similar and parallel growth of the left and
right fetal kidneys during gestation (fig. 4). This indicates
that with regard to size and shape fetal kidney development
is minimally affected by the surrounding organs. This was
even the case in the 32 fetuses (total of 60 scans) with
congenital diaphragmatic hernia, when kidney growth could
be affected by the surrounding tissues due to intrathoracic
herniation of the abdominal organs.12 However, postnatally
there seems to be disproportional growth of the left and right
kidney, resulting in smaller right kidney by ages 7 to 15
years.13
Surprisingly another interesting finding of our study was
that bipolar growth of the left and right kidneys per week is
about twice that of AP kidney growth (1.15 and 1.33 mm vs
0.56 and 0.58, fig. 5). This proves longitudinal growth of the
fetal kidney rather than thickening.
Measuring fetal kidney size is useful not only to detect
growth retardation or abnormal fetal kidney development,
but also to assess kidney function at birth. To this end the
experimental study by Gubhaju and Black in a baboon
model of human fetal kidney development showed that kidney size correlates with the amount of nephrons and subsequently with kidney function at birth.14 In addition, several
studies of the postnatal kidney are ongoing to allow the
measurement of renal function and drainage, to be compared to standard nuclear scan images to validate the MRI
methods.
In this study it was our goal to measure bipolar and AP
diameters of fetal kidneys between weeks 20 and 36 of
gestation on MRI since these parameters are commonly used
for prenatal US screening evaluation. However, in postnatal
1573
investigations it is common to use renal volumes to assess

the normal evolution or growth of the kidneys. Renal volume
calculation is performed using MRI or CT. It has proved to
be the best estimator of actual renal size and it allows the
best prediction of renal function.15 Several studies show that
measuring renal volume with the voxel count method on
MRI or CT better estimates renal size and function than
using the ellipsoid formula on US.15,16 Measurement of renal volume using the voxel count method also seems to be
more accurate on MRI than on CT.15 Of course, it may be a
prolongation of this study to measure renal fetal volume.
Additional arguments for using and studying the delineation of the 2 perpendicular kidney diameters (bipolar and
AP) in a first phase are 2-fold. 1) The delineation of kidney
diameters is easy to perform with good reproducibility and it
can be done in a short time, even during the scan. 2) Separating the diameter delineations allows the detection of differences between the longitudinal and transversal growth of
the fetal kidney, which would not be possible using the voxel
count method.
In this study we extrapolated fetal renal growth to GA
instead of to a biometric parameter, such as fetal body
volume or weight. Specifically the study of Konje et al
showed that kidney length is a better estimation of unknown GA than the usual biometric parameters of femur
length, crown-body, biparietal, ossification center Mahoney or intestinal development Goldstein evaluated on
US.17 Correlating kidney size with a biometric value such
as fetal body volume could probably provide more distinction of normal vs abnormal renal development than with
GA. The most important reason for this conviction is that
genetic differences and environmental factors can interfere with fetal growth.18,19 These differences have a similar influence on kidney size and fetal body volume, allowing kidney size to be compared to normal kidney size in
fetuses of the same fetal body volume. However, accurate
fetal body weights on the exact day of MRI were not
available.
This study demonstrates the ease and reproducibility of
using diameter to assess kidney size, as performed in the
current study, and to relate it to GA instead of to fetal body
weight. This study was limited due to its retrospective character. Therefore, not all data sets allowed the accurate use of
the voxel count method due to incomplete or tilted scan
volumes, or partial volume effects. In addition, some fetuses
in the study were scanned more than once and, thus, these
measurements included in analyses were not totally independent. To minimize this effect the repeated scans were
done at an interval of at least 24 hours, in a random way and
blinded to earlier results.
CONCLUSIONS
Our results of fetal MRI indicate a significant linear growth
pattern of the fetal kidneys in relation to GA. This normative curve may help determine whether there is pathological
kidney growth during gestation. These data may eventually
be used to predict renal function in premature neonates, so
that we can anticipate neonatal renal problems and support
them better in the future.
1574

AP
CT
GA
MRI
US
anteroposterior
computerized tomography
gestational age
magnetic resonance imaging
ultrasound
REFERENCES
1.
2.
3.
4.
5.
6.
7.
8.
9.
Cassart M, Massez A, Metens T, Rypens F, Lambot MA, Hall

M et al: Complementary role of MRI after sonography in
assessing bilateral urinary tract anomalies in the fetus.
AJR Am J Roentgenol 2004; 182: 689.
Anderson N, Clautice-Engle T, Allan R, Abbott G and Wells
JE: Detection of obstructive uropathy in the fetus: predictive value of sonographic measurements of renal pelvic
diameter at various gestational ages. AJR Am J Roentgenol
1995; 164: 719.
Greig JD, Raine PA, Young DG, Azmy AF, MacKenzie JR,
Danskin F et al: Value of antenatal diagnosis of abnormalities of the urinary tract. BMJ 1989; 298: 1417.
Thomas DF: Fetal uropathy. Br J Urol 1990; 66: 225.
Witzani L, Brugger PC, Hormann M, Kasprian G, CsaponeBalassy C and Prayer D: Normal renal development investigated with fetal MRI. Eur J Radiol 2006; 57: 294.
Woolf AS, Price KL, Scambler PJ and Winyard PJ: Evolving
concepts in human renal dysplasia. J Am Soc Nephrol 2004;
15: 998.
Sivit CJ, Hill MC, Larsen JW, Kent SG and Lande IM: The
sonographic evaluation of fetal anomalies in oligohydramnios between 16 and 30 weeks gestation. AJR Am J Roentgenol 1986; 146: 1277.
Chitty LS and Altman DG: Charts of fetal size: kidney and
renal pelvis measurements. Prenatal Diagn 2003; 23: 891.
Cohen HL, Cooper J, Eisenberg P, Mandel FS, Gross BR,
Goldman MA et al: Normal length of fetal kidneys: sono-
10.
11.
12.
13.
14.
15.
16.
17.
18.
19.
graphic study in 397 obstetric patients. AJR Am J

Roentgenol 1991; 175: 545.
Scott JE, Wright B, Wilson G, Pearson IA, Matthews JN and
Rose PG: Measuring the fetal kidney with ultrasonography.
Br J Urol 1995; 76: 769.
Poutamo J, Vanninen R, Partanen K and Kirkinen P: Diagnosing fetal urinary tract abnormalities: benefits of MRI
compared to ultrasonography. Acta Obstet Gynecol Scand
2000; 79: 65.
Rittler M, Mazzitelli N, Grandi C, Vauthay L, Fuksman R and
Bernal L: Renal enlargement in the fetus and newborn with
congenital diaphragmatic hernia: a refuted hypothesis.
J Pediatr Surg 2000; 35: 442.
Safak AA, Simsek E and Bahcebasi T: Sonographic assessment
of the normal limits and percentile curves of liver, spleen
and kidney dimensions in healthy school aged children.
J Ultrasound Med 2005; 24: 1359.
Gubhaju L and Black MJ: The baboon as a good model for
studies of the human kidney development. Pediatr Res
2005; 58: 505.
Heuer R, Sommer G and Shortliffe LD: Evaluation of renal
growth by magnetic resonance imaging and computerizes
tomography volumes. J Urol 2003; 170: 1659.
Bakker J, Olree M, Kaatee R, de Lange EE, Moons KG, Beutler
JJ et al: Renal volume measurements: accuracy and repeatability of US compared with that of MR imaging. Radiology
1999; 211: 623.
Konje JC, Abrams KR, Bell SC and Taylor DJ: Determination
of gestational age after 24th week of gestation from fetal
kidney length measurements. Ultrasound Obstet Gynecol
2002; 19: 592.
Lampl M, Kuzawa CW and Jeanty P: Growth patterns of heart
and kidney suggest inter-organ collaboration in facultative
fetal growth. Am J Human Biol 2005; 17: 178.
Farley TA, Mason K, Rice J, Habel JD, Scribner R and Cohen
DA: The relationship between the neighbourhood environment and adverse birth outcomes. Paediatr Perinatal Epidemiol 2006; 20: 188.
Minimally Invasive Open Renal Surgery

Job K. Chacko, Martin A. Koyle, Gerald C. Mingin and Peter D. Furness, III*
From the Department of Pediatric Urology, Childrens Hospital and Department of Surgery, University of Colorado Health Science Center,
Denver, Colorado
Purpose: Improved pediatric laparoscopic techniques and instruments have led to the increased popularity of laparoscopic
pyeloplasty and nephrectomy at some centers. The recent trend has compared laparoscopic to open techniques to draw
parallel conclusions from the adult literature that laparoscopic surgery in children offers the same advantages. Historically
open renal surgery in the pediatric population has been done successfully but usually through more traumatic incisions. We
present our experience with minimally invasive open renal surgery.
Materials and Methods: A retrospective review of the last 6 years was performed of consecutive open pyeloplasties for
ureteropelvic junction obstruction and open nephrectomy for multicystic dysplastic kidneys and renal duplication anomalies
at a tertiary hospital for children. Parameters evaluated were patient age at surgery, surgical incision size, operative time,
hospital stay and the need for postoperative narcotics.
Results: A total of 135 patients underwent open renal surgery using an open retroperitoneal flank incision, including 66
younger than 1, 32 who were 1 to 5, 11 who were 5 to 10 and 26 who were older than 10 years. Mean SD operative time
was 101.4 44.7 minutes in patients younger than 1 year, 87.7 39.3 minutes in those 1 to 5 years old, 127.1 62.7 minutes
in those 5 to 10 years old and 127.8 38.4 minutes in those older than 10 years. Incision size for the groups was 1.9 0.61,
1.9 0.72, 3.0 1.3 and 3.8 1.6 cm, respectively. The last 20 patients younger than 1 year who underwent open pyeloplasty
had an incision of between 1 and 1.5 cm. Most incisions were performed through a posterior, subcostal muscle splitting
approach. All patients received postoperative ketorolac. Supplemental narcotics were not required in any patients younger
than 10 years. Of the patients 90% were discharged home in less than 23 hours.
Conclusions: The minimally invasive approach to open renal surgery is a safe and effective treatment choice in pediatric
urology. The procedure can be easily performed through a small incision with minimal morbidity, comparable operative time
and excellent cosmesis without excessive postoperative pain issues, allowing early discharge home. Perhaps this refined open
surgery technique should be the benchmark for comparing new techniques.
Key Words: kidney, nephrectomy, ureteral obstruction, hydronephrosis, abnormalities
e have previously described our experience with

minimally invasive open pyeloplasty.1 The benefit
of laparoscopy has been shown in the adult population and it has clear advantages.2 Laparoscopy in pediatric
urology is increasing for excisional and reconstructive operations.3 8 We reviewed our series of open renal surgery to
describe our success and experience with a minimally invasive open technique.
RESULTS

We retrospectively reviewed the records of open pyeloplasty
for ureteropelvic junction obstruction, and open renal surgery for multicystic dysplastic kidneys and renal duplication
anomalies between 2000 and 2006 at our tertiary institution. Children were evaluated preoperatively with renal ultrasound or computerized tomography to assess the degree
of the pathological condition. Further evaluation included
nucleotide renal scans with furosemide washout, which were
performed to assess and support the need for surgery. An
* Correspondence: Department of Pediatric Urology, Childrens

Hospital and University of Colorado, School of Medicine, 1056 East
19th Ave., B-463, Denver, Colorado 80218 (FAX: 303-864-5572;
e-mail: Furness.Peter@tchden.org).
0022-5347/07/1784-1575/0
indwelling catheter was left in place overnight. Diet was

started when the patients were awake and alert after anesthesia. Patients undergoing pyeloplasty and partial nephrectomy were discharged home with a flank drain, which
was removed within 1 week. Intraoperative and postoperative parameters evaluated were operative time, incision
length, hospital stay and postoperative analgesic use.
A total of 135 patients underwent surgery during the study

period. Table 1 shows the procedural breakdown. Evaluated
parameters and followup data were available on all 135
patients. Mean age at surgery was 3.6 years (range 1 month
to 17 years), including younger than 1 year in 66 patients, 1
to 5 years in 32, 5 to 10 years in 11 and older than 10 years
in 26. Average operative time for all procedures was 105.3
minutes. Average operative time for pyeloplasty was 117.3
minutes and for other renal surgeries it was 93.3 minutes.
Mean SD operative time was 101.4 44.7 minutes in
patients younger than 1 year, 87.7 39.3 minutes in those
1 to 5 years old, 127.1 62.7 minutes in those 5 to 10 years
old and 127.8 38.4 minutes in those older than 10 years.
Incision size in the groups was 1.9 0.61, 1.9 0.72, 3.0
1.3 and 3.8 1.6 cm, respectively (table 2). The last 20
1575
Vol. 178, 1575-1578, October 2007

Printed in U.S.A.
DOI:10.1016/j.juro.2007.05.099
1576
MINIMALLY INVASIVE OPEN RENAL SURGERY

TABLE 1
Surgery
No. Pts
Pyeloplasty
Nephrectomy
Upper pole heminephrectomy
Ureteropyelostomy
91
37
5
2
Total
135
patients younger than 1 year undergoing open renal surgery

had incisions between 1 and 1.5 cm. Most incisions were
performed through a posterior, subcostal muscle splitting
approach. All patients received ketorolac postoperatively.
Supplemental narcotics were not required in any patients
younger than 10 years. Of the patients 90% were discharged
home in less than 23 hours.
Of the 91 patients 40 (44%) had indwelling stents that
required a second procedure for removal. Five children required repeat pyeloplasty secondary to persistent hydronephrosis and/or symptoms for a success rate of 95%. Two
patients had persistent flank pain with hydronephrosis and
they subsequently underwent successful repeat pyeloplasty.
Three children with unimproved asymptomatic hydronephrosis underwent reoperation. Of note, these 3 patients
had a patent anastomotic site at exploration after ureteropelvic junction takedown. Another patient with pyeloplasty
had a percutaneous nephrostomy tube placed for postoperative infection and persistent hydronephrosis. An antegrade
stent was placed. After removal of the stent at 6 weeks
hydronephrosis remained improved 4 years postoperatively.
One patient in the nephrectomy group had a wound abscess,
which required drainage. Nonoperative complications included a retained Penrose drain that migrated internally
but was easily removed at the bedside with a cystoscope and
flexible grasper through the drain site.
DISCUSSION
Improved pediatric laparoscopic techniques and instruments have led to the increased popularity of laparoscopic
pyeloplasty and nephrectomy at some centers. The recent
trend has compared laparoscopic to open techniques to draw
parallel conclusions from the adult literature that laparoscopic surgery in children offers the same advantages.
Bonnard et al reported that hospital stay was shorter for the
laparoscopic vs the open groups for dismembered pyeloplasty (2.4 vs 5 days, p 0.05).3 In addition, a comparison of
robot assisted laparoscopic pyeloplasty to open pyeloplasty
was performed. Lee et al observed that with robotic assistance hospital stay was shorter (2.3 vs 3.5 days, p 0.001)
with significantly decreased narcotic use.4 Recently Kutikov
et al reported the feasibility of laparoscopic pyeloplasty in
patients younger than 6 months with a good success rate
and an average hospital stay of 1.2 days.5
There is no argument that laparoscopic surgery in adults
is associated with a beneficial decrease in LOS.2 Many benefits in adults, including LOS, decreased pain and more
rapid return to activity, are not as easily translated to children because they recover much better than adults regardless of the technique used. The use of open surgery in the
pediatric population is well established. Historically open
renal surgery in the pediatric population has been done
successfully but usually through a more traumatic incision.

There is evidence that major procedures can be done in
children even on an outpatient basis. This was shown in the
past with procedures such as nephrectomy, pyeloplasty and
pyelolithotomy.9-11 Often the need for hospital admission or
readmission is based on parental comfort levels rather than
on patient medical necessity.11
Our experience is that appropriate preoperative parental
counseling with the minimally invasive open renal technique has been successful for performing most of these operations in a 23-hour postoperative hospital stay setting. In
fact, our LOS is significantly less than the time reported in
pediatric laparoscopic vs open renal technique studies. We
believe that the conclusion drawn by these groups that laparoscopic techniques offer an LOS advantage is not justified.
One of the known drawbacks of laparoscopic surgery is
the increased time compared to that of open surgery. The
operative times reported by Bonnard3 and Lee4 et al for their
laparoscopic techniques showed that in the best laparoscopic
hands reconstructive laparoscopic renal techniques require
a learning curve. Bonnard et al reported an operative time of
219 vs 96 minutes for the laparoscopic technique vs the open
technique of pyeloplasty (p 0.05).3 Lee et al reported that
operative time was longer in the robotic assisted laparoscopic technique vs open pyeloplasty (219 vs 181 minutes,
p 0.031).4 These data suggest that a significant learning
curve is inevitable, as shown by the increased operative
times for reconstructive laparoscopic techniques. In our experience with our reconstructive renal procedures, mainly
pyeloplasty, the average time of 117.3 minutes compares
favorably to the reported times for laparoscopic and open
pyeloplasty.
Ku et al compared a retroperitoneal laparoscopic approach with an open approach for extirpative renal needs in
children.6 They again noted that the laparoscopic group had
a shorter hospital stay (2.5 vs 4 days, p 0.018) but with an
extirpative surgery operative time was similar when
matched against that of the open group (150 vs 145 minutes). On the other hand, Hamilton et al found a shorter
hospital stay (22.5 vs 41.3 hours, p 0.03) with a longer
operative time (175.6 vs 120.2 minutes, p 0.01) for their
laparoscopic nephrectomy group vs the open group.7 We
found that the minimally invasive open approach for benign
kidney disease is ideal. Most of our open nephrectomies were
done for multicystic dysplastic kidneys. These kidneys are
typically smaller with a minimal blood supply and they can
be brought through a less than 2 cm incision with minimal
manipulation. In our experience with our extirpative renal
cases the average operative time was 93.3 minutes, which
compares favorably to the reported times of laparoscopic and
open renal extirpative procedures.
Ku et al also examined the learning curve for pediatric
laparoscopic nephrectomy and nephroureterectomy in 20
consecutive children to better understand what is required
TABLE 2
Age (yrs)
No. Pts
Mean SD
Operative Time (mins)
Mean SD
Incision Length (cm)
Younger than 1
15
510
Older than 10
66
32
11
26
101.4 44.7
87.7 39.3
127.1 62.7
127.8 37.4
1.9 .61
1.9 .72
3.0 1.3
3.8 1.6

to become facile with laparoscopic techniques.12 Of note is
that the techniques studied were extirpative in nature and
not reconstructive. They noted that the median time in the
first 10 patients was 181 minutes compared to a median of
125 minutes for the second 10 (p 0.02). There was no
significant difference in blood loss or time to oral intake but
the first 10 patients had a significantly longer hospital stay
(5.4 vs 2.5 days, p 0.009). Piaggio et al noted a similar
outcome for laparoscopic partial nephrectomy with a shorter
operative time in patients as experience increased.8 Median
hospital stay was 2 days for the laparoscopic group.
It is undeniable that the current laparoscopic renal techniques reported require a longer operative time initially.
However, with experience, new instrumentation and refined
laparoscopic techniques one would expect that operative
time would eventually approach that of open operative time.
It is likely that the postoperative parameters measured for
children undergoing extirpative laparoscopic procedure vs
minimally invasive open procedure would be indistinguishable in the future. However, we believe at this time that our
technique of minimally invasive renal surgery might be
more appealing to most pediatric urologists. The presented
technique of open surgery uses instruments, flank exposure,
tissue handling principles and suture techniques that are
already familiar to pediatric urologists. To some extent there
will be a learning curve with minimally invasive open techniques but they will likely be more rapid with a smaller
investment of time and money.
As mentioned, we were able to keep hospital stay to less
than 23 hours in 90% of the patients. Also, our average
operative time for all open surgeries at all ages was 105.3
minutes, which was lower than in the laparoscopic groups
for these different surgeries. By adhering to a strict schedule
of ketorolac and acetaminophen we have found that we are
able to limit narcotic use to children 10 years or older, which
facilitates early discharge home. We believe that this attests
to the minimally invasive nature of our open technique.
In our series we found that 49% of the patients were
younger than 1 year and 72% were younger than 5 years. In
this group average incision length was 1.9 cm. In addition,
operative time was shorter in this group younger than 5
years. Based on these results we believe that there is a
definite role for open renal surgery and there is no advantage to laparoscopic surgery in this age group. At our institution we have performed laparoscopic and robot assisted
laparoscopic renal surgery in patients older than 5 years and
in obese patients. We have found that the open and laparoscopic approaches to renal surgery are effective in these 2
groups. However, since most of our patients present at a
young age and are physically smaller, we think that minimally invasive open surgery is the preferred approach.
CONCLUSIONS
As pediatric urologists gain more expertise with laparoscopic surgery and the advances in technology associated
with it, there is little doubt that the role of such forms of
therapy will expand in the reconstructive renal arena. However, currently the learning curve is steep and expensive.
Thus, until such experience and expertise are achieved and
the results associated with it mirror those of open repairs
standard open surgery, with well placed mini incisions and
1577
careful tissue handling, remains the gold standard to which

these new techniques must be compared.

LOS length of stay
REFERENCES
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
Chacko JK, Koyle MA, Mingin GC and Furness PD III: The

minimally invasive open pyeloplasty. J Pediatr Urol 2006;
2: 368.
Moore RG, Averch TD, Schulam PG, Adams JB II, Chen RN
and Kavoussi LR: Laparoscopic pyeloplasty: experience
with the initial 30 cases. J Urol 1997; 157: 459.
Bonnard A, Fouquet V, Carricaburu E, Aigrain Y and ElGhoneimi A: Retroperitoneal laparoscopic versus open pyeloplasty in children. J Urol 2005; 173: 1710.
Lee RS, Retik AB, Borer JG and Peters CA: Pediatric robot
assisted laparoscopic dismembered pyeloplasty: comparison with a cohort of open surgery. J Urol 2006; 175: 683.
Kutikov A, Resnick M and Casale P: Laparoscopic pyeloplasty
in the infant younger than 6 months: is it technically possible? J Urol 2006; 175: 1477.
Ku JH, Yeo WG, Choi H and Kim HH: Comparison of retroperitoneal laparoscopic and open nephrectomy for benign
renal diseases in children. Urology 2004; 63: 566.
Hamilton BD, Gatti JM, Cartwright PC and Snow BW: Comparison of laparoscopic versus open nephrectomy in the
pediatric population. J Urol 2000; 163: 937.
Piaggio L, Franc-Guimond J, Figueroa TE, Barthold JS and
Gonzalez R: Comparison of laparoscopic and open partial
nephrectomy for duplication anomalies in children. J Urol
2006; 175: 2269.
Elder JS, Hladky D and Selzman AA: Outpatient nephrectomy
for nonfunctioning kidneys. J Urol 1995; 154: 712.
Sprunger JK, Reese CT and Decter RM: Can standard open
urological procedures be performed on an outpatient basis?
J Urol 2001; 166: 1062.
Mohamed M, Hollins G and Eissa M: Experience in performing
pyelolithotomy and pyeloplasty in children on day-surgery
basis. Urology 2004; 64: 1220.
Ku JH, Yeo WG, Kim HH and Choi H: Laparoscopic nephrectomy for renal disease in children: is there a learning
curve? J Pediatr Urol 2005; 40: 1173.
EDITORIAL COMMENT
These authors add their voice to those in our pediatric urology community who dispute the putative advantages of laparoscopy in the pediatric patient. Their short operative times,
incisions and hospital stays are data that on the surface
make a compelling case for open pediatric renal surgery.
However, let us examine the data and arguments a bit more
closely.
Half of all patients in this study were younger than 1
year. It may be that this is why their open renal surgery
outcomes compared so favorably with the laparoscopic outcomes of Lee and Ku et al (references 4 and 6 in article),
whose patients were almost 8 and 4 to 5.5 years old, respectively. In those age matched studies the results were better
in the laparoscopy groups.
Pediatric laparoscopic operative times decrease with experience. However, even if the operative time never equals
that of open surgery, a philosophical question arises. Is it
1578
preferable to do a procedure that consumes a bit less surgeon time or one that delivers less morbidity? Additionally,
while open incisions seem minimal in childhood, scars grow
with time. With growing societal emphasis on cosmetics a
3 cm incision that grows 3 or 4-fold may not be viewed as a
small scar by the adult years later.
The authors add their data to this discussion and force us
all to confront our biases. Although I favor laparoscopy for
pediatric renal surgery, I admit that there is a tipping point,
ie a young age at which the advantages of laparoscopy tend
to recede. Most honest laparoscopists would agree with this
statement. By the same token most candid surgeons would

agree that there is an older age at which laparoscopy is
superior to open surgery. Future studies comparing apples
to apples, ie infant open vs infant laparoscopic surgery,
would hopefully bring the open/laparoscopy line into clearer
focus.
Richard N. Schlussel
Department of Urology
Columbia University Medical Center
New York, New York
Transperitoneal Laparoscopic Pyeloplasty for

Primary Repair of Ureteropelvic Junction Obstruction
in Infants and Children: Comparison With Open Surgery
Lisandro A. Piaggio, Julie Franc-Guimond,* Paul H. Noh, Mark Wehry, T. Ernesto Figueroa,
Julia Barthold and Ricardo Gonzlez
From the Division of Urology, Department of Surgery, A. I. duPont Hospital for Children, Wilmington, Delaware, and Department of
Urology, Thomas Jefferson University, Philadelphia, Pennsylvania
Purpose: Pediatric laparoscopic pyeloplasty to treat ureteropelvic junction obstruction has been reported to decrease
hospitalization stay and the analgesic requirement compared to open pyeloplasty. It is not clear if all age groups benefit from
this procedure. We compared primary laparoscopic and open pyeloplasty in infants and children.
Materials and Methods: We retrospectively reviewed the records of consecutive primary pyeloplasties at a single institution
during a 4.5-year period. Demographic data, body measurements, presentation, indications for surgery, operation type,
surgical time, complications, analgesia requirement, hospital stay and outcome were recorded.
Results: Patients were divided in the open pyeloplasty group of 41 and the laparoscopic pyeloplasty group of 37. The groups
were similar with regard to sex and laterality. There were more patients 14 months or younger in the open pyeloplasty group.
Mean surgical time was longer in laparoscopic pyeloplasty but it decreased significantly after the first 15 cases (each
p 0.001). Hospitalization and postoperative analgesia requirements were similar in the 2 groups. There was a higher
success rate for laparoscopic pyeloplasty in patients older than 14 months (p 0.05). In the open pyeloplasty group there were
more re-interventions as well as a trend toward more complications and readmissions.
Conclusions: Transperitoneal laparoscopic pyeloplasty was performed safely in all pediatric age groups with minimal
morbidity and excellent short-term results. In our experience laparoscopic pyeloplasty in infants and children is more difficult
and time-consuming surgery than open pyeloplasty. However, it may provide a better outcome with fewer complications and
better cosmesis. Prospective studies are needed to confirm these results.
Key Words: ureter, kidney, obstruction, laparoscopy, ureteral obstruction
ince the first report in 1993, LP has rapidly grown in

acceptance for treating adults with UPJO.1 4 The
transperineal or retroperitoneal approach with or
without robotic assistance is reported to decrease hospitalization and analgesic requirements compared to open surgery in children.5,6 Studies comparing open vs transperitoneal LP are sparse and most are in the adult literature.2,7,8
We reviewed our experience with all primary repairs of
UPJO in the last 4.5 years and compared LP to OP.

A retrospective chart review was performed in consecutive
patients undergoing primary repair of UPJO from January
2002 to June 2006. Approval was obtained from the internal
review board at our hospital. Patients with bilateral repair,
solitary kidney or incomplete data were excluded.
We recorded demographic data, weight, body surface
area, BMI, laterality, indications for surgery, presentation,
Study received hospital internal review board approval.

* Current address: Division of Pediatric Urology, Centre Hospitalier Universitaire Sainte-Justine and Universit de Montreal,
Montreal, Quebec, Canada.
Correspondence: A. I. duPont Hospital for Children, 1600 Rockland Rd., Wilmington, Delaware 19899 (telephone: 302-651-5107;
FAX: 302-651-6410; e-mail: ricardo_gonzalez33154@yahoo.com).
0022-5347/07/1784-1579/0
operation type, surgical time, blood loss, drain and stent use,
postoperative analgesic requirements, perioperative complications, length of hospitalization, success rate, the need for
readmission and subsequent procedures.
Patients were admitted to the hospital on the day of
surgery. All patients underwent dismembered pyeloplasty
with or without pelvic reduction. Most open and laparoscopic
procedures were performed by a pediatric urology fellow
(LP, MW or JFG) assisted by a staff member (PHN, JSB,
TEF or RG). Antegrade or retrograde pyelogram was attempted immediately before surgery in the majority of
cases.9
Patients who underwent open surgery were approached
extraperitoneally through a lateral or posterior lumbotomy
incision according to surgeon preference. Repair was performed with 6-zero or 7-zero polydioxanone. Stenting was
performed according to surgeon preference. A perinephric
drain was left in all except 1 open case.
LP was approached transperitoneally. Cystoscopy and
retrograde pyelogram with stent placement were performed
at the beginning of the procedure. In all except 2 cases 3
ports were used, including a 5 mm port for the camera and
2 working 3 or 5 mm ports according to patient size. Intraperitoneal access was obtained using the Bailez technique.10
The renal pelvis was exposed with medial mobilization of the
colon except in 2 patients, in whom the transmesocolonic
1579
Vol. 178, 1579-1583, October 2007

Printed in U.S.A.
DOI:10.1016/j.juro.2007.03.159
1580
LAPAROSCOPIC PYELOPLASTY FOR URETEROPELVIC JUNCTION OBSTRUCTION
approach was used. A percutaneous hitch stitch was placed

in the pelvis and/or ureter to facilitate exposure. Repair was
performed with a running suture of 5-zero or 6-zero polyglactin, or 6-zero polydioxanone. An intra-abdominal round
Jackson Pratt drain was left in all except 3 cases through a
port site.
Children weighing less than 20 kg were prepared with a
U drape to include the abdomen and genitalia in the operative field, permitting cystoscopy and stent manipulation during surgery. A Foley catheter was left indwelling at the end
of the procedure in OP and LP cases. The record of surgical
time was started when cystoscopy or antegrade pyelogram
began and concluded when the dressing was finished.
If not contraindicated, for postoperative pain management all patients were prescribed acetaminophen, a narcotic
and/or ketorolac, which they could receive at 2 to 6-hour
intervals. Patients older than 4 years who were eligible for
patient controlled analgesia received constant narcotic infusion plus additional boluses with doses determined by the
pain service. Otherwise the need for and administration of
analgesics was determined by the nursing staff according to
standard pediatric nursing practice and guided by validated
pain scales, as reported previously.11 Doses of opioids other
than morphine were converted to morphine equivalents.
Acetaminophen doses were not considered.
The success rate was defined as symptom resolution,
marked reduction of hydronephrosis on ultrasonography or
improvement in the drainage curve on diuresis renography.
Only patients with at least 3 months of FU and 2 postoperative ultrasound studies were considered for evaluating outcome. Patients lost to FU were not considered in success rate
analysis.
Statistical analysis was performed using SPSS, version
11.5. For ordinal data Fishers exact test was used. For
continuous variables the nonparametric independent 2-sample Mann-Whitney U tests was used. Pearson correlation
coefficients were determined using log transformed data.
Since our population presented a bimodal distribution, comparison between groups was assessed for the whole group
and after dividing the groups by age using a cutoff of 14
months.
RESULTS
Demographic Data
The 82 patients identified for the study were divided into 2
groups, including 37 who underwent LP and 45 who underwent OP. In the OP group 4 patients were excluded due to
solitary kidney in 2, bilateral repair in 1 and incomplete
TABLE 1. Patient characteristics
No. pts
No. females/males
No. lt/rt side
Average age in months
(range)
No. 14 mos or younger
Av kg wt (range)
Av m ht (range)
Av m2 body surface
(range)
Av kg/m2 BMI (range)
Open
Laparoscopic
p Value
41
9/32
29/12
44
(0.5205)
37
11/26
24/13
61
(0.5216)
0.3
0.4
0.05
24
16
(2.774)
0.9 (0.461.74)
0.6
(0.21.7)
11
24.7
(3.974)
1.1 (0.561.84)
0.9 (0.31.96)
0.05
0.05
0.05
0.05
17
17.8
0.05
(1321)
(1228)
TABLE 2. Presentation
Overall
Prenatal hydronephrosis
Pain/Dietls crisis
Vomiting
Other (urinary tract
infection, hematuria,
abdominal mass)
No. OP
No. LP
p Value
41
28
11/6
1
1
37
17
14/11
3
3
0.03
Not significant
Not significant
records in 1. Table 1 lists patient characteristics. The groups

were similar with regard to sex and laterality. There were
more patients 14 months or younger in the OP group. Mean
age, weight, BMI and body surface area were smaller in this
group.
Indications for Surgery
In patients with prenatally detected hydronephrosis surgery
was favored if at presentation they had severe hydronephrosis (Society for Fetal Urology grade III or IV) and in the
diuretic renogram the affected side had 40% differential
kidney function or less, or it did not respond to furosemide.
In patients with prenatal hydronephrosis under surveillance
indications for surgery were symptoms, unresolving or
worsening hydronephrosis, loss of relative renal function
(less than 40% in the affected kidney) or an obstructive
pattern on diuretic renogram.
Presentation
There were more patients with a diagnosis of prenatal hydronephrosis of those who underwent OP (p 0.05, table 2). In
symptomatic patients the most common complaint was flank
pain (75%), which was often associated with vomiting (52%).
Four patients (12%) in the LP group who presented with symptoms had a history of prenatally detected hydronephrosis.
Table 3 lists surgical time and outcome. Mean surgical
time for LP was longer regardless of patient age or team
experience (p 0.01). Analysis of operative time showed a
significant negative Pearson correlation for LP as experience
increased but no change in OP (r 0.72, p 0.001 and r
0.015, p 0.92, respectively, see figure). Surgical time for
laparoscopic procedures decreased significantly after 15
cases (p 0.01). There was a trend toward a higher success
rate in the laparoscopic group (97% vs 83%). This difference
was statistically significant when patients older than 14
months were compared (p 0.05, table 3).
Intraoperative Radiological
Studies, and Drain and Stent Use
Antegrade or retrograde pyelogram immediately before the
procedure was performed in all except 2 patients in each
group in whom cannulation of the ureter or percutaneous
access to the kidney was not achieved (table 3). More patients in the LP group were stented (p 0.001). Drains were
kept indwelling for a mean of 2.5 and 6.3 days in LP and OP
cases, and stents remained indwelling for 6 and 3 weeks,
respectively.
Intraoperative Complications
There were no conversions to open surgery for LP. There was
no need for blood transfusion in either group. Mean differ-
1581
TABLE 3. Surgical time and outcome

OP
No. primary pyeloplasty
Mean mins procedure duration (range)
Mean mins anesthesia induction recovery (range)
No. intraop pyelogram:
Antegrade
Retrograde
No. nephrostent or Double-J stent
No. lost in FU - short FU
No. failure/% success:
Older than 14 mos
14 Mos or younger
41
144
65
LP
(78230)
(3890)
37
2
19
6
6/83
4/17
2/24
ence in preoperative and postoperative hemoglobin and hematocrit was the same. Two patients in the laparoscopic
group and 1 in the open group experienced intraoperative
bleeding.
Hospital Course, Postoperative Complications,
Rehospitalization and Repeat Intervention
Tables 4 and 5 list hospital stay, the need for rehospitalization, re-interventions and complications. Complications and
readmissions related to recurrent obstruction in patients
with failed pyeloplasty were not considered. Hospitalization
negatively correlated with experience with LP but not with
OP (r 0.38, p 0.05 and r 0.047, p 0.77, respectively).
Analgesia Requirement
A caudal block or epidural catheter was used in 23 OP cases
but in no LP cases (p 0.001). The postoperative intravenous narcotic requirement was similar in the 2 groups. Ketorolac administration was higher in the OP group but it
was not statistically significant (table 6). Postoperative intravenous narcotic use decreased with time in the LP group
but not in the OP group (r 0.35, p 0.05 and r 0.16, p 0.3,
respectively).
37
278
67
p Value
(156413)
(33120)
0.001
Not significant
0.001
Not significant
0.11
0.05
Not significant
35
37
7
1/97
0/26
1/11
FU
Mean FU for OP and LP was 24 (range 3 to 48) and 6.3
months (range 1 to 24), respectively. Seven patients in the
LP group (short FU in 5 and 2 lost to FU) and 6 in the OP
group (lost to FU) were not considered in the success rate
analysis. All patients considered a success after pyeloplasty
had significant improvement in hydronephrosis and they
were symptom-free except 3 with OP and 1 with LP, who had
persistent significant calicectasis on ultrasonography. Repeat diuretic renogram showed improved drainage curve in
all 4 cases.
DISCUSSION
We analyzed 82 consecutive cases (4 were excluded) of primary UPJO repair performed at the same institution in a
4.5-year period. During the beginning of the laparoscopic
experience only patients 5 years or older underwent repair
by this method. Our historical OP control was skewed toward smaller patient size and age. In the LP group we
included 9 patients (24%) who were 6 months or younger.
Each group included patients in whom surgery was done by
the same surgeons. For the stated reasons we believe that
the comparison between the groups is valid. Although this
study is limited by its nature, it supports the efficacy and
safety of LP in infants and children, and serves as a basis for
designing future studies.
Since first report of LP in adults12,13 and children,14
minimally invasive surgery has evolved and the same if not
better results can be achieved with smaller instruments and
fewer port sites. We can now perform LP in small children
and infants without compromising patient safety.1518 The
question that remains is whether there is an advantage of
LP over open surgery in any age group. Comparative studies
TABLE 4. Hospitalization, rehospitalization and complications
No. pts
Mean days hospitalization
(range)
No. postop complications (%)
No. rehospitalized
No. secondary procedure
Linear regression analysis of operative time revealed significant negative Pearson correlation since experience increased for LP (r 0.72,
p 0.001) but there was no change for OP (r 0.015, p 0.92).
OP
LP
41
2.5 (19)
37
2.4 (15)
9
3*
7
(35)
3
0
0
(36)
p Value
0.63
0.06
0.11
0.01
Complications and hospitalizations related to obstruction in patients with

failed pyeloplasty were not considered.
* Urinary leakage with associated febrile urinary tract infection in 2
patients and poor oral intake in 1.
Stent placement in 5 patients, nephrostomy placement in 1 and reintervention for bleeding in 1.
1582
using transperitoneal robotic assisted LP for primary UPJO

repair in the pediatric population show a decreased hospital
stay and postoperative narcotic requirement compared to
those of open procedures.6,19 Similar results were reported
by Bonnard et al using the retroperitoneal approach.5 In our
study these differences were not apparent but hospitalization in the OP group in these 2 studies6,19 was longer than in
our series of open procedures. The narcotic requirement in
the study from Boston was higher for each group than in our
study but they did not report ketorolac use, which may in
part be responsible for that difference.6 We also noted that,
when the pain service was consulted, our patients were
prone to receive more narcotics regardless of the type of
operation, which may have also been their experience since
54% of their patients with OP received epidural analgesia.
More than half of the patients whom we included in the
OP group were younger than 1 year. These patients usually
recover more rapidly and require less postoperative narcotics. In the study from Boston mean patient age for the open
and laparoscopic procedures was higher6 than in our study.
The same was true in the series from Paris, in which patients younger than 2 years were not offered LP.5 This may
explain the differences in results among these studies. When
we compared patients younger than 15 months in the 2
groups, median hospital stay for LP and OP was 1 and 2
days, respectively, and narcotic use was the same with more
patients in the OP group receiving a caudal block
(p 0.001). In older patients there was a trend toward fewer
postoperative intravenous narcotic and ketorolac requirements for LP, and hospital stay for this group decreased
with experience but remained the same for the OP group.
Another explanation for the differences compared with
other reports is that, since our laparoscopic experience is
short, we were conservative in the treatment of these cases,
which were routinely stented, drained and left with a Foley
catheter. Others have reported a different approach.6,15 It
can be hypothesized that not leaving a drain could have
expedited hospital discharge in the LP group since many
patients meet the criteria that we used to send patients
home after OP on postoperative day 1. Rather, they remained hospitalized for observation of the drain output for
at least 24 hours after Foley catheter removal.
TABLE 5. Perioperative complications
No. pts
No. complications:
Bleeding
Leakage
Febrile urinary tract infection
Ileus
Wound seroma
OP
LP
10
10
2
4
3
2
1
5
5
2
1
1
1
0
Profuse bleeding at nephrostent placement stopped after 350 ml blood loss

in 1 patient and bleeding from the drain site in 1 needed immediate
postoperative re-intervention in the OP group, bleeding from a crossing
vessel in 1 was controlled with a clip (postoperative renogram showed no
change in differential kidney function) and injury to the main renal vein in
1 was repaired with suturing after placement of 2 additional working ports
to clamp the vein in the LP groups, leakage was managed with stent
placement, Foley catheter and nephrostomy in 1 each, and it stopped
spontaneously in 1 in the OP group, leakage resolved with Foley catheter
replacement in the LP group, urinary tract infection in 3 needed procedure
to improve drainage and 2 also had urinary leakage in the OP group, and
urinary tract infection was documented after stent removal in the LP
group.
TABLE 6. Analgesia requirement
No. pts
No. caudal/epidural
Postop intravenous morphine
(mg/kg)
Intravenous ketorolac (mg/kg)
OP
LP
p Value
41
23
0.27
37
0
0.28
0.001
Not significant
0.6
0.36
Not significant
In contrast to the Boston study, in which removal was a

criterion for patient discharge home in all groups,6 80% of
our patients with OP were discharged home with an indwelling drain, which was later removed at the clinic. Some of
these patients had complications (urinary leakage) but they
were treated on an outpatient basis unless associated signs
or symptoms were present. Since readmissions were not
considered as part of the postoperative stay, the overall
hospitalization for OP was comparable to that of the LP
group even with more complications and rehospitalizations.
Of note, most complications, readmissions and need for secondary interventions in the OP group occurred in unstented
cases.
Whether the higher success rate in the LP group was due
to the improved visualization and magnification provided by
laparoscopy, or because of the universal use of Double-J
stents in the LP group cannot be determined by the data
available. Another explanation for the improved success rate
in the LP group is the shorter FU. However, Langaki et al
reported that most LP failures occurred in the first 6 months
after repair,20 and that was the median FU in the LP group
in our study.
An argument against using laparoscopic reconstructive
procedures in the pediatric population, especially in small
children, is that laparoscopic suturing is challenging, timeconsuming and associated with a prolonged learning curve.
Similar benefits, such as a small incision and rapid recovery,
can be achieved with open surgery. Trying to minimize incisional size with open surgery carries the drawback of a
small field and the need for significant tissues retraction for
exposure. In our experience even in infants transperitoneal
LP could be done safely with good exposure, excellent visualization and no need for conversion. This age group also
appears to benefit from LP compared to OP since there was
a trend toward a higher success rate and decreased
morbidity.
In the series from Paris all cases except 1 were performed
by the same surgeon.5 In this group the main concern is
teaching LP to surgeons in training. Robotic assistance was
proposed as a tool to overcome the need for advanced laparoscopic skills and shorten the learning curve. Our experience is that experienced laparoscopic surgeons trained other
surgeons and fellows, that every new surgeon using the
procedure benefits from the prior experiences of colleagues.
Thus, despite the challenging nature of the procedure the
experience is somehow transmissible. Our learning curve
does not seem to be different from that of the Boston group
with robotic assisted LP.6
CONCLUSIONS
In this comparative, retrospective study transperitoneal LP
was performed safely in all age groups with no need for
conversion to open procedures, minimal morbidity and ex-

cellent short-term results. In our experience LP in infants
and children is a more difficult and time-consuming surgery
than OP. However, it may provide a comparable or better
outcome with fewer complications, more rapid recovery and
better cosmesis. Prospective studies are needed to better
assess the role of laparoscopy in pediatric pyeloplasty.
7.
10.
BMI
FU
LP
OP
UPJO
body mass index

followup
laparoscopic pyeloplasty
open pyeloplasty
ureteropelvic junction obstruction
8.
9.
11.
12.
REFERENCES
1.
Razdan S, Bagley and McGinnis D: Minimizing minimally

invasive surgery: the 5-mm trocar laparoscopic pyeloplasty.
J Endourol 2005; 19: 533.
2. Klingler H, Remzi M, Janetschek G, Kratzik C and Marberger
M: Comparison of open versus laparoscopic pyeloplasty
techniques in treatment of uretero-pelvic junction obstruction. Eur Urol 2003; 44: 340.
3. Jarrett T, Chan D, Charambura T, Fugita O and Kavoussi L:
Laparoscopic pyeloplasty: the first 100 cases. J Urol 2002;
167: 1253.
4. Eden C, Gianduzzo T, Chang C, Thiruchelvam N and Jones A:
Extraperitoneal laparoscopic pyeloplasty for primary and
secondary ureteropelvic junction obstruction. J Urol 2004;
172: 2308.
5. Bonnard A, Fouquet V, Carricaburu E, Aigrain Y and ElGhoneimi A: Retroperitoneal laparoscopic versus open pyeloplasty in children. J Urol 2005; 173: 1710.
6. Lee R, Retik A, Borer J and Peters C: Pediatric robot assisted
laparoscopic dismembered pyeloplasty: comparison with a
cohort of open surgery. J Urol 2006; 175: 683.
13.
14.
15.
16.
17.
18.
19.
20.
1583
Clayman R: Comparison of open versus laparoscopic pyeloplasty techniques in treatment of ureteropelvic junction
obstruction. J Urol 2005; 173: 2023.
Bauer J, Bishoff J, Moore R, Chen R and Iverson A: Laparoscopic versus open pyeloplasty: assessment of objective and
subjective outcome. J Urol 1999; 162: 692.
Duel B, Vates T, Heiser D, Barthold J and Gonzlez R: The
utility of antegrade pyelography prior to pyeloplasty via
dorsal lumbotomy. J Urol 1999; 162: 174.
Franc-Guimond J, Kryger J and Gonzlez R: Experience with
the Bailez technique for laparoscopic access in children.
J Urol 2003; 170: 936.
Piaggio L, Franc-Guimond J, Figueroa T, Barthold J and
Gonzalez R: Comparison of laparoscopic and open partial
nephrectomy for duplication anomalies in children. J Urol
2006; 175: 2269.
Kavoussi L and Peters C: Laparoscopic pyeloplasty. J Urol
1993; 150: 1891.
Schuessler WW, Grune MT and Tecuanhuey LV: Laparoscopic
dismembered pyeloplasty. J Urol 1993; 150: 1795.
Peters C, Schlussel R and Retik A: Pediatric laparoscopic dismembered pyeloplasty. J Urol 1995; 153: 1962.
Metzelder M, Schier F, Petersen C, Truss M and Ure B: Laparoscopic transabdominal pyeloplasty in children is feasible
irrespective of age. J Urol 2006; 175: 688.
in the infant younger than 6 monthsis it technically possible? J Urol 2006; 175: 1477.
Reddy M, Nerli RB and Bashetty R: Laparoscopic dismembered pyeloplasty in children. J Urol 2005; 174: 700.
Yeung C, Tam Y, Sihoe J, Lee K and Liu K: Retroperitoneoscopic dismembered pyeloplasty for pelvi-ureteric junction
obstruction in infants and children. BJU Int 2001; 87: 509.
Yee D, Shanberg A, Duel B, Rodriguez E, Eiche LL et al: Initial
comparison of robotic-assisted laparoscopic versus open pyeloplasty in children. Urology 2006; 67: 599.
Inagaki TRK, Ong AM, Kavoussi LR and Jarrett TW: Laparoscopic pyeloplasty: current status. BJU Int 2005; 95: 102.
Clinical Prize Finalists

Relationship of Varicocele Grade and Testicular
Hypotrophy to Semen Parameters in Adolescents
David A. Diamond,* David Zurakowski, Stuart B. Bauer, Joseph G. Borer, Craig A. Peters,
Bartley G. Cilento, Jr., Harriet J. Paltiel, Ilina Rosoklija and Alan B. Retik
From the Departments of Urology and Radiology (HJP), Childrens Hospital Boston, Harvard Medical School, Boston, Massachusetts
Purpose: Surgical indications for correcting adolescent varicocele include high varicocele grade and testicular hypotrophy.
To our knowledge these findings have not been correlated with semen parameters to date. We examined the relationship
between unilateral left varicocele grade or the testicular volume differential and semen parameters in adolescents.
Materials and Methods: Semen analyses were done in 57 Tanner stage V adolescent males at ages 14 to 20 years
(mean SD 18 1.6). Varicocele grade was determined by the attending urologist. Testicular volumes were determined by
scrotal ultrasound performed by an attending sonologist. Data analysis was performed using nonparametric statistical
methods.
Results: Boys with testicular volume differentials greater than 10% vs those with differentials less than 10% had significantly lower sperm concentration and total motile sperm counts. For differentials greater than 20% these decreases were
more dramatic. Percent motile sperm was significantly lower for boys with volume differentials greater than 20% vs those in
the less than 10% and 10% to 20% categories. No significant differences were detected in percent normal morphology among
the volume differentials. No significant differences were detected for any semen analysis parameter as a function of varicocele
grade.
Conclusions: Sonographically derived volume differentials greater than 10% between normal and affected testes correlate
with a significantly decreased sperm concentration and total motile sperm count. This finding may serve as a marker to
identify adolescents with unilateral left varicocele who are at greatest risk for future infertility. At least semen analysis and
close annual followup of these adolescents seem warranted.
Key Words: testis, varicocele, semen, adolescent
lear indications for surgical correction of the asymptomatic adolescent varicocele remain to be well defined. Whereas varicocele grade and testicular hypotrophy have been routinely assessed and used variably as
indicators for surgical intervention, their correlation with
fertility has been indeterminate. We determined whether
semen analysis parameters correlate with varicocele grade
or testicular volume differential in Tanner V adolescents
with unilateral left varicocele.

Semen analyses were done in 57 Tanner stage V adolescent
males at ages 14 to 20 years (mean SD 18 1.6) with
clinically apparent varicoceles. Varicocele grade I to III according to the method of Dubin and Amelar1 was determined
by the attending urologist. Testicular volumes were calculated with scrotal ultrasound by an attending sonologist
using a previously described technique.2 Briefly, testes were
scanned using high frequency linear or curved array transducers. Transverse and longitudinal images of each testis
Study received Childrens Hospital institutional review board approval.

* Correspondence: Department of Urology, Childrens Hospital,
300 Longwood Ave., Boston, Massachusetts 02115 (telephone: 617355-7796; FAX: 617-730-0474; e-mail: david.diamond@childrens.
harvard.edu).
0022-5347/07/1784-1584/0
were obtained, and length, width and height measurements

were made using electronic calipers. Testicular volumes
were calculated using the empirical formula of Lambert,
length width height 0.71.3 A single semen analysis
was performed in 61% of the boys and multiple semen analyses were done in 39%. There was no selection bias in terms
of patients suggested to undergo semen analysis. For internal study consistency the most recent semen analysis for
each patient was analyzed. Semen analyses were performed
by the Reproductive Laboratory at the Brigham and
Womens Hospital, Boston, in accordance with WHO guidelines.4 This study was approved by the Childrens Hospital
institutional review board.
STATISTICAL ANALYSIS
Semen analysis parameters, including sperm concentration,
percent motile sperm, percent normal morphology and total
motile sperm count, were compared among the volume differential categories (less than 10%, 10% to 20% and greater
than 20%) and varicocele grade using nonparametric statistical methods because the data were not normally distributed with an indication of skewedness, ie asymmetry,
according to the Kolmogorov-Smirnov test. The KruskalWallis test was used to assess overall differences among
groups and the Mann-Whitney U test was applied for pairwise comparisons.5 The median and IQR (25th75th percen-
1584
Vol. 178, 1584-1588, October 2007

Printed in U.S.A.
DOI:10.1016/j.juro.2007.03.169
VARICOCELE GRADE AND TESTICULAR HYPOTROPHY

tiles) were used as descriptive statistics for all 4 semen
analysis parameters. The Spearman correlation coefficient
rs was used to assess associations between semen analysis
variables and volume differential as a continuous variable,
whereas the Pearson correlation r was used to determine
any potential association with patient age. In addition, total
motile sperm in millions was calculated according to the
formula, sperm density in 106/ml percent motile sperm/
100 semen volume in ml. Total motile sperm counts were
grouped into 3 categories (less than 10 million, 10 to 20
million and greater than 20 million) and compared according
to volume differential categories (3 3 table) and with the
Pearson chi-square test with 4 df to evaluate associations.6
We also tested by 3-way ANCOVA the effects of testicular
volume differential category, varicocele grade and age on
each of the 4 semen analysis parameters to control for any
possible confounding among the predictor variables and establish whether volume differential was significantly related
to each parameter independent of grade and age. Volume
differential category and varicocele grade were treated as
factors with 3 levels and age was treated as a continuous
covariate.7 Statistical analysis was performed with SPSS,
version 14.0 with 2-tailed p 0.05 considered statistically
significant.
1585
FIG. 1. Scatterplot of total motile sperm count in individuals reveals

significant overall differences in median counts among volume differential categories (p 0.018). Horizontal lines represent median.
gested that total motile sperm counts were lower for patients
in the 10% to 20% volume differential category compared to
those with a volume differential of less than 10% (p 0.07).
These findings were supported by the significant inverse
Spearman correlations between volume differential as a continuous (nonnormally distributed) variable and sperm concentration (rs 0.40, p 0.003), percent motility (rs
0.28, p 0.04) and total motile sperm (rs 0.33,
p 0.02).
When pooling patients in the 10% to 20% and greater
than 20% volume differential categories, sperm concentration (p 0.001) and total motile sperm (p 0.01) were lower
than the values in the less than 10% volume differential
category. No significant differences were detected in percent
normal morphology among the volume differential categories (p 0.57).
Figure 1 shows individuals in each of the 3 volume differential categories with respect to total motile sperm count
in millions. Overall differences among categories were observed. The distribution of patients among the 3 volume
RESULTS
Semen analysis parameters for all 3 volume differential
categories revealed overall group differences for sperm concentration, percent motility and total motile sperm count
(p 0.01, 0.02 and 0.018, respectively, table 1). Pairwise
comparisons between groups using the Mann-Whitney U
test indicated significantly lower sperm concentration for
patients in the 10% to 20% and greater than 20% volume
differential categories compared to those in the less than
10% volume differential category (p 0.01). Percent motility
was significantly lower for patients in the greater than 20%
volume differential category than in those in the less than
10% and 10% to 20% categories (p 0.01). Total motile
sperm was significantly lower for patients with a volume
differential of greater than 20% compared to that in patients
with a differential less of than 10% (p 0.01). Trends sug-
TABLE 1. Semen analysis parameters according to volume differential category

Parameter
No. pts
Sperm concentration (million):
Median
IQR
Range
% Motile:
Median
IQR
Range
% Normal morphology:
Median
IQR
Range
Total motile sperm (million):
Median
IQR
Range
Less Than 10%
10%20%
Greater Than 20%
27
18
12
70
36131
20273
35
1852
4110
15
680
1166
63
4075
2390
58
5069
3284
41,
2454
1067
27
1938
699
25
1335
163
22
650
475
64
32134
101,120
32
1394
1309
10
382
1222
p Value
(Kruskal-Wallis test)
0.01*
0.02*
0.57
0.02*
* Statistically significant.
Vs less than 10% p 0.01.
Vs 10% to 20% Mann-Whitney U test p 0.01.
1586
FIG. 2. Distribution of total motile sperm count (TMC) in patients in

each volume differential category revealed overall differences (chisquare test p 0.001). Low total motile counts of less than 10
million were found in 59% of patients with greater than 20% volume
differential, in 11% of those in 10% to 20% volume differential
category and in 0% of those with volume differential less than 10%.
differential categories revealed that patients with larger

volume differentials had a significantly greater percent of
semen analyses in the lower total motile sperm ranges (chisquare 22.46 on 4 df, p 0.001). For example, of patients
in the greater than 20% volume differential category 59%
had low total motile sperm counts (less than 10 million),
whereas 11% had total motile counts less than 10 million in
the 10% to 20% differential category and none in the less
than 10% volume differential category (fig. 2). Excellent
total motile counts (greater than 20 million) were observed
in 85% of patients in the less than 10% volume differential
category, in 67% in the 10% to 20% category and in only 33%
in the greater than 20% volume differential category. Thus,
there was a significant relationship between the severity of
testicular volume differential and total motile sperm count.
When stratifying patients according to varicocele grade
severity (I to III), no significant differences were detected for
any semen analysis parameters (each Kruskal-Wallis test
p 0.50, table 2).
Based on Pearson correlations none of the critical semen

analysis parameters significantly correlated with age, including sperm concentration (r 0.00, p 0.99), percent
motile sperm (r 0.18, p 0.16), percent normal morphology (r 0.02, p 0.89) or total motile sperm (r 0.19,
p 0.15). The only variable that correlated with patient age
was specimen volume with older boys producing larger volumes (r 0.47, p 0.001).
Multivariate analysis using ANCOVA revealed that testicular volume differential was significantly associated with
sperm concentration (p 0.005), sperm motility (p 0.007)
and total motile sperm count (p 0.04) independent of
varicocele grade and age, and it was not associated with
normal morphology (p 0.60, table 3). Varicocele grade and
age did not have any association with the 4 semen parameters evaluated (each p 0.05). These results confirm that a
larger volume differential is associated with significantly
lower sperm concentration and motility regardless of grade
or patient age.
DISCUSSION
Since Tulloch described varicocelectomy as a treatment for
male infertility in 1954, the optimal indications for this
surgery have been controversial.8 The prevalence of varicocele in adolescent boys is 15%, making this the most common
surgically correctable congenital anomaly in boys. However,
it seems clear from the adult literature that not all boys
require prophylactic surgical intervention since 80% of men
with varicocele are fertile, whereas only 20% have problems
with infertility.9 Therefore, how do we decide which asymptomatic boys should undergo surgery? Numerous studies
support the belief that varicocelectomy improves sperm
concentration but has variable effects on sperm motility and
morphology.10 In a meta-analysis of varicocele studies
Schlesinger et al noted improvement postoperatively in
sperm concentration in 12 of 16 published studies, improved
sperm motility in 5 of 12 and improved sperm morphology in
5 of 10.11 Sperm morphology improved only when there was
also improvement in sperm concentration. Therefore, if adolescent varicocele was known to be associated with a significantly decreased sperm concentration, varicocelectomy
would be a logical intervention.
TABLE 2. Semen analysis parameters according to varicocele grade

Parameter
No. pts
Sperm concentration (million):
Median
IQR
Range
% Motile:
Median
IQR
Range
% Normal morphology:
Median
IQR
Range
Total motile sperm (million):
Median
IQR
Range
p not significant.
Grade I
4
Grade II
Grade III
24
29
63
12102
5106
50
2193
4273
41
2499
1230
60
3580
3084
59
4570
2182
51
3869
1090
16
830
634
25
1337
490
27
1341
199
44
7248
1309
36
16131
11,120
44
1786
2344
p Value (Kruskal-Wallis test)

0.99
0.66
0.53
0.97

TABLE 3. Multivariate analysis of testicular volume differential,
varicocele grade and age vs semen analysis parameters
Parameter
Sperm concentration:
F
p value (3-way ANCOVA)
Sperm % motility:
F
Sperm morphology:
F
Total motile sperm count:
F
Vol Differential
Varicocele Grade
Age
5.90
0.005*
0.05
0.95
0.02
0.98
5.41
0.007*
0.19
0.83
3.36
0.07
0.52
0.60
0.57
0.58
0.01
0.99
3.27
0.04*
1.44
0.25
1.39
0.24
For p values volume differential and grade were factors and age was a
continuous covariate.
* Statistically significant.
Clinical parameters that are readily accessible to the

urologist include varicocele grade (I to III), testicular volume
and turgor assessments. Scrotal ultrasound has gained
widespread use as an alternative objective means of assessing volume compared to Rochester, Takahara or Prader orchidometry. Testicular turgor differences remain subjective.
The majority of pediatric urologists have used testicular
volume differentials as a primary indication and varicocele
grade as a secondary indication for surgical correction of
adolescent varicocele. Semen analysis data has been more
difficult to obtain due to clinician or patient reluctance.12
Therefore, what represents a physiologically significant volume differential or varicocele grade in adolescents due to its
effect on semen parameters has been indeterminate. We
correlated varicocele grade and volume differentials with
semen analysis results. Our finding that varicocele grade
has no correlation with semen analysis results is consistent
with the adult literature, demonstrating that varicocele
grade is not predictive of fertility, although this remains
controversial.11,13,14
On the other hand, the finding that a testicular volume
differential greater than 10% correlates with a significantly
lower sperm concentration appears to be clinically important. In addition, the correlation between increased testicular volume differentials (10% to 20% and greater than 20%)
with lower total motile sperm counts appears significant.
Prior studies demonstrated an increased likelihood of conception without invasive techniques if the total motile sperm
count exceeds 20 million.15 Some groups have suggested
moving directly to in vitro fertilization for adults with total
motile sperm counts less than 5 million. We stratified our
subnormal adolescent cases as total motile sperm counts
less than 10 million (poor) or 10 to 20 million (intermediate),
whereas greater than 20 million was regarded as normal.
Few previous studies have examined testicular volume
effect due to adolescent varicocele and its relationship to
semen quality. Haans16 and Laven17 et al noted that left
testicular growth failure in patients with varicocele relative
to controls correlated with total sperm number but not with
sperm concentration. They studied absolute left testicular
volume rather than volume differentials between affected
and unaffected testes. Paduch and Niedzielski observed that
sperm concentration varied with total testicular volume but
they did not stratify varicocele cases based on the severity of
the volume differential.18 On the other hand, Guarino et al
1587
found no predictive value of testicular volume measurement

with regard to semen analysis in Tanner V adolescents.19
However, in that study there was no significant difference in
testicular volume differential among patients with varicocele who had normal vs abnormal semen analyses and stratification by volume differential was not performed. In the
adult varicocele literature Sigman and Jarow noted a progressive decrease in total motile sperm count with a greater
reduction in left testicular volume regardless of fertility
status.20 Based on our results when evaluating adolescents,
a physical finding, ie testicular volume based on ultrasound
measurements, can be used to stratify boys into higher vs
lower risk categories regarding future fertility.
Our study was appropriately confined to Tanner stage V
patients. Some studies of adolescent varicocele have not
used this developmental criterion and, therefore, improvement in semen parameters may have reflected developmental maturation. The lack of correlation between age and
semen parameters other than volume is consistent with
proper Tanner staging.
There are some limitations to our study. Of the boys 61%
submitted 1 semen specimen. It is generally recognized that
2 or 3 specimens obtained a number of weeks apart provide
greater accuracy. In addition, normative data on semen
analysis in adolescents is lacking and we did not have a
control group for comparison. In addition, our number of
patients with grade I varicocele available for study was
modest.
CONCLUSIONS
The results of our study indicate that sonographically derived volume differentials greater than 10% between normal
and affected testes correlate with decreased total motile
sperm count and may serve as a marker to identify adolescents with unilateral left varicocele who are at greatest risk
for future infertility. Our findings do not support prophylactic spermatic vein ligation for all boys with a volume differential of greater than 10% since many of these patients have
total motile sperm counts within the normal range according
to WHO criteria, ie greater than 20 million. However, as a
group at risk for subfertility, we believe that these adolescents should at least have semen parameters studied and
followed annually. Long-term followup of this patient population with regard to semen parameters and fertility seems
particularly important.
ACKNOWLEDGMENTS
Semen analyses were performed at the Reproductive Laboratory, Brigham and Womens Hospital, Boston.
REFERENCES
1.
Dubin L and Amelar RD: Varicocele size and results of varicocelectomy in selected subfertile men with varicocele. Fertil
Steril 1970; 21: 606.
2. Paltiel HJ, Diamond DA, DiCanzio J, Zurakowski D, Borer
JG and Atala A: Testicular volume: comparison of
orchidometer and US measurements in dogs. Radiology
2002; 222: 114.
3. Lambert B: The frequency of mumps and of mumps orchitis.
Acta Genet Stat Med, suppl., 1951; 2: 1.
1588
4. Collection and examination of human semen. In: WHO Laboratory Manual for the Examination of Human Semen and
Sperm-Cervical Mucus Interaction, 3rd ed. World Health Organization. Edited by RJ Aitken, A Aribarg, K Gopalkrishnan,
DW Hamilton, DF Katz, D Mortimer et al. New York: Cambridge University Press 1992: sect 2, pp 326.
5. Bland M: An Introduction to Medical Statistics, 3rd ed. New
York: Oxford University Press 2000; pp 211229.
6. Rosner B: Fundamentals of Biostatistics, 6th ed. Belmont,
California: Duxbury Press 2006; pp 426 435.
7. Katz MH: Multivariable Analysis: A Practical Guide for Clinicians, 2nd ed. New York: Cambridge University Press 2006,
pp 24 34.
8. Tulloch WS: A consideration of sterility factors in the light of
subsequent pregnancies. II. Subfertility in the male. Trans
Edinb Obstet Soc 19511952; 104: 129.
9. Skoog SJ, Roberts KP, Goldstein M and Pryor JL: The adolescent varicocele: whats new with an old problem in young
patients? Pediatrics 1997; 100: 112.
10. Seftel AD, Rutchik SD, Chen H, Stovsky M, Goldfarb J and
Desai N: Effects of subinguinal varicocele ligation on sperm
concentration, motility and Kruger morphology. J Urol
1997; 158: 1800.
11. Schlesinger MH, Wilets FF and Nagler HM: Treatment outcome after varicocelectomy: a critical analysis. Urol Clin
North Am 1994; 21: 517.
12.
13.
14.
15.
16.
17.
18.
19.
20.
Jarow JP: Effects of varicocele on male fertility. Hum Reprod

Update 2001; 7: 59.
Gentile DP and Cockett ATK: The effect of varicocelectomy on
testicular volume in 89 infertile adult males with varicoceles. Fertil Steril 1992; 58: 209.
Vereecken RL and Boeckx G: Does fertility improvement after
varicocele treatment justify preventive treatment at puberty? Urology 1986; 28: 122.
Matkov TG, Zenni M, Sandlow J and Levine LA: Preoperative
semen analysis as a predictor of seminal improvement following varicocelectomy. Fertil Steril 2001; 75: 63.
Haans LCF, Laven JSE, Mali WPTM, te Velde ER and
Wensing CJG: Testis volumes, semen quality, and hormonal patterns in adolescents with and without a varicocele. Fertil Steril 1991; 56: 731.
Laven JSE, Haans LCF, Mali WPTM, te Velde ER, Wensing
CJG and Eimers JM: Effects of varicocele treatment in
adolescents: a randomized study. Fertil Steril 1992; 58:
756.
Paduch DA and Niedzielski J: Semen analysis in young men
with varicocele: preliminary study. J Urol 1996; 156: 788.
Guarino N, Tadini B and Bianchi M: The adolescent varicocele:
The crucial role of hormonal tests in selecting patients with
testicular dysfunction. J Pediatr Surg 2003; 38: 120.
Sigman M and Jarow JP: Ipsilateral testicular hypotrophy is
associated with decreased sperm counts in infertile men
with varicoceles. J Urol 1997; 158: 605.
Surgical Treatment of Unilaterally

Undescended Testes: Testicular Growth After
Randomization to Orchiopexy at Age 9 Months or 3 Years
Claude Kollin,* Bengt Karpe, Ulf Hesser, Tina Granholm and E. Martin Ritzn
From the Department of Woman and Child Health, Karolinska Institutet, Stockholm, Sweden
Purpose: We compared the growth of congenital, unilaterally undescended testes following orchiopexy at age 9 months or 3
years.
Materials and Methods: Patients were randomized to surgery at age 9 months (72) or 3 years (83). Testicular volume was
measured by ultrasonography at ages 6, 12, 24, 39 and 48 months.
Results: Orchiopexy at age 9 months resulted in an increase in testicular volume at subsequent measurements at ages 2, 3
and 4 years compared to the volume at 6 months (p 0.001). In contrast, no significant growth was noted in the group treated
at age 3 years. The improved testicular growth after early orchiopexy was also demonstrated by a gradual increase in the ratio
of the previously retained testis and the scrotal testis in individual boys from 6 months to 4 years (0.68 to 0.81, p 0.001).
For the late treatment group a significant decrease in this ratio was noted during the same period (0.68 to 0.56, p 0.01).
Conclusions: Surgical treatment at 9 months resulted in partial catch-up of testicular growth until at least age 4 years
compared to surgery at 3 years, clearly indicating that early surgery has a beneficial effect on testicular growth. Since
testicular volume is an approximate indirect measure of spermatogenic activity, this gives hope that orchiopexy at this age
may improve future spermatogenesis.
Key Words: testis; cryptorchidism; infertility, male; spermatozoa; abnormalities
ndescended testes is the most common congenital

anomaly in boys.1 The prevalence in full-term newborns (birth weight 2,500 gm or greater) is approximately 3%, decreasing by spontaneous descent to 1% to 2%
by age 6 months. The incidence seems to be increasing in
Denmark, suggesting an environmental influences on genital development.2 Unilateral cryptorchidism accounts for
about 85% of all cases.1
Failure of bilateral testicular descent results in severely
impaired spermatogenesis. If untreated, this results in infertility in adulthood.3 In agreement with these studies a
decreased paternity rate has been observed after surgically
treated bilateral cryptorchidism but not after unilateral
cryptorchidism.4 Orchiopexy for bilateral cryptorchidism
leads to a higher normal sperm count if surgery was performed before age 3 years compared to surgery after age 4
years.5 Therefore, it is generally accepted that retained testes should be brought into the scrotum at some time during
early childhood.
The means of accomplishing this are hormonal or surgical
treatment. Two modes of hormone treatment have been
suggested but each has been of doubtful efficacy and safety.
Study received approval from the Committee for Ethics in Medical

Research at Karolinska Institutet.
Supported by Stiftelsen Frimurare Barnhuset, Sllskapet Barnavrd,
Kronprinsessans Lovisas och Axel Tielmans fond fr barnsjukvrd
and Sven Jerrings Fond.
* Correspondence: Department of Woman and Child Health, Karolinska Institutet, Pediatric Surgery (Q3:03), Karolinska University Hospital
Solna, SE-171 76 Stockholm, Sweden (telephone: 46 8 51777714; FAX:
46 8 51777715; e-mail: claude.kollin@karolinska.se).
0022-5347/07/1784-1589/0
Several groups have disputed the effectiveness of gonadotropin releasing hormone or human chorionic gonadotropin
treatment.6 Therefore, surgical positioning of the retained
testes into the scrotum has gained renewed popularity.7
However, the optimal age for surgical treatment remains
controversial. Based on histopathological studies it was suggested that surgery should be performed within the first 2
years of life.8,9 To our knowledge there are no published,
prospective studies that randomized boys with undescended
testes to surgery at different ages with postoperative followup of testicular growth and function. Since the final goal
of treatment is to achieve normal adult spermatogenesis,
repeat testicular biopsies would seem to be the optimal way
to monitor treatment success. However, this would not be
ethically acceptable. Instead, repeat measurements of testicular volume may serve as a surrogate since there is a good
correlation between the spermatogenic activity of a testicle,
and its volume in adults10,11 and in prepubertal children.12
Therefore, we focused on testicular growth before and after
orchiopexy at age 9 months or 3 years a followup to age 4
years.
Ultrasonography is a reliable method for determining
testicular volume13 with precision and reproducibility also
during the first years of life.14 Thus, an accurate determination of the volume of the retained and scrotal testes using
ultrasonography before and repeatedly after treatment
seems to be acceptable as an early marker of the success of
treatment.
This report is the followup of a previous study in which
patients with and without orchiopexy at age 9 months were
followed until age 24 months.15 Further details of study
1589
Vol. 178, 1589-1593, October 2007

Printed in U.S.A.
DOI:10.1016/j.juro.2007.03.173
1590
SURGICAL TREATMENT OF UNILATERALLY UNDESCENDED TESTES
design, surgical procedure, definitions and methods can be

found in the previous publication.
PATIENTS AND METHODS
This study began in 1998 at the Astrid Lindgren Childrens
Hospital, Karolinska University Hospital. A total of 164
boys with unilateral palpable undescended testis were included in the study, of whom 117 were recruited at ages 0 to
3 weeks and 47 were referred later and included at age 6
months. According to neonatal records the latter boys had
unilaterally palpable undescended testes also at birth. At
age 6 months the boys were randomized to surgery at age 9
months (72) or 3 years (83). Nine boys who were included at
birth withdrew before age 6 months due to lack of parental
compliance and they were consequently not randomized.
Some patients did not undergo all ultrasound examinations
because they missed appointments, were ill or came too late
for ultrasonography.
Clinical and ultrasound examinations were used to define
the position and volume of the 2 testicles at ages 0 to 3
weeks, 2 months (no ultrasonography), 6 months, and 12, 24,
36 and 39 months in the late treatment group, and 48 and 60
months at all time points 2 weeks. The number of ultrasound examinations varied somewhat at different time
points, although patients remained in the study. Blood samples for hormone measurements were collected at each visit
and at surgery biopsy was done for histological evaluation
(data not shown).
One pediatric surgeon (CK) performed more than 95% of
the clinical examinations and surgical procedures. Inclusion
criteria were congenital unilateral palpable undescended
testis at ages 0 to 3 weeks or at 6 months. Exclusion criteria
were spontaneous descent, recognized syndromes, other
birth defects or other pathological conditions afflicting the
external genitalia, prematurity (less than 37 weeks of gestation) and prior groin surgery.
Definition of Palpable Undescended Testes
All testes that could not be brought down into the scrotum at
clinical examination were included. If a testis in a suprascrotal position could be pulled down to the upper part of
scrotum but did not remain there after about 30 seconds of
traction (after the exhaustion of cremasteric activity), it was
also defined as an undescended testis. If it remained in the
upper part of the scrotum for approximately 10 seconds or

longer after traction was released or it could be pulled down
to the lower part of the scrotal position, it was defined as a
retractile testis and not included.
Ultrasound was performed with high resolution scanners.
Testicular volume was calculated using the approximation
for a prolapsed ellipsoid, V /6 length width height.
The majority of procedures were performed by 1 pediatric
radiologist (UH). Examiners were blinded to previous results and preliminary data but obviously the scars of orchiopexy were clearly visible. At the 4-year ultrasound this was
not a problem since a 1-year-old scar could not be distinguished from a 3-year-old scar. All parts of this study were
approved by the Committee for Ethics in Medical Research
at the Karolinska Institutet.
Statistical Methods
The measured volume of the testicles of 164 boys was analyzed with a generalized linear model for longitudinal data.
Due to skewed distribution of residuals the logarithm of
testicular volumes was used for calculations. Missing values
were assumed to be missing at random. Due to the number
of missing values data were analyzed with PROC MIXED in
SAS.
RESULTS
Testicular volumes previously reported up to age 2 years15
are included in the total numbers to show the full development from ages birth to 4 years. Testicular size was determined in the retained and the scrotal testis in each patient.
Development of the Scrotal
and Retained Testes From Birth to 6 Months
Between ages 0 to 3 weeks and 6 months median volume of
the normally descended testis increased from 0.30 to 0.51 ml
(p 0.001, see table and fig. 1). The retained testis increased
from 0.25 to 0.35 ml (p 0.001, see table and fig. 2). There
was a significant difference in median volume between the
scrotal and retained testes at ages 0 to 3 weeks and 6
months (each p 0.001). Thus, during the first 6 months
postnatally the volume of the descended testis increased by
70%, while the increase in the retained testis was only 40%.
Volume at various ages of retained and scrotal testes, and ratio between them in early and late treated groups
Undescended Testis
Descended Testis
Undescended/Descended
Age group
No. Testes
Median Vol (ml)
No. Testes
Median Vol (ml)
No. Testes
Median
03 Wks
6 Mos
12 Mos:
Early surgery
Late surgery
24 Mos:
Early surgery
Late surgery
36 Mos:
Early surgery
Late surgery
48 Mos:
Early surgery
Late surgery
110
139
0.25
0.35
112
144
0.30
0.51
110
139
0.84
0.68
67
72
0.38
0.33
67
72
0.51
0.51
67
72
0.74
0.66
53
63
0.42
0.33
54
66
0.56
0.54
53
63
0.73
0.60
43
44
0.49
0.38
44
45
0.59
0.58
43
44
0.86
0.64
33
44
0.50
0.38
33
44
0.65
0.64
33
44
0.80
0.56
Orchiopexy was performed at ages 9 months and 3 years.
FIG. 1. Development of volume of initially retained testes in boys

with palpable unilaterally undescended testes from ages 0 to 3
weeks (w) to 4 years (y). At age 6 months (mo) all boys were
randomized to orchiopexy at 9 months (72) or 3 years (83). At ages
2, 3 and 4 years there was significant difference in median volume
of initially retained testes between early and late treated groups
(p 0.001). Curves show median and 95% CI. Values indicate number of testes.
Further Growth of the Scrotal

Testes From 6 Months to 4 Years
Between 6 months and 4 years an increase in the median
volume of the scrotal testes was noted (p 0.001, see table
and fig. 1). There were no significant differences in scrotal
testis volume between the 2 groups at any age.
Further Growth of The Retained
Testes From 6 Months to 4 Years and
the Influence of Orchiopexy at 9 Months or 3 Years
Orchiopexy at age 9 months resulted in an increase in median testicular volume from 0.35 ml at 6 months to 0.42 ml
at 2 years, 0.49 at 3 years and 0.50 ml at 4 years (p 0.001,
see table and fig. 1). Thus, orchiopexy was followed by significant partial catch-up growth. In contrast, during the
1591

randomized to orchiopexy at 9 months (72) or 3 years (83). In early
treated group there was increase in median ratio from ages 6
months to 3 and 4 years (p 0.001). At ages 2, 3 and 4 years there
was significant difference in this ratio between early and late
treated groups, again showing partial catch-up growth of early
treated testis (p 0.001). Curves show median and 95% CI. Values
indicate number of ultrasound examinations.
same period there was no significant growth of the retained

testes before or after surgery in the group randomized to
orchiopexy at 3 years. At ages 2, 3 and 4 years there were
significant differences in median volume between testes that
were surgically treated at 9 months compared to the group
with treatment at age 3 years (p 0.001, see table). The
volume of the initially retained testis in each group still had
not reached the size of its scrotal counterpart on any measurements at age 2, 3 or 4 years.
Retained-to-Scrotal Testis Ratio
The ratio of the volume of the undescended testis and its
contralateral scrotal counterpart provided a direct comparison of the 2 testes in individual boys. Thus, it is an index of
the degree of growth deficit of the retained testis.
In the early treatment group operated at 9 months there
was an increase in the median ratio from 0.68 at 6 months to
0.81 at 4 years (p 0.001). In contrast, a decrease in the
median ratio was noted in the late treatment group (at 3
years) from 0.68 at 6 months to 0.56 at 4 years (p 0.001, see
table and fig. 3). At ages 2, 3 and 4 years there were significant differences in this ratio between the early and late
treatment groups, again demonstrating the partial catch-up
growth of early treated testes (p 0.001).
DISCUSSION

randomized to orchiopexy at 9 months (72) or 3 years (83). There
was no significant difference in scrotal testicular volume between
early and late treated groups at any age. Curves show median and
95% CI. Values indicate number of testes.
There is a general consensus that cryptorchidism should be

treated in early childhood to achieve normal spermatogenesis and normal fertility. In the search for improvements
interest has focused on age at treatment. In the classic
histopathological studies of Hedinger the number of spermatogonia per tubular cross-section was decreased in undescended testes from age 2 years and thereafter compared to
a normal population.16 This was later confirmed by others.
However, this proves neither that treatment at age 1 year
1592
would save normal spermatogenesis nor that treatment at a

later age would lead to a worse result in adulthood. Thus,
prospective, randomized, controlled studies of the effect of
orchiopexy in infancy compared to later surgery are needed.
In the current study boys were randomized to surgical treatment at age 9 months or 3 years.
A way of assessing the effect of surgery on undescended
testes during childhood would be to follow testicular growth.
In normal boys testicular volume almost doubles during the
first 3 months of life with little or no further increase between ages 3 and 18 months.14 Also, during the following
years testicular growth is slow, as demonstrated in the current study by the development of the scrotal testes during
the first 4 years of life (see table and fig. 2), showing slow,
statistically nonsignificant growth during the second and
fourth years of life. Thus, comparing the volume of the previously retained testes during the first year after surgery at age
9 or 36 months should be a way to compare the effectiveness of
treatment at these 2 ages. Also, a comparison of the ratio
between the retained and the scrotal testis in the 2 groups can
be a good measurement of therapeutic success.
To our knowledge this prospective, randomized study
shows for the first time that early orchiopexy at age 9
months results in significant catch-up growth of the initially
retained testes up to age 4 years, which contrasts with the
complete absence of growth of testes that are still undescended until age 3 years. The growth of these latter testes
could not be salvaged by surgery at 3 years. No resumed
growth was noted until the end of the current followup at age
4 years. Also, the ratio between the previously retained testis
and its scrotal counterpart showed a significant increase following early surgery, while in the late treated group this ratio
showed a significant decrease even after surgery. These results
strongly suggest that surgery at age 9 months rather than 3
years is beneficial for testicular growth.
Our results are consistent with those in previous studies
showing that the volume of the retained testes is usually less
than volume of scrotal testes with the difference persisting
into adulthood.17,18 However, those studies were not prospective with measurements of testicular size at birth, nor
did they assess testicular growth during the first months of
life, a period when the testes show more activity than at any
other time before puberty.
We found a significant difference in volume between scrotal and retained testes at the first measurement at age 0 to
3 weeks. This suggests that prenatal factors have a role in
impaired testicular growth. However, in the first report of
this study it was shown that retained testes that at birth were
close to equal in volume with the contralateral scrotal testis
showing a significant difference in volume at 6 months.15 This
suggests that impaired growth of the retained testis was not
only due to prenatal factors. Increased temperature was suggested to be the most probable causative factor.19
Also, a biopsy study of iatrogenic cryptorchidism after
hernia repair showed that these testes have a decreased
number of germ cells per tubular cross-section as time in a
suprascrotal position increases, indicating that these testes
undergo secondary adverse changes and accumulated time
in the suprascrotal environment has an important role.20 In
addition, it was reported that surgical treatment at a
greater age increases the risk of finding no germ cells in a
testicular biopsy compared to whether surgery was performed at a younger age.9
The finding that growth of a retained testis is already

impaired during the first 6 months of life suggests that
surgery during the neonatal period might be required to
avoid further damage to undescended testes. However, before beginning such a study methods must be developed to
distinguish testes that will descend spontaneously soon after birth from those that will remain suprascrotal. Also, the
natural course of development of testicles that are suprascrotal at birth but descend during the first 6 months of life
must be established.
CONCLUSIONS
The current study shows that early surgical treatment at
age 9 months resulted in partial catch-up of testicular
growth until age 4 years. This was not the case if surgery
was delayed until age 3 years, clearly indicating that early
surgery has a beneficial effect on testicular growth. Since
testicular volume is an approximate measure of spermatogenic activity, this gives hope that early orchiopexy may
improve future spermatogenesis.
ACKNOWLEDGMENTS
Pia Kjellholm provided assistance and Magnus Backheden
provided statistical support.
REFERENCES
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
Berkowitz GS, Lapinski RH, Dolgin SE, Gazella JG, Bodian

CA and Holzman IR: Prevalence and natural history of
cryptorchidism. Pediatrics 1993; 92: 44.
Boisen KA, Kaleva M, Main KM, Virtanen H, Haavisto AM,
Schmidt IM et al: High and increasing prevalence of cryptorchidism in Denmark. Ugeskr Laeger 2004; 22: 4372.
Lee PA and Coughlin MT: Fertility after bilateral cryptorchidism. Evaluation by paternity, hormone, and semen data.
Horm Res 2001; 55: 28.
Lee PA and Coughlin MT: The single testis: paternity after presentation as unilateral cryptorchidism. J Urol 2002; 168: 1680.
Engeler DS, Hsli PO, John H, Bannwart F, Sulser T, Amin
MB et al: Early orchiopexy: prepubertal intratubular germ
cell neoplasia and fertility outcome. Urology 2000; 56: 144.
Ong C, Hasthorpe S and Hutson JM: Germ cell development in
the descended and cryptorchid testis and the effects of
hormonal manipulation. Pediatr Surg Int 2005; 21: 240.
Ritzen EM, Bergh A, Bjerknes R, Christiansen P, Cortes D, Haugen
SE et al: Nordic consensus on treatment of undescended testes.
Acta Paediatr 2007; 96: 638.
Hadziselimovic F and Herzog B: The importance of both an
early orchidopexy and germ cell maturation for fertility.
Lancet 2001; 358: 1156.
Cortes D, Petersen BL and Thorup J: Testicular histology in
cryptorchid boysaspects of fertility. Unpublished data.
Takihara H, Cosentino MJ, Sakatoku J and Cockett AT: Significance of testicular size measurement in andrology: II.
Correlation of testicular size with testicular function.
J Urol 1987; 137: 416.
Lenz S, Giwercman A, Elsborg A, Cohr KH, Jelnes JE, Carlsen
E et al: Ultrasonic testicular texture and size in 444 men
from the general population: correlation to semen quality.
Eur Urol 1993; 24: 231.
Noh PH, Cooper CS, Snyder HM, Zderic SA, Canning DA and
Huff DS: Testicular volume does not predict germ cell count
in patients with cryptorchidism. J Urol 2000; 163: 593.

13.
14.
15.
16.
17.
18.
19.
20.
Diamond DA, Paltiel HJ, DiCanzio J, Zurakowski D, Bauer

SB, Atala A et al: Comparative assessment of pediatric
testicular volume: orchidometer versus ultrasound. J Urol
2000; 164: 111.
Main KM, Toppari J, Suomi AM, Kaleva M, Chellakoty M,
Schimidt IM et al: Larger testes and higher inhibin B levels
in Finnish than in Danish newborn boys. J Clin Endocrinol
Metab 2006; 91: 2732.
Kollin C, Hesser U, Ritzn M and Karpe B: Testicular growth
from birth to two years of age, and the effect of orchidopexy
at age nine months: a randomized, controlled study. Acta
Paediatr 2006; 95: 318.
Hedinger C: Histological data in cryptorchidism. In: Pediatric
Adolescent Endocrinology: Cryptorchidism: Diagnosis and
Treatment. Basel: Karger 1979; vol 6, pp 313.
Puri P and Sparnon A: Relationship of primary site of testis to
final testicular size in cryptorchid patients. Br J Urol 1990;
66: 208.
Cendron M, Huff DS, Keating MA, Snyder HM 3rd and
Duckett JW: Anatomical, morphological and volumetric
analysis: a review of 759 cases of testicular maldescent.
J Urol 1993; 149: 570.
Bedford JM: Effects of elevated temperatures on the epididymis and testis: experimental studies. Adv Exp Med Biol
1991; 286: 19.
Fenig DM, Snyder HM, Wu H, Canning DA and Huff DS: The
histopathology of iatrogenic cryptorchid testis: an insight
into etiology. J Urol 2001; 165: 1258.
EDITORIAL COMMENT
This prospective, randomized trial addresses an important
question of the timing of orchiopexy (age 9 months vs 3
years). The authors conclude that early orchiopexy resulted
in partial catch-up of testicular growth compared to the
contralateral descended mate. The largest increase in
growth for the early group occurred by 2 years at 3 months
postoperatively. To fully compare the early and late surgery
groups the same followup should have been used. Although
the prepubertal testis is often viewed as quiescent, some
1593
groups have reported a small, second postnatal surge of

luteinizing hormone and resultant testosterone linked with
the eventual appearance of primary spermatocytes around
age 5 years,1,2 although others did not see this surge.3 It
would be interesting to continue testicular volume measurements through this period to see whether any further
changes are evident.
While the authors state that testicular volume may be
used as a surrogate for a measure of spermatogenesis, this
association is most commonly seen in adulthood. In children
Noh et al found that, although the generalizing estimating
equation demonstrated a direct correlation between testicular volume and germ cell count, individual germ cell count/
volume measures varied widely within the CIs (reference 12
in article). They concluded that testis volumes at orchiopexy
could not be used to reliably predict individual germ cell
histology. Further studies may elucidate whether early orchiopexy in a period of testis plasticity is able to evoke
changes in histology that will allow improved spermatogenesis and paternity in adulthood.
Thomas F. Kolon
Childrens Hospital of Philadelphia
University of Pennsylvania School of Medicine
Philadelphia, Pennsylvania
1.
Waaler PE, Thorsen T, Stoa KF and Aarskog D: Studies in

normal male puberty. Acta Paediatr Scand, suppl., 1974;
249: 1.
2. Huff DS, Fenig DM, Canning DA, Carr MC, Zderic SA and
Snyder HM III: Abnormal germ cell development in cryptorchidism. Horm Res 2001; 55: 11.
3. Chada M, Prusa R, Bronsky J, Kotaska K, Sidlova K, Pechova M
et al: Inhibin B, follicle stimulating hormone, luteinizing
hormone and testosterone during childhood and puberty in
males: changes in serum concentrations in relation to age
and stage of puberty. Physiol Res 2003; 52: 45.
Endoscopic Treatment of
Vesicoureteral Reflux Associated With Ureterocele
Boris Chertin,* Nochiparambil Mohanan, Amicur Farkas and Prem Puri
From the Departments of Urology, Shaare Zedek Medical Centre, Jerusalem and Ben-Gurion University of Negev, Beer Sheva, Israel, and
Childrens Research Centre, Our Ladys Hospital for Sick Children, University College Dublin (NM, PP), Dublin, Ireland
Purpose: We determined the value of endoscopic treatment for vesicoureteral reflux associated with ureterocele.
Materials and Methods: From 1984 to 2005, 109 children with a median age of 6 months underwent endoscopic ureterocele
puncture. Ureterocele presented as a part of a duplex system in 97 of patients (89%) and as part of a single system in 12 (11%).
Vesicoureteral reflux was seen to the lower ipsilateral moiety in 53 patients and in 32 contralateral kidneys (85 refluxing
renal units). Puncture was performed with a 3Fr Bugbee electrode. High grade vesicoureteral reflux or breakthrough
infection while on antibiotic prophylaxis served as the indication for the surgical correction of vesicoureteral reflux. Median
followup after endoscopic correction was 10 years (range 1 to 21).
Results: Spontaneous vesicoureteral reflux resolution following successful ureterocele puncture was seen in 36 of the 85
refluxing renal units (42%) and in 5 (6%) reflux was downgraded. The latter patients were withdrawn from antibiotic
prophylaxis and they did well. A total of 33 refluxing renal units with vesicoureteral reflux into the lower moiety of the
ureterocele kidney and 11 contralateral refluxing renal units underwent endoscopic correction. Reflux was corrected in 31 of
the 44 refluxing renal units (70%) after a single injection and it resolved after a second injection in another 9 (21%). In 4
refluxing renal units (9%) endoscopic correction failed and open reimplantation was done. Of the 109 patients (13%) 14 had
vesicoureteral reflux to the ureterocele moiety following endoscopic puncture. Of those patients endoscopic correction resolved
reflux in 3, reflux resolved spontaneously in 5 and upper pole partial nephrectomy was performed in 4 due to a nonfunctioning
moiety. The remaining 2 patients did well without antibiotic prophylaxis.
Conclusions: Our data show that endoscopic treatment of vesicoureteral reflux associated with ureterocele is a simple,
long-term effective and safe procedure, avoiding the need for open surgery in the majority of patients following endoscopic
puncture of ureterocele.
Key Words: ureter, ureterocele, vesico-ureteral reflux, endoscopy, hyaluronic acid
ecently surgical treatment for ureterocele has evolved

from complicated major surgery to minimally invasive endoscopic treatment, namely endoscopic puncture.13 However, some groups still advocate a routine aggressive surgical approach to ureterocele, including partial
nephrectomy and open ureteral reimplantation.1,4 They justify this management based on a high percent of the patients
who still have VUR to the lower moiety following puncture
or frequently de novo reflux to the ureterocele moiety associated with breakthrough infection and a low incidence of
spontaneous resolution. We previously reported our experience with endoscopic ureterocele puncture, describing a success rate of greater than 90% in terms of complete ureterocele decompression in 2 series.2,3 Moreover, a significant
number of patients demonstrated a high ratio of spontaneous VUR resolution following successful puncture. In those
who still have high grade VUR or breakthrough infection
while on antibiotic prophylaxis, endoscopic correction of
VUR was performed.
Although endoscopic treatment (STING) has a high success rate for primary VUR, to our knowledge its role for VUR
* Correspondence: Department of Urology, Shaare Zedek Medical

Center, Jerusalem, 91031, P. O. 3235, Israel (telephone: 972-26555560; FAX: 972-2-6555062; e-mail: bchertin@yahoo.com).
0022-5347/07/1784-1594/0
associated with ureterocele has not yet been explored.

Therefore, we performed a retrospective study to evaluate
the effectiveness of endoscopic treatment for VUR associated
with ureterocele.

We retrospectively reviewed the medical records of all patients who underwent primary endoscopic puncture of ureterocele at our 2 institutions from 1984 to 2005. A total of
109 consecutive children with a median age of 6 months
(range 1 month to 14 years) underwent endoscopic ureterocele puncture. Ureterocele presented as part of a duplex
system in 97 patients (89%) and as part of a single system in
12 (11%). Prenatal ultrasound detected hydronephrosis
leading to the postnatal diagnosis of ureterocele in 41 patients (38%), whereas in the remaining 68 (62%) the diagnosis was made on investigation for UTI.
Our regimen of preoperative evaluation and postoperative followup was identical at the 2 institutions and it
was previously published.2,3 Briefly, it included renal and
bladder ultrasound, VCUG, and repeat renal scans with
diethylenetriaminepentaacetic acid in the past, mercaptoacetyltriglycine-3 in the last decade and 99mtechnetium
dimercapto succinic acid. Kidney uptake of 45% to 55% of
1594
Vol. 178, 1594-1597, October 2007

Printed in U.S.A.
DOI:10.1016/j.juro.2007.03.170
ENDOSCOPIC TREATMENT OF VESICOURETERAL REFLUX AND URETEROCELE

total renal activity was considered normal. VUR was seen
in the lower moiety of the ipsilateral kidney in 53 cases
and in 32 contralateral kidneys (85 RRUs). According to
the International Classification System of the International Reflux Study Committee5 VUR was grades I to IV in
11, 46, 18 and 10 RRUs, respectively.
Puncture was performed using a 3Fr Bugbee electrode.
All patients underwent renal and bladder scans 2 weeks
following puncture. All patients received antibiotic prophylaxis until VCUG showed spontaneous VUR resolution or
definitive treatment cured VUR. High grade VUR or breakthrough UTI while on antibiotic prophylaxis served as indications for surgical correction of VUR.
Endoscopic correction was initially performed using polytetrafluoroethylene as a tissue augmenting substance. After
Food and Drug Administration approval in 2001 dextranomer/hyaluronic acid copolymer (Deflux) was used. In patients with a duplex system and VUR into a lower moiety
injection was performed by introducing the needle submucosally under the ureteral orifice of the upper moiety at the
6 oclock position (see figure).6,7 In cases in which the gap
between the 2 orifices was large, injection was done at the 6
oclock position in the submucosal part of the refluxing ureter. In patients with grade I to III VUR we used the usual
STING technique, while in those with grade IV VUR or a
widely open orifice injection was performed inside the orifice, as we previously described.8 In patients with de novo
reflux into the ureterocele moiety, requiring endoscopic correction, injection was performed by introducing the needle
under the ureteral orifice when identified, and under the
collapsed ureterocele when anatomical landmarks were not
clearly recognized to achieve the volcanic appearance of the
distal ureter.
VCUG was performed 3 to 6 months after endoscopic
correction. Antibiotic prophylaxis was administered when
VCUG showed no reflux. In those patients annual ultra-
1595
VUR outcome after endoscopic ureterocele puncture in 85 RRUs

Outcome
No RRUs (%)
Spontaneous resolution
Downgrading
Endoscopic correction after injection 1
Endoscopic correction after injection 2
Failed endoscopic treatment open reimplantation
36 (42)
5 (6)
31 (70)
9 (21)
4 (9)
sound was performed as long-term followup. Median followup after puncture was 10 years (range 1 to 21).
RESULTS
Spontaneous VUR resolution following successful ureterocele puncture was seen in 36 of the 85 RRUs (42%) and in 5
(6%) VUR was downgraded. The latter patients were withdrawn from antibiotic prophylaxis and they did well (see
table). A total of 33 RRUs with VUR into the lower moiety of
the ureterocele kidney and 11 contralateral RRUs underwent endoscopic correction. Of those 44 RRUs 18 (41%)
underwent STING using polytetrafluoroethylene as a tissue
augmenting substance. The remaining 26 RRUs (59%) received STING with Deflux. Reflux was corrected in 31 of the
44 RRUs (70%) after a single injection and it resolved after
a second injection in 9 (21%). In 4 RRUs (9%) endoscopic
correction failed. Those patients underwent open reimplantation.
Of the 109 patients 14 (13%) had VUR to the upper
ureterocele moiety following endoscopic puncture. Eight
children had an ectopic ureterocele and 6 had an intravesical ureterocele associated with a duplex system before
puncture. There was no difference in reflux grade between
children with an ectopic vs an intravesical ureterocele. In
3 patients endoscopic correction resolved VUR, while reflux resolved spontaneously in 5 cases and upper pole
partial nephrectomy was performed in 4 due to a nonfunctioning moiety at the beginning of our learning curve. The
remaining 2 patients with a poorly functioning kidney did
well without antibiotic prophylaxis. Five patients (4.5%)
had lower UTI following successful endoscopic correction
of VUR during long-term followup. In none of these children did repeat VCUG show recurrent VUR. No difference
was observed in the outcome of surgery between children
who underwent endoscopic correction using polytetrafluoroethylene vs Deflux as the tissue augmenting substance.
DISCUSSION
Endoscopic correction technique in duplex system with upper ureterocele and VUR of lower moiety. Technique is identical to that for
duplex system.6,7 A, needle is inserted under ureterocele moiety
orifice. B, needle is introduced at 6 oclock position. C, whole needle
length is introduced under 2 ureters. D, implant position after
injection.
The concept of endoscopic VUR correction provides a minimally invasive treatment option for UTI or renal parenchymal damage associated with reflux.9 Since the introduction of STING 2 decades ago and Deflux approval by
the Food and Drug Administration as a suitable implant
for subureteral injection, endoscopic correction of VUR
has became first line treatment in the majority of children
with primary and secondary VUR.10,11 The wide use of
minimally invasive approaches has changed the management not only of VUR, but also of different congenital
anomalies. We and others have reported high success with
endoscopic puncture of ureterocele.13 Endoscopic treatment of obstructive ureterocele has became an initial
1596
treatment for children with this anomaly as well as definitive treatment at the majority of centers.
However, some urologists oppose this approach, pointing
out the need for secondary surgery related to the association
of ureterocele with VUR.1,4 Open ureteral reimplantation
was for years an integral part of the complex approach to
ureterocele treatment, including upper pole partial nephrectomy and ureterocele incision. Although the possibility of
endoscopic VUR correction was reported previously, most
practicing pediatric urologists have not accepted it as a
recommended choice.12,13 Recently Perez-Brayfield et al reported the results of endoscopic treatment of complex VUR
cases with Deflux, including 5 with ureterocele after incision
or puncture.11 They noted a 60% success rate with this
approach. We argued previously that even in cases in which
further surgery for VUR is needed following endoscopic ureterocele decompression endoscopic injection may solve the
problem and, thus, spare the patients unnecessary aggressive surgery.2,3 Our data support that. In patients who presented with VUR following endoscopic puncture endoscopic
correction failed in only 4 RRUs (9%) and they underwent
open reimplantation. Moreover, spontaneous VUR resolution following successful ureterocele puncture was seen in 36
of the 85 RRUs (42%). High spontaneous resolution of VUR
may be explained by the fact that successful decompression
eventually leads to the restoration of normal trigonal anatomy.
Another crucial point raised by the opponents of the endoscopic approach to ureterocele is a high incidence of de
novo reflux into the ureterocele moiety following endoscopic
puncture. We have stressed that a meticulous puncture
technique avoids iatrogenic reflux.2,3 However, in cases in
which VUR occurs into a ureterocele moiety it is usually of
low grade and may be solved endoscopically. Of the 109
patients 14 (13%) had VUR to the ureterocele moiety following endoscopic puncture. VUR was grade III in only 3 patients and the remaining 11 presented with grades I and II
VUR. Of those patients endoscopic correction resolved VUR
in 3, VUR resolved spontaneously in 5 and upper pole partial
nephrectomy was performed in 4 due to a nonfunctioning
moiety. In the 4 patients who underwent partial nephrectomy a nonfunctioning pole was the only indication for surgery.
Opponents of the endoscopic technique for VUR into a
ureterocele moiety have usually pointed out the lack of the
proper submucosal injection site in these cases. To obtain
clear anatomical landmarks Perez-Brayfield et al suggested
using fluoroscopy and/or guidewires as a useful adjunct.11 It
is true that following endoscopic correction there is no support for submucosal injection at the 6 oclock position in
some patients, especially in children with ectopic ureterocele. In this case we suggest injecting the tissue augmenting
substance inside the ureteral orifice, as recommended for
grades IV and V VUR. When that is not possible, the surgeon
should introduce the whole length of the needle under the
collapsed ureterocele to achieve the volcanic appearance following injection.
Another interesting finding in this study is that there is a
group of patients with ongoing problems, such as a nonfunctioning or poorly functioning renal moiety, or persistent
VUR into a lower or ureterocele moiety, who do not require
any treatment and may be withdrawn from antibiotic prophylaxis without a risk of UTI and secondary surgery. We
recently reported data demonstrating that a nonfunctioning

pole may be left in place following successful ureterocele
decompression without causing morbidity during long-term
followup.14 Data in the current study further confirm these
observations. Seven RRUs are doing well without antibiotic
prophylaxis despite grade II VUR into a ureterocele or lower
moiety, of which 5 had a poorly functioning or nonfunctioning moiety. No doubt, further studies should answer the
question of whether persistent VUR in children after successful ureterocele decompression who do not have a history
of UTI would be at additional risk for renal parenchymal
damage.
CONCLUSIONS
Our data show that there is a high percent of spontaneous
reflux resolution following successful endoscopic ureterocele
puncture. When surgical correction of VUR is required, endoscopic treatment appears to be a simple, long-term effective and safe procedure, avoiding the need for open surgery
in the majority of patients.

RRU renal refluxing unit
STING subureteral polytetrafluoroethylene
injection
UTI urinary tract infection
VCUG voiding cystourethrogram
VUR vesicoureteral reflux
REFERENCES
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
Cooper CS and Snyder HM III: Ureteral duplication, ectopy,

and ureteroceles. In: Pediatric Urology. Edited by JP
Gearhart, RC Rink and P Mouriquand. Philadelphia: WB
Saunders Co 2001; pp 430 449.
Chertin B, Fridmans A, Hadas-Halperen I and Farkas A: Endoscopic puncture of ureterocele as a minimally invasive
and effective long-term procedure in children. Eur Urol
2001; 39: 332.
Chertin B, De Caluwe D and Puri P: Is primary endoscopic
puncture of ureterocele a long-term effective procedure?
Coplen DE and Barthold JC: Controversies in the management
of ectopic ureteroceles. Urology 2000; 56: 665.
Medical versus surgical treatment of primary vesicoureteral reflux:
report of the International Reflux Study Committee. International Reflux Study Committee. Pediatrics 1981; 67: 392.
Farkas A, Moriel EZ and Lupa S: Endoscopic correction of
vesicoureteral reflux our experience with 115 ureters.
J Urol 1990; 144: 534.
Miyakita H, Ninan G and Puri P: Endoscopic correction of
vesicoureteral reflux in duplex systems. Eur Urol 1983; 24:
111.
Chertin B, De Caluw D and Puri P: Endoscopic treatment of
primary grades IV and V vesicoureteral reflux in children
with subureteral injection of polytetrafluoroethylene.
J Urol 2003; 169: 1804.
ODonnell B and Puri P: Treatment of vesicoureteral reflux by
endoscopic injection of Teflon. Br Med J 1984; 289: 79.
Lackgren G, Wahlin N, Skoldenberg E, Neveus T and Stenberg
A: Endoscopic treatment of vesicoureteral reflux with dextranomer/hyaluronic acid copolymer is effective in either
double ureters or a small kidney. J Urol 2003; 170: 1551.

11.
12.
Perez-Brayfield M, Kirsch AJ, Hensle TW, Koyle M, Furness P

and Scherz CH: Endoscopic treatment dextranomer/hyaluronic acid for complex cases of vesicoureteral reflux. J Urol
2004; 172: 1614.
Diamond T and Boston VE: Reflux following endoscopic puncture of ureteroceles. A new approach using subureteric
Teflon injections. Br J Urol 1987; 71: 279.
13.
1597
Yachia D: Endoscopic treatment of ureterocele in a duplex

system. Br J Urol 1993; 71: 105.
14. Chertin B, Rabinowitz R, Pollack A, Koulikov D, Fridmans A,
Hadas-Halperen I et al: Does prenatal diagnosis influence
the morbidity associated with left in situ nonfunctioning or
poorly functioning renal moiety after endoscopic puncture
of ureterocele? J Urol 2005; 173: 1349.
Nerve Sparing Ventral Clitoroplasty:

Analysis of Clitoral Sensitivity and Viability
Jennifer Yang, Diane Felsen and Dix P. Poppas*
From the Institute for Pediatric Urology, New York Presbyterian Hospital, Weill Medical College of Cornell University, New York,
New York
Purpose: Enlargement of the clitoris is often a prominent manifestation of virilizing congenital adrenal hyperplasia and
other disorders of sexual development. Controversy persists regarding the viability and sensitivity of the clitoris following
clitoroplasty. We present 51 consecutive cases of nerve sparing ventral clitoroplasty performed by a single surgeon.
Materials and Methods: Nerve sparing ventral clitoroplasty was performed in all patients. Postoperative evaluation for
clitoral viability included gross examination and capillary perfusion testing. Patients older than 5 years were evaluated for
clitoral sensory testing and vibratory sensory testing.
Results: A total of 51 patients 4 months to 24 years old (mean age SD 4.6 6.8 years) with clitorimegaly underwent nerve
sparing ventral clitoroplasty. Of the patients 41 had capillary perfusion testing of the clitoris, of whom all had a viable clitoris.
Ten of the 41 patients underwent clitoral sensory testing. Patients reported an average degree of sensation of 3.6 0.9 at the
labia minora and 4.8 0.4 at the clitoris. Nine of the 10 patients also underwent vibratory sensory testing. Average values
for the introitus, clitoris, labia and thigh were 3.56, 1.61, 5.08, and 5.83, respectively. Mean time after surgery for the patients
who underwent clitoral sensory testing/vibratory sensory testing was 2.0 0.8 years. No variations in the sensitivity results
were reported at followup in 2 patients.
Conclusions: To our knowledge this is the largest report of followup testing of clitoral viability and sensation after
clitoroplasty. Continued long-term followup is ongoing to document long-term sexual function using this nerve sparing
ventral approach for clitoroplasty.
Key Words: abnormalities; clitoris; adrenal hyperplasia, congenital; surgery, plastic
pproximately 1/2,000 infants is born each year worldwide with genital ambiguity.1 Management of the
physical and potential psychological effects faced by
these children and their families requires a compassionate
and multidisciplinary approach.1 The surgical treatment of
patients with ambiguous genitalia has been the focus of
debate for many years and it remains in large part unresolved. This is especially true with regard to the surgical
reconstruction of the masculinized clitoris. Several concerns
regarding the current body of knowledge exist and relate to
a lack of early and long-term followup of patients undergoing clitoral surgery. Additionally, long-term outcome studies
that are available examine clitoral sensation and cosmesis
in women after surgical techniques that are no longer used
or advised.2 Furthermore, it is likely that many of these
women were treated at centers where there was limited
experience.
Surgical correction of the enlarged clitoris has undergone
many changes in the past 3 decades. Initially, clitoridectomy
was performed. Recession clitoroplasty was later devised as
an alternative. Today, reduction clitoroplasty, where the
glans is preserved and part of the erectile bodies are excised,
Study received medical institution review board approval.

* Correspondence: Institute for Pediatric Urology, Rodgers Family
Professor of Pediatric Urology, New York Presbyterian Hospital,
Weill Medical College of Cornell University, 525 East 68th St., Box
94, New York, NY 10021 (telephone: 212-746-5337; FAX: 212-7468065; e-mail: dpoppas@med.cornell.edu).
0022-5347/07/1784-1598/0
is the most widely accepted and used technique.3 Despite

advancements in surgical technique controversy persists regarding the viability and sensitivity of the clitoris following
reduction clitoroplasty. Additionally, the long-term outcome
of sexual function using this technique remains unclear. At
a minimum, optimal sexual function in patients undergoing
reduction clitoroplasty requires preservation of the innervation and vascular supply to the glans clitoris.4 We report our
results using an NSVC technique for reduction clitoroplasty
based on the current understanding of female clitoral anatomy.

Under institution review board approval from our medical
institution, we retrospectively reviewed the charts of patients that underwent NSVC, as described by Poppas et al,5
as performed by a single surgeon at 1 institution between
1996 and 2005 for clitorimegaly associated with ambiguous
genitalia. Records were also reviewed with respect to diagnosis, patient age at presentation, age at surgery, diagnostic
procedures and gender assignments. All patients were assessed and treated by a multidisciplinary team consisting of
a pediatric urologist, pediatric endocrinologist, pediatric
endo-psychiatrist and geneticist.
All patients followed routine preoperative protocol, including a liquid diet 48 hours before surgery. Patients were
admitted to the hospital the day before surgery for final
1598
Vol. 178, 1598-1601, October 2007

Printed in U.S.A.
DOI:10.1016/j.juro.2007.05.097
NERVE SPARING VENTRAL CLITOROPLASTY

assessment by the multidisciplinary team. Medical treatment was initiated by the endocrinologist in patients who
required medical therapy. Patients also received preoperative antibiotics on call to the operating room.
A total of 51 patients 4 months to 24 years old (mean
age SD 4.6 6.8 years) with clitorimegaly associated with
ambiguous genitalia underwent NSVC. Of the patients 32
were between 0 and 2 years old, 5 were 2 to 5 years old and
11 were older than 5 years. Of these patients 46 (90%) were
genetic females with congenital adrenal hyperplasia, 3 (6%)
were 46 XY who had undergone sex reassignment surgeries
and 2 (4%) were 46 XX disorder of sexual development. On
initial examination, 4 (8%), 22 (43%), 11 (21%) and 6 patients (12%) were rated Prader score II to V, respectively.
The remaining 16% of the patients were undocumented.
Phallus length was 1.0 to 4.5 cm (average 2.4).
Clitoral Testing
Clitoral viability was evaluated postoperatively by gross
examination and CPT. Capillary perfusion of the clitoris was
performed by applying pressure to the clitoris until it
blanched, when pressure was removed. Capillary perfusion
time was determined by the number of seconds until complete reperfusion of the glans. This was compared to the nail
bed capillary perfusion test.
Patients older than 5 years were considered candidates
for CST. CST was performed using a cotton tip applicator.
Using a scale of 0 no sensation to 5maximum sensation,
the patient was asked to report the degree of sensation at
various points of the inner thigh and genitalia (labia majora,
labia minora, vaginal introitus and clitoris). Inner thigh
stimulation was set at level 3 for each patient and used as a
baseline to compare other areas tested. In addition, these
patients also had vibratory sensory testing performed using
a biothesiometer designed to quantify the ability of patients
to detect vibratory stimuli. A biothesiometer is most commonly used in testing for neurological diseases, such as
peripheral neuropathy.6 The device generates a vibratory
stimulus of varying amplitudes that can be gradually increased until the sensation is perceived by patients. A quantitative measure was established by recording the amplitude
of vibration (on a scale from 1 to 10), which correlated with
the threshold. The amplitude of vibration was inversely
proportional to vibratory sensing ability. Two patients underwent sensory evaluation before and after surgery.
RESULTS
Evaluation of postoperative results included assessment of
overall cosmesis, CPT and in select patients CST of the
clitoris. Followup was 1 week to 8 years (mean 24.4 months).
Two patients were lost to followup. No infections, ischemic
tissue loss or other postoperative complications were identified. None of the patients in the study group were identified
as requiring further revision of the clitoris.
A total of 49 patients (96%) had CPT of the clitoris. All
patients had a viable clitoris with normal CPT compared to
nail bed testing (less than 5 seconds). Average clitoral reperfusion was 2.8 0.4 seconds. Ten of the 49 patients (20%)
were older than 5 years and were considered candidates for
CST. Patients reported an average degree of sensation of 3.6
0.9 at the labia minora and 4.8 0.4 at the clitoris. Nine of
the 10 patients also underwent clitoral vibratory testing.
1599
Area Tested
No. Pts
Mean SD
Capillary refill
Clitoral sensitivity:
Inner thigh
Labia minora
Clitoris
Vibratory sensory testing:
Inner thigh
Labia minora
Vaginal introitus
Clitoris
49
10
2.8 0.4
Referent
3.6 0.9
4.8 0.4
5.83 1.2
5.08 1.7
3.56 0.5
1.61 0.6
Average values for the introitus, clitoris, labia and thigh

were 3.56, 1.61, 5.08, and 5,.83 respectively (see table).
Mean time after surgery for the patients who underwent
clitoral testing was 2.0 0.8 years. Two patients showed no
change in sensation when tested before and after clitoroplasty.
DISCUSSION
Various techniques have been described to correct clitorimegaly. Until the 1960s, the clitoris was typically amputated,
resulting in a female appearing perineum but poor sexual
function.7 Alternatives to clitoridectomy were later described,
including recession and relocation of the enlarged clitoris.8 The
clitoris was preserved by burying and imbricating the corporeal shaft and excess glans clitoris beneath the pubis. While
these procedures maintained clitoral innervation and function
with good cosmetic result, there were many disadvantages,
including pain during stimulation due to the trapping of erectile tissue beneath scar. In 1970, Randolph and Hung devised
a new technique for the correction of enlarged clitoris, that is
reduction clitoroplasty.9 A reduction clitoroplasty involves excision of a portion or all of the shaft with preservation of the
glans. Spence and Allen later described subtotal resection of
the clitoral shaft with preservation of the glans.10 The major
disadvantage to these procedures is that the primary vascular
and nervous supply of the glans may be interrupted and the
viability of the glans clitoris relies on collateral supply from
surrounding soft tissue.
These alternative surgical techniques were followed by
descriptions of methods to preserve the neurovascular bundle through a ventral dissection.11,12 Wedge resection of the
glans clitoris has been described for the management of
marked clitorimegaly. The glans and corporeal bodies are
bivalved in a coronal plane, removing approximately 50% of
the girth of the ventral corporeal tissue down to the level of
the base of the shaft. No bulging has been reported with
stimulation and patients have good sensation and clitoral
function.13 Despite these advances in operative technique,
the published literature describing surgical outcomes and
functional results reflect complications of older procedures
that are no longer used.
Early studies by Money et al showed satisfactory sexual
response in females following clitorectomy.14 In 1966 Gross
et al reported a series of 47 patients who underwent clitoridectomy with good cosmetic results.15 Clitoral recession and
relocation of the enlarged clitoris has been reported with
disadvantages, including painful erections, abnormalities in
sexual function, compromised appearance and progression
of clitoral enlargement in poorly controlled patients on steroids. Reduction clitoroplasty, as described by Randolph and
Hung, was performed on 14 patients with good results in
1600
terms of erection and cosmetic appearance.9 However, no

followup information on the sexual function or clitoral sensation of these patients was reported.
Reduction clitoroplasty in which part or the entire shaft
is excised with glans preservation is the most accepted and
widely used technique today. Various problems have been
associated with reduction clitoroplasty, ranging from loss of
sensation and sexual function to sloughing of the glans.16,17
There are a number of published studies of clitoral sensitivity after clitoroplasty. Minto et al examined 39 adults with a
history of ambiguous genitalia who underwent feminizing
genitoplasty using a validated sexual function questionnaire
and genital examination.18 They found that patients who
had undergone clitoral surgery had significantly higher difficulties with sensuality compared to the no surgery group.
The 2 groups had overall difficulties with orgasm. Krege
et al assessed long-term followup in 27 patients who underwent clitoroplasty and vaginoplasty. They reported the presence of clitoral sensation in 6 patients who underwent clitoroplasty, but only through qualitative description from a
nonvalidated questionnaire.2 However, none of these reports
provided an objective assessment of clitoral viability or followup testing of clitoral sensitivity.
Crouch et al described the initial results from a pilot
study involving 6 women who had undergone varying degrees of clitoral surgery.19 They obtained post-coital questionnaire results as well as specialized sexual function assessment. Light touch, thermal and vibratory sensation
thresholds were evaluated for the clitoris and vagina using a
genitosensory analyzer and Von Frey filaments. All 6 women
were shown to have highly abnormal results for sensation of
the clitoris, but these women underwent clitoral surgical
procedures that are no longer used. Frost-Arner et al described a similar group of postpubertal women who underwent reduction clitoroplasty at a mean age of 20 years.20
They compared vibration and light touch/deep pressure sensation in this cohort compared to those in a group of normal
women. In contrast to the previous studies, they found no
difference between women who had single clitoral reduction
and the normal cohort. The heterogeneous nature of these
data regarding clitoral sensitivity following clitoroplasty
stems from the wide variety of surgical techniques used
historically in this specialized patient population.
In 2002, a consensus statement recommended that it is
crucial to preserve the neurovascular bundle, the glans and the
preputial skin of the glans if clitoral reduction is to be performed.4 Preservation of the neurovascular bundle relies on an
intimate understanding of anatomy. Baskin et al reported a
computerized 3-dimensional picture analysis of 14 normal human fetal clitoral specimens.21 This permitted a detailed description of the normal fetal clitoral nerve bundles and corporeal bodies. Gearhart et al performed dorsal neurovascular
bundle nerve conduction studies at the time of feminizing genitoplasty in 6 patients with an average age of 13 months. They
concluded that modern, nerve sparing techniques allow preservation of the neurovascular bundle, which may permit normal sexual function in adulthood.22 The Intersex Society of
North America responded by citing the experience of adults
who had undergone feminizing genitoplasty as infants who
have normal nerve conduction studies and yet have an absence
of clitoral sensation and orgasmic experience. Therefore, despite precise anatomical models and careful surgical technique
during clitoroplasty, short-term and long-term followup are

needed.
There is an obvious concern relating to attempting to
quantify outcomes in clitoral surgery. There are no substantiated tests for assessment of the viability and function of
the clitoris in unaffected women. Therefore, we have chosen
the capillary perfusion test as a potential model for viability
related to clitoral perfusion. Capillary refill testing of the
nail bed has long since been used as an overall measure of
perfusion, in the setting of dehydration or hypovolemia. The
recent meta-analysis by Steiner et al cited capillary refill
time, in combination with other clinical parameters such as
skin turgor and respiratory pattern, as a useful initial assessment of dehydration.23 One study used capillary nail bed
refill time as an indication of low upper body blood flow in
premature infants.24 Similarly, we used this simple technique
as estimation of the perfusion of the clitoris, as compared to
normal values for capillary refill time of the nail bed.25
Based on a clear understanding of the female clitoral
anatomy we used a modified NSVC technique.5 This technique resulted in excellent cosmetic outcome and no postoperative ischemic complications. In addition, the vascular
supply to the clitoris was well maintained, as evidenced by a
mean CPT of 2.96 seconds (normal less than 5) in all patients tested. CST and vibratory testing results revealed
heightened clitoral sensitivity relative to surrounding genitalia. Unlike other studies in the literature, all patients in
this study underwent NSVC, which was performed by a
single surgeon. To our knowledge this is the largest cohort
with followup clitoral viability and sensitivity ever reported.
A limitation of this study was the lack of a normal cohort for
comparison. Although the results presented suggest relative
normal clitoral sensitivity in these patients, continued followup is indicated as well as the use of validated instruments
to determine sexual function after the cohort becomes sexually
active. At that time, it may be feasible to obtain similar test
results from an age matched normal cohort for comparison.
CONCLUSIONS
To our knowledge this is the first report of followup testing of
clitoral viability and sensation after reduction clitoroplasty.
The nerve sparing reduction clitoroplasty described in this
report leaves the dorsal neurovascular bundles of the corporeal
bodies and the glans clitoris intact. This is a safe and reliable
approach to correct the enlarged clitoris. Sexual and social
function of our patient cohort is difficult to assess until all
patients reach sexual maturity and adolescence. Continued,
long-term followup is ongoing to document long-term sexual
function using this nerve sparing approach for clitoroplasty.

CPT capillary perfusion testing
CST clitoral sensitivity testing
NSVC nerve sparing ventral clitoroplasty
REFERENCES
1.
Maharaj NR, Dhai A, Wiersma R and Moodley J: Intersex

conditions in children and adolescents: surgical, ethical,
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
and legal considerations. J Pediatr Adolesc Gynecol 2005;

18: 399.
Krege S, Walz KH, Hauffa BP, Korner I and Ruben H: Longterm follow-up of female patients with congenital adrenal
hyperplasia from 21-hydroxylase deficiency, with special
emphasis on the results of vaginoplasty. BJU Int 2000; 86:
253.
Lee PA and Witchel SF: Genital surgery among females with
congenital adrenal hyperplasia: changes over the past five
decades. J Pediatr Endocrinol 2002; 15: 1473.
Clayton PE, Miller WL, Oberfield SE, Ritzen EM, Sippell WG
and Speiser PW: ESPE/ LWPES CAH Working Group. Consensus statement on 21-hydroxylase deficiency from the
Lawson Wilkins Pediatric Endocrine Society and the European Society for Paediatric Endocrinology. J Clin Endocrinol Met 2002; 87: 4048.
Poppas DP, Hochsztein AA, Baergen RN, Loyd E, Chen J and
Felsen D: Nerve sparing ventral clitoroplasty preserves
dorsal nerves in congenital adrenal hyperplasia. J Urol,
part 2, 2007; 178: 1802.
Vanden Broucke H, Everaert K, Peersman W, Claes H,
Vanderschueren D and Van Kampen M: Ejaculation latency times and their relationship to penile sensitivity in
men with normal sexual function. J Urol 2007; 177: 237.
Jones HW and Wilkins L: Gynecological operations in 94 patients with intersexuality. Am J Obstet Gynecol 1961; 82:
1142.
Lattimer JK: Relocation and recession of the enlarged clitoris
with preservation of the glans: an alternative to amputation. J Urol 1961; 86: 113.
Randolph JG and Hung W: Reduction clitoroplasty in females
with hypertrophied clitoris. J Pediatr Surg 1970; 5: 224.
Spence HM and Allen TD: Genital reconstruction in the female
with adrenogenital syndrome. Br J Urol 1976; 45: 126.
Mollard P, Jusskiewenskim S and Sarkissian J: Clitoroplasty
in intersex: a new technique. Br J Urol 1982; 53: 341.
Rajfer J, Ehrlich RM and Goodwin WE: Reduction clitoroplasty via ventral approach. J Urol 1982; 128: 341.
Hutson JM, Voigt RW, Luthra M, Kelly JH and Fowler R:
Girth-reduction clitoroplastya new technique: experience
with 37 patients. Pediatr Surg Int 1991; 6: 336.
Money J, Hampson JG and Hampson JL: Hermaphroditism:
recommendations concerning assignment of sex, change of
15.
16.
17.
18.
19.
20.
21.
22.
23.
24.
25.
1601
sex and psychologic management. Bull Johns Hopkins

Hosp 1955; 97: 284.
Gross RE, Randolph J and Crigler JF: Clitorectomy for sexual
abnormalities: indications and technique. Surgery 1966;
59: 300.
Farkas A, Chertin B and Hadas-Halperin I: 1-Stage feminizing
genitoplasty: 8 years of experience with 49 cases. J Urol
2001; 165: 2341.
Sircili MHP, de Mendonca BB, Denes FT, Mareira G, SanchezBachega TAS and Arnaldo de Queiroz e Silva F: Anatomical
and functional outcomes of feminizing genitoplasty in patients with virilizing congenital adrenal hyperplasia. Clinics 2006; 61: 209.
Minto CL, Liao LM, Woodhouse CR, Ransley PG and
Creighton SM: The effect of clitoral surgery on sexual outcome in individuals who have intersex conditions with ambiguous genitalia: a cross-sectional study. Lancet 2003;
361: 1252.
Crouch NS, Minto CL, Laio LM, Woodhouse CRK and
Creighton SM: Genital sensation after feminizing genitoplasty for congenital adrenal hyperplasia: a pilot study.
BJU Int 2004; 93: 135.
Frost-Arner L, Aberg M and Jacobson S: Clitoral sensitivity
after surgical correction in women with adrenogenital syndrome: a long-term follow-up. Scand J Plast Reconstr Hand
Surg 2003; 37: 356.
Baskin LS, Erol A, Li YW, Liu WH, Kurzrock E and Cuhn GR:
Anatomical studies of the human clitoris. J Urol 1999; 162:
1015.
Gearhart JP, Burnett A and Owen JH: Measurement of pudendal evoked potentials during feminizing genitoplasty:
technique and application. J Urol 1995; 153: 486.
Steiner MJ, DeWalt DA and Byerley JS: Is this child dehydrated? JAMA 2004; 291: 2746.
Osborn DA, Evans N and Kluckow M: Clinical detection of low
upper body blood flow in very premature infants using
blood pressure, capillary refill time and central-peripheral
temperature difference. Arch Dis Child Fetal Neonatal Ed
2004; 89: F168.
Sherber NS, Wigley FM and Scher RK: Autoimmune disorders:
nail signs and therapeutic approaches. Dermatol Ther
2007; 20: 17.
Guest Lecture
Challenges for the Pediatric Urologic Surgeon in the 21st Century
n the business world challenges of the future are called
externalities, and they occur outside ones control but
profoundly affect us. We need to be smart enough to
respond to those challenges. The issues to be discussed are
manpower in pediatric urology, demographics, changes in the
health care marketplace, government regulatory changes and
leadership.
Husmann performed a wonderful survey of the manpower in pediatric urology, and there are currently 40 open
jobs, which is quite good. Not all are perfect opportunities,
but there are a lot of good jobs. There are different feelings
about the impact of subspecialty certification, as 33% of
pediatric urologists think it will bring more business to them
but 67% say we are already busy and it probably will not
have an effect. However, I think it will drive the quality
agenda in pediatric urology. Of practicing pediatric urologists 58% are in academics but that means 42% are not, and
that rate is growing 5% to 6% with each survey. Currently,
there are approximately 280 pediatric urologists nationally,
and 8 to 12 pediatric urologists retire per year. We have 22
fellowship sites but in the last few years we only 8 matched
and then 12 matched, and those numbers really are not
replacing the physicians who are retiring. However, this
year 20 people matched (50% male and 50% female). A
challenge for the future is we just do not know how many
pediatric urologists we are going to need.
What are some of the challenges on the outside? Medicare
reimbursement is a real key issue because it is related to
commercial reimbursement, which directly affects us. Last
year the federal government tried to reduce reimbursement
5.6% and it was staved off. This year they are proposing a
5.1% cut for 2007 and they have proposed a 37% cut during
the next 7 years. This is not a good trend.
The next thing is what will Generation X, Generation Y
and Generation Z, the silent generation, think about health
care? Will they believe in the mission of health care and/or
the business of health care? Will they seek other options? Do
they follow a path to medical school and incur hundreds of
thousands of dollars in debt to practice in a profession that
is increasingly finding it difficult to match the rising costs
with reimbursement? This is a challenge to attracting physicians and good people.
The war on terror changes things because it will have an
impact on foreign trained medical students. Nearly 19% of
surgical fellows are internationally trained, and approximately half of them are born overseas. Will they be able to
obtain J-1 or other visas to practice in this country? At this
time relatively few physicians practicing outside primary
care qualify for these waivers.
Another challenge is the graying of America. Beginning
in 2013, 10,000 Americans will turn 65 every single day, and
the implication of that is that their medical costs are 3 to 5
times the amount for people under 65. By 2030, 20% of our
0022-5347/07/1784-1602/0
entire population will have passed that 65th birthday. Medicare spending was $34 billion in 1980 and $252 billion in
2002, and it will double easily by 2012. We all know that
there are a smaller number of workers to pay the bill. It is a
collision.
Federal and state funding will be directed toward this
elderly population, and it will affect medical jobs and opportunities in health care. My niece goes to Cornell Medical
School. The first speech she got on her first day at medical
school was that geriatrics was a health career that she should
definitely consider. It is being preached in medical schools.
Our challenges to attracting talent are higher educational debt and lifestyle changes. People who are going into
pediatric urology now do not have the same issues I had 10
years ago when I was a pediatric urologist. They are dual
caregivers in families, and do not want to take calls.
While the number of surgical cases in pediatric urology
has increased, according to Husmanns survey the complexity of these operations has decreased in the last 5 years by
45%. It is changing but this should not be a concern as new
opportunities within this specialty will be created.
Another reality is that there are salary discrepancies
between pediatric urology and our adult colleagues, and the
additional 2 years of training has an impact on actual cash
flow for people who decide to choose pediatric urology as a
profession.
The demographics of the pediatric patient pipeline are
robust. The female population demographics indicate that
20% of the population this year was younger than 15 years.
However there is a demographic twist. The childbearing
ages are basically the early 20s, and so 13% of the population, as it exists now, are in childbearing age.
The fact that the crude birth weight and fertility rate in
this country are decreasing is interesting. The total number
of pregnancies has declined steadily by an average of more
than 1% per year. Of the pregnancies 62% ended in live
births with the balance resulting in induced abortion or fetal
loss. The rate of pregnancies in a larger cohort per 1,000
women between 15 and 44 years old was 9% lower in 1996
than it was in 1990. The pregnancy rates are highest for
women in their early 20s but in that group the most rapidly
growing potential childbearing population between 15 and
20 years old is the Hispanic and black nonHispanic populations.
So what are the observations and demographics? Our
specialty is changing, and so are the people that we hope to
attract. The demographic trends in population aging may
make recruitment and compensation more challenging, the
aging baby boomers will test medical resources and the
progressively decreasing birth rate, which is matching
trends that are much worse in Europe, may diminish our
patient pipeline.
1602
Vol. 178, 1602-1605, October 2007

Printed in U.S.A.
DOI:10.1016/j.juro.2007.07.070
CHALLENGES IN 21st CENTURY

Changes in the health care marketplace are rapid. We are
at 12% of gross domestic product in our medical spending,
and it is projected to be 17% by 2012. The big problem is the
gap between the national health expenditure and the consumer price index, which is why it seems like health care is
more costly. We talk about robots, minimally invasive surgery
and the pharmaceutical pipeline, as well as a whole host of
innovations in medicine, all of which drives the cost up but the
consumer price index still has a gap.
The problem is the American public has this great sense
of entitlement. They want the best care, they want it right
now and they want somebody else to pay for it. Industry and
the employers are looking at this saying, How do we deal
with that psychology? What is happening is that the American public has asked our industry to change the paradigm
by making competition occur in the marketplace. The employers role as a benefit sponsor is eroding, and as a consequence the employees are transitioning from recipients of
health benefits to buyers of health benefits. Employers have
driven the benefit buy-downs, which means the benefits are
not as rich as they used to be and they cost less, but at least
the employees can have health benefits. However that is
creating a gap in health benefits.
The most common current consumer directed solutions
are the high deductible health plans ($1,000, $2,000, $5,000
deductibles) coupled with the health savings account. What
this means is that the employees are paying more out-ofpocket for their health care, and the products they are using
require their direct involvement in selecting care because it
is their money. That is moving the market to what is called
retail.
The behavior of these people in the market is based on
their health status, income, employer support, involvement
in health care, benefits experience, confidence in decision
making, and trust in the health care system. They are everything from active health managers to disengaged fatalists. I believe that unconcerned minimizers represent a large
number of the people in this population. Now it is retail and
the American public is going to be buying their health care,
and scrutinizing the providers, hospitals and health plans.
Everybody is under the gun, which is called transparency.
It is a market that is ripe for what is called in business,
disruptive innovation, which is when a leapfrog occurs, such
as when we went from phones to e-mail and from fax to
e-mail. It totally changes the marketplace. An example of
what disrupts the marketplace is minute clinics or ready
clinics. These clinics employ a nurse practitioner and are
setup in a supermarket, Wal-Mart or CVS Pharmacy. You
visit the clinic for a sore throat, headache or cold, and pay
$44. The nurse practitioner gives you a beeper and while you
are shopping she/he beeps you. You return to the clinic for an
examination or throat culture, the beeper is given back to
you and you continue to shop. You are beeped again to
return to the clinic, the Strep test is positive and you are
given a prescription for ampicillin, which you have filled at
the store. You do not go to a pediatrician or family practitioner. That is the first step in retail and the retail marketplace, just like there is a first step in robotics.
There is an immense amount of disruptive innovation,
including new consumer management models. Humana and
Aetna have physician provider details on the web site, including outcomes of the physician. For surgeons their length
of stay, cost of office visits and cost of surgical procedures are
1603
available to the consumer, because they are going to be

spending their money.
What we have here is a circumstance of creating a whole
different environment, where there is a direct connection
between the provider and the consumer of health care. However, this entitlement will disappear as national standards
occur. We will rethink the clinical paradigm of patient dependence on the physician. Patients will look at the dollars
first and if it is their dollars, they will make decisions about
outcomes and quality of care, and go to the appropriate
physicians. This will be driven by a digitized marketplace and,
as we move to the electronic medical record, private health
record and national provider database, all of these advances
move us toward a market that is going to be totally transparent. Health plan providers are going to have to put their rates
out there and we are going to have to put our editing software
out there, and so it all has to be transparent. That is a big
change.
When there is little involvement but the cost is high, the
idea is that consumer engagement, and provider and health
plan transparency will drive down the cost. However, this
will involve people being involved in health care, which
means we will have predictive modeling, evidence-based medicine, coordination of care and electronic medical records. All of
those initiatives will remove waste in the system, and there is
an immense amount of waste. There will be more precision,
which is what the government and employers are pushing us to
do. And frankly, since employers pay for health care premiums, we have to listen to them.
Rapid consolidation within health plans is occurring
through mergers and acquisitions. The twin towers are
WellPoint and United, and there are 7 other plans. This
could ultimately become an oligopoly in the health care
market. Companies like United are vertically integrating by
purchasing companies that provide provider software such
as case management, predictive modeling and profiling.
They own the company that does this work for almost everybody in the country. PacifiCares behavior health organization and office management software, WebMD, is owned
by United, and they are vertically integrating as a company.
Thats a pretty scary thing. Also, there used to be 123 Blues
plans 10 years ago but now there are only 32 and they are
deeply consolidating. There is significant consolidation
within the marketplace. Why shouldnt there be consolidation? EBITAD is earnings before interest, taxes and depreciation, and this is what business people look at. If you are
making $6 to $8 billion or $4 to $6 billion a year, you can buy
other health plans and consolidate, which is exactly what is
happening.
Combined Aetna, Cigna, Coventry, HealthNet, Humana,
United and WellPoint insure 50 million people, or basically
a quarter of the total population in the United States. That
is a deep consolidation of power and a deep consolidation of
money. It has disruptive effects in the marketplace, particularly for those in a not-for-profit health plan.
The money spent on direct-to-consumer advertising from
June 2005 to January 2006 increased 20%. The ad dollars
are primarily spent on sedatives, statins, and asthma and
osteoporosis medications. The total market last year in direct-to-consumer advertising was $4 billion, which is about
quadruple the budget that Coca-Cola would spend on 1
product. At the same time the direct-to-consumer advertising has gone up, the detailing and information to physician
1604
and the money for research grants have decreased 22%. The
message is that the advertising dollars being spent on direct-to-consumer products are taking away from research
dollars because there is a higher return on investment for
the pharmaceutical companies. And as a consequence, consumers are requesting drugs directly from the physician,
some of which are the lifestyle drugs they see advertised.
The next area of change is the specialty biotech market.
Specialty infusion drug therapy is a $200 billion a year
market and it is growing 10% to 15% a year. It is the biggest
challenge we have as a health plan. Therapy may cost
$25,000 per patient per year and is used by 4% of the
population for diseases such as psoriasis, hepatitis C, rheumatoid arthritis and multiple sclerosis. There are also drugs
called orphan drugs created by the orphan drug act of 1983.
These drugs are used by only 0.1% of the population but each
therapy costs between $75,000 and $100,000 per patient per
year for sickle cell disease, multiple myeloma and cystic
fibrosis. In addition, there are ultra rare therapies for Gautiers disease, Fabrys disease and mucopolysaccharidosis,
which cost $300,000 a year. In my 400,000 member health
plan we have 8 people with Gautiers disease in 2 families. A
lot of money and direction are going that way, and in 2012
we believe that at least 50% to 60% of the medical loss ratio,
which is the cost of health care, will be in pharmaceuticals
paying for drugs.
In the marketplace employers are putting pressures on
employees and health plans for solutions. For-profit health
plans are consolidating and gaining market power, which is
good for stockholders but bad for doctors. That money comes
out of our pockets as well as the hospitals and consumers
pockets. It is bad for not-for-profit health plans that are
basically thin margin businesses and it is not good for the
public at large. Market competition in retail products and
transparency of provider performance and pricing are here
to stay, and will be a driving force in market change in the
future. The minute clinics are just the beginning of retail
medicine, and electronic medical records, private health information records and national provider database will enable and drive the process of data transfer, efficiency in
providing health care cost savings, improved care and transparency of patient outcomes.
With regard to challenges from the government, I have
already mentioned Medicare provider reimbursement. We
pay 130% of Medicare to our providers in the Capitol region,
and as a physician driven plan we are sensitive about that.
Based on the national provider idea, you will be asked to
submit a national provider ID into the national provider
database in the future if you have Medicare patients, as that
will be the only way you will get paid. This process will soon
apply to Medicaid as well because it is driven by Centers for
Medicare and Medicaid Services.
We were spending close to $30 billion on Medicaid in
2004. Of the 52.4 million enrollees in Medicaid about 50%
are children, and the balance are adults, the blind and
disabled, and the elderly. I always thought Medicaid was for
kids. It turns out that children get 19% of the expenditure
and the bulk of it goes to the blind, disabled and elderly.
Because of the demographic trend that 65-year-old patients
consume 3% to 5% more than someone younger than 65,
from 2003 to 2004 there was an $11.3 billion increase but the
increase went to the disabled and elderly. Only 38% of it was
growth within the system. At that trough the pediatric urol-
ogist is last in line. Although I want to pay people Medicaid

rates higher than the state gives us, the pool that we have
continues to erode for the pediatric populations. It is a real
challenge.
ICD-10 replaces the ICD-9 coding, and it is an expanded
alphanumeric system. Much of the rest of the world uses
ICD-10, and the reason that the government is pushing so
hard to do this is because we cannot do international medical research in disease surveillance worldwide without using ICD-10 like everyone else. ICD-10 provides greater detail in coding. If there are 12,000 ICD-9 diagnostic codes
there will be 120,000 ICD-10 codes. The codes are 3 to 5
digits for ICD-9 and 7 alphanumeric for ICD-10. So ICD-10
will cover body section, system, root operation, body part,
approach (laparoscopic, transperitoneal), device (robotic,
laparoscopic)and deal with modifiers and qualifiers, all of
which will be built into the alphanumeric system. In urology, benign prostatic hyperplasia goes from 2 codes to 5
codes.
The impact on health plans is going to be unbelievable
because we have to determine how to pay for all these
different codes and build them into our computer systems.
However with our computer systems and software configuration, it is a multiple year project. This will require modifications to all electronic transactions using diagnostic and
procedural coding, complete retraining of staff in hospitals
and offices, changes in correspondence, potential loss of revenue and rework from inexperience. The learning curve
associated with this change is going to be challenging. You
can no longer use eponyms in medical record documentation.
Reporting will change, contracting will change with hospitals and providers, and the diagnosis related group structure
will change within the hospital system. This is unbelievably
huge.
How much will this cost? The Blue Cross Blue Shield
Association believes it will cost $10 billion. The RAND have
predicted a lesser number. The fact remains that this technology advance law has already been passed by the House.
They are pushing hard to have it activated by 2010 but
consultants have told Congress that this is so big, difficult,
challenging and disruptive that you should not even dream
about it until October 2012, which is 5 years away. Although
the details on timing and specific coding are not finalized, we
have to get this on our radar screen and begin to plan now
for the impact.
How do you live with all of these challenges? There is a
book entitled Heroic Leadership: Best Practices from a 450Year-Old Company That Changed the World by Chris
Lowney, and it is a history of the Jesuit orders. How did they
work as a business and how were they successful for 450
years? They used 4 principles.
The first principle is self-awareness. We need to selfevaluate, look at ourselves, what they called examine yourself. We need to understand our strengths or weaknesses,
values and world view, and keep a good focus on who we are.
As Walt Whitman said, be true to yourself. Every heart
vibrates to that iron strength. The second principle is we
have to have ingenuity, the ability to innovate, to remain
flexible, to adapt constantly. The Jesuits had to because they
were everywhere in the world. We have to have heroism, and
this is referred to as heroic leadership. We have to energize
ourselves through ambitious goals, motivating and inspiring
ourselves. The third principle is we have to inspire and

motivate people around us because we all do great things
but it takes a heroic vision to do it. Finally, the fourth
principle is we have to love what we do and our fellow man.
This is a quid pro quo for us because we spent our lives
devoted to children.
Now, what do you do? You approach problems with a
sense of urgency and full accountability, and you measure
your progress against goals. This is called after-action review, and is what the military does every day. They train
people in Iraq and Afghanistan to do that. We do this within
our company. What is after-action review? You make a plan,
deciding what you are going to do, and you execute the plan.
At the end of it, you ask what worked, what did not work,
what do we need to do to change it and what do we need to
do to improve it. The fact is, without sensitivity, look at what
you do, evaluate it and do it frequently. In the battlefield in
Iraq they do this every day. Also you need to take risks.
1605
In conclusion, the issue of heroic leadership is the word

Magus. What does that mean? It is a Latin word that means
more. You do more than other people, and you do it with
enthusiasm, love, self-awareness and drive. My belief is that
for pediatric urology big challenges are out there but with
innovation in surgical procedures and diversification in the
fields of medical endeavor, we may want to expand our
portfolio of medical interests, and we should embrace the
quality agenda associated with certification. That will be
critical to our specialty. Take the quality agenda, and use
that certification wisely.
You are all surgical leaders. You have talent, energy and
the tools to create your own future during this time. I know
you will do it.
William J. Cromie
Capital District Physicians Health Plan
Albany, New York
Outcomes Research
Esthetic Outcomes of Genitoplasty in Males
Born With Bladder Exstrophy and Epispadias
Brian A. VanderBrink, Jeffrey A. Stock and Moneer K. Hanna*
From the Departments of Pediatric Urology, Schneiders Children Hospital (BAV, MKH), New Hyde Park and New York Hospital-Cornell
University (MKH), New York, New York, and Saint Barnabas Medical Center (JAS, MKH), Livingston, New Jersey
Purpose: Male children born with bladder exstrophy-epispadias complex may have poor self-image because the esthetic
aspects of the genitalia acquire greater significance with age. We reviewed our long-term outcomes of genitoplasty in males
born with exstrophy-epispadias complex with emphasis on the esthetic results.
Materials and Methods: A total of 70 male patients with exstrophy-epispadias complex were the basis of this retrospective
review. Primary and secondary genitoplasty was performed in 44 and 26 patients, respectively. Various surgical methods
were implemented for genitoplasty, including Cantwell Ransley technique with or without paraexstrophy flaps in 36 cases,
partial corporeal mobilization with or without dermal graft in 16, subperiosteal corporeal mobilization from the pubic rami
in 7 and complete penile disassembly in 6. The parent and when appropriate the patient determined satisfaction with the
cosmetic appearance. The incidence of postoperative complications and need for surgical revision were noted.
Results: Long-term followup was available in 65 of 70 patients. Esthetic satisfaction was present in 60 of 65 patients (92%).
A total of 19 revisions (29%) were required with 4 patients undergoing multiple revisions. These revisions included fistula
repair, repair of urethral stricture, glanular contouring and recurrent chordee release.
Conclusions: The penile malformations observed in exstrophy-epispadias complex may be due to a short urethral strip,
corporeal disproportion and/or iatrogenic scarring. The optimal method to correct corporeal disproportion includes placement
of a dermal graft. In select cases careful and partial mobilization of the crura from the pubic rami allows penile lengthening.
In secondary reconstructive efforts subperiosteal dissection provides a virgin surgical field, while protecting neurovascular
and erectile structures.
Key Words: abnormalities, bladder exstrophy, epispadias, esthetics, penis
istorically the importance of achieving continence

and preservation of renal function in patients born
with bladder EEC resulted in focus on the urinary
tract during complex reconstructive surgeries. Advancements in urinary continence for children born with EEC
have been achieved primarily based on the surgical techniques of Jeffs and Gearhart along with the recently developed complete primary closure technique developed by
Grady and Mitchell.1,2 The associated external genital abnormalities in males with EEC include a penis that appears
to be short, broad, flat and tethered upward by a dorsal
urethral strip. In a staged reconstructive approach penile
reconstruction is performed at age approximately 1 year,
whereas the complete primary closure technique addresses
the problem during the neonatal period.
Regardless of when corrective surgery of the phallus is
done, this can have a dramatic impact on the psychological
well-being of the male child. Long-term followup reveals
that many of these patients have poor self-image because
the esthetic aspects of their genitalia acquire greater significance with age.3 Recognizing the fact that as these patients
reach puberty, appearance is equally if not more important
than function, reconstructive surgery has been refined by
* Correspondence: 101 Old Short Hills Rd., Suite 203, West Orange, New Jersey 07052 (telephone: 973-325-7188; FAX: 973-3257409; e-mail: mhanna@mkhanna.com).
0022-5347/07/1784-1606/0
many surgeons. Thus, attention to cosmesis and esthetics

has begun to assume a larger role in reconstructive surgery
for the patient with EEC. We previously described our experience with esthetic aspects of surgery of the pubic area
and umbilicus in the patient with EEC.4,5 We present our
27-year clinical experience with reconstructing the male external genitalia in EEC with a focus on esthetic outcomes.
A total of 70 male children born with EEC between 1977 and
2006 were under the care of the senior author. Long-term
followup (1 to 29 years) was available on 65 of 70 patients
and it served as the basis of the data analyzed. Retrospective
review of the office charts of these 65 patients was performed
to assess the esthetic and functional outcomes of their reconstructive surgeries. Genitoplasty was performed concomitantly with other genitourinary reconstructive surgeries, ie
bladder augmentation, bladder neck reconstruction, construction of continent urinary reservoir, etc, in the majority
of patients. The parent and when appropriate the patient
determined satisfaction with the cosmetic outcome in a subjective manner. The incidence of postoperative complications
from each procedure and need for revision were noted.
Surgical Techniques
Various surgical methods were implemented when performing male genitoplasty in the study population (table 1).
1606
Vol. 178, 1606-1610, October 2007

Printed in U.S.A.
DOI:10.1016/j.juro.2007.03.192
ESTHETIC OUTCOMES OF GENITOPLASTY
1607
TABLE 1. Male genitoplasty surgical methods

Method
No. Treated
Cantwell-Ransley Technique with/without paraexstrophy

flaps6,7
Johnston technique (partial corporeal mobilization):9
Dermal graft
No dermal graft
Kelley Eraklis technique (extensive corporeal
mobilization)10
Mitchell Bagli technique (complete penile
dissassemby)11
36
Total
16
11
5
7
6
65
Which procedure selected depended on preoperative and

intraoperative findings, such as the severity of phallic length
inadequacy, degree of dorsal curvature and whether it was
primary or reoperative phalloplasty. Early in the series the
Jeffs-Gearhart staged approach, which included the modified Cantwell-Ransley technique with or without paraexstrophy skin flaps, was used in 36 patients.6,7 Paraexstrophy flaps, as described by Duckett,8 were used when the
proximal urethral plate was transected during primary
bladder closure and the skin flaps adjacent to the exstrophied bladder were raised, rotated and interposed between
the edges of the divided urethral plate.
When genitoplasty using the standard Cantwell-Ransley
technique was assessed intraoperatively and the results
were thought to yield an inadequate phallus, alternate reconstructive techniques were used. In these difficult situations penile lengthening was achieved by partial corporeal
mobilization from the inferior pubic rami, as described by
Johnston (fig. 1).9 Following mobilization of the corporeal
bodies from the rami the migration of the medial results of
these structures, and realignment of the right and left corpora can provide a longer length of visible penis (fig. 2).
FIG. 2. A, Preoperative view of short, barely visible phallus in 24year-old male wishing to pursue penile lengthening procedure. Patient also has absent umbilicus. B, intraoperative view shows patient after corporeal mobilization and alignment. C, 6 months
postoperatively note significant penile lengthening and reconstructed umbilicus after penile lengthening and umbilicoplasty.
Dermal grafts were necessary in 11 patients to correct

chordee, while attempting to maximize penile length. These
grafts were harvested from a hairless region of skin over the
lower anterior abdominal wall near the anterior iliac superior spine. The tunica albuginea was incised transversely
along the dorsal portion directly opposite the site of maximal
curvature following artificial erection with saline infection
until a straight penis was achieved. The dermal graft was
then sutured using running absorbable material to bridge
the gap between the severed tunical edges. Artificial erection was repeated after graft placement.
Seven patients who underwent secondary genitoplasty
and revision for penile lengthening also underwent subperiosteal corporeal mobilization, as described by Kelley and
Eraklis.10 Briefly, the periosteum of the inferior pubic rami
is incised with diathermy and subperiosteal dissection using
an osteotome is used to mobilize the corpora cavernosa to
decrease the risk of direct injury to the neurovascular bundles and tunica albuginea. The complete penile disassembly
technique, as described by Mitchell and Bagli11 for the repair of epispadias with or without associated bladder exstrophy, has been the method of choice to repair epispadias in
our patients since 2000, with 6 patients having undergone
this procedure (fig. 3).
RESULTS
FIG. 1. Partial corporeal cavernous mobilization off of inferior pubic

ramus and midline alignment to augment visible penile length, as
originally described by Johnston.9
Of the male patients undergoing genitoplasty 43 were born

with classic bladder exstrophy and 22 were born with epispadias. Of the 44 patients who underwent primary repair
35 underwent this within 72 hours of life. Five patients
underwent repair between ages 3 weeks and 6 months, while
the remaining 4 underwent repair after age 1 year. The male
patients who underwent secondary repair were 4 to 25 years
old. Followup in the 11 patients with dermal grafting was 1
to 20 years.
On physical examination 63 of 65 patients had a horizontal or downward angled penis while standing. Satisfaction
with the esthetic outcome was high in our series. A total of
60 patients (92%) were satisfied, while 5 (8%) were unsatisfied with the appearance of the penis after surgery (table 2).
Three of the 5 patients who were dissatisfied with the outcomes underwent esthetic revisions, including glanular contouring and release of recurrent chordee with cosmetic results that were acceptable to the patients.
Excluding the most recently operated group of patients
who underwent the Mitchell technique, 22 revisions were
required following genitoplasty with 4 patients undergoing
1608
FIG. 3. A, preoperative view of newborn male with classic bladder EEC. B, intraoperative view shows complete primary exstrophy repair with
dissection of urethral plate from corpora cavernosa. Urethral plate was stained with methylene blue to assist in defining its borders.
C, intraoperative view after bladder closure, division of corporeal bodies and while closing urethra. Penile urethra now resides in its normal
ventral position. D, immediate postoperative view after complete primary closure.
multiple revisions. The most common reasons for reoperation were urethrocutaneous fistula and urethral stricture.
These 2 diagnoses were seen most frequently in boys in
whom paraexstrophy flaps were used during urethral reconstruction. Two of the 6 boys had a hypospadiac meatus
following penile disassembly repair.
DISCUSSION
While surgical techniques have improved the functional outcomes of EEC, caregivers have realized that body image and
cosmesis are a priority to patients as they go through puberty into adulthood. Psychological reviews indicate that
patients are concerned about the appearance of the body and
it is now known that reconstruction can provide an almost
normal appearance. In fact, adolescents consider the abnormal appearance of the external genitalia to be a greater
importance than incontinence.3 As physicians, it is important to realize the quality of life issues that these patients
face when dealing with their bodies. Efforts directed toward
esthetic reconstruction of the lower external genitalia may
help alleviate some of their concern.
The decreased penile length in male patients born with
EEC has been speculated to result from diastasis of the
pubic symphysis separating the bodies of the corpora, creating the illusion of a short penis despite normal corporeal
length. Alternatively a second hypothesis is that the corpora
cavernosa are congenitally short. Our impression is that a
combination of the 2 factors may coexist. Several studies
have been done to better understand the factors behind the
observed decreased penile length in patients with EEC and
determine which techniques were efficacious to augment
length. Silver et al used pelvic magnetic resonance imaging
in 10 men who underwent exstrophy reconstruction in childhood and compared the results to those in 10 age and race
matched controls.12 Using specific corporeal measurements
the overall length of the corpora cavernosa in the exstrophy
group was shorter than in the control group. The decrease in
length was a result of a shorter anterior segment of crura,
rather than the posterior segment attached to the pubic
ramus. Clinically Schillinger and Wiley examined the relationship between the intersymphyseal distance and penile
length retrospectively.13 They found that patients with an
intersymphyseal distance of greater than 4.0 cm following
iliac osteotomies and division of the sacrotuberous and sacrospinous ligaments resulted in an inadequate phallus.
Most of these failures occurred in postpubertal boys. The
findings of these studies demonstrate that the decreased
penile length observed in males with EEC can be attributable to widely separate, congenitally short corpora cavernous bodies.
Using this knowledge it would seem obvious that efforts
toward correcting the diastasis of the pubic bones would
adequately augment penile length. However, it was observed clinically that approximation of the pubic bones by
iliac osteotomies, when possible, does not always increase
functional penile length. This is believed to be due to the fact
that correction of the skeletal anomaly leads to pelvic tilting,
resulting in a more horizontal lie of the pubo-ischial rami,
while the penis retracts, becoming less prominent than before.9 Furthermore, it is difficult to perform osteotomies in
men with EEC who complain of decreased penile length and
desire reconstructive surgery.
Various surgical techniques were used in our series for
phallic reconstruction. Most patients underwent CantwellRansley epispadias repair with good results. Using this technique satisfaction with the esthetic outcomes was achieved
in 34 of 36 patients (94%). However, surgical revisions were
necessary because of the increased incidence of urethral
stricture and urethrocutaneous fistulas in this group. This
complication is not unexpected and up to 40% of patients in
whom paraexstrophy skin flaps are used in bladder closure
may experience a complication as a result of the flaps.14 The
observed urethrocutaneous fistulas following CantwellRansley repair in our series are consistent with other reported fistula rates in the literature.15
Kelley and Eraklis first proposed mobilization of the corpora from their attachment to the widely separated ischiopubic rami, resulting in migration of these structures medial
TABLE 2. Esthetic outcomes based on male genitoplasty surgical technique

No. Surgical Technique
Esthetic Outcome
Jeffs et al
14
Johnston
Kelley and Eraklis10
Mitchell and Bagli11
Total No.
Satisfied
Dissatisfied
34
2
15
1
7
0
4
2
60
5
Totals
36
16
65

and superior to provide an adequate length of visible penis.10 While in these initial report they commented on satisfactory results in a small number of patients, concern over
the safety of this procedure has been raised. The apprehension stems from the location of the neurovascular bundles in
relation to the crura. In the exstrophic-epispadias penis the
position of the neurovascular bundles was found to be more
laterally displaced, rather than the normal dorsal location of
the neurovascular bundles on the penile shaft.16 Hurwitz
et al noted that, as the bundles became more proximal and
approached the point where the corpora joined the pubic
bone, they coursed more anterior and the bundles became
difficult to trace, merging into the cleft between the corpora
and pubic bone.16 Injury to these neurovascular bundles
may occur if meticulous surgical techniques are not followed.
Thus, knowledge of the vascularity of the penis is essential
for safely performing penile reconstructive procedures.
To minimize the devascularization risk Johnston proposed partial9 rather than complete mobilization of the corpora, as originally described by Kelley and Eraklis.10 In our
series when penile lengthening was deemed necessary in
primary phalloplasty, partial crural mobilization using the
Johnston technique was the initial preferred method (fig. 1).
However, in cases of secondary genitoplasty penile lengthening by crural mobilization was accomplished using an
osteotome to stay subperiosteal and avoid injury resulting in
devascularization. In secondary or tertiary reconstructions a
subperiosteal approach provided a virgin surgical field,
while protecting neurovascular and erectile structures.
Mitchell and Bagli developed a surgical technique
based on the paired penile vascular anatomy to correct
epispadias.11 The basis of the repair is complete penile
disassembly into 3 discrete components, including the
urethral plate, and the right and left hemicorporeal glandular bodies. The blood supply to the disassembled penis is
based on the paired dorsal and neurovascular bundles to
each hemiglans, the deep cavernous arteries to the corporeal
bodies, with the spongiosal tissue supplying the proximal
urethral plate. The upward pulling effect on the penis is
eliminated with the urethral plate dissected and separated
from the corporeal bodies. Furthermore, mobilization allows
corporeal rotation, which in turn contributes further to the
correction of dorsal chordee.17 It is sometimes necessary to
correct dorsoventral length discrepancies by dermal graft
insertion and suturing together adjacent corporotomies.18
We performed 11 dermal grafts to minimize penile shortening with no adverse effects on erectile dysfunction and no
recurrent chordee. The additional penile length achieved
with this dermal grafting resulted in a dependent phallus
and cosmesis preferable to that of ventral plication alone,
which is likely to shorten the phallus.
Penile disassembly dissection enables the penis to hang
down at a normal angle but hypospadias may result due to
the shortness of the urethral plate following positioning of
the bladder posterior in the pelvis. Hypospadias was observed in 33% of patients in the current series who underwent the Mitchell technique. The penis was resurfaced using
reversed Byars skin flaps in our patients. However, new
techniques to further improve the cosmetic appearance using rotational flaps to avoid a longitudinal dorsal scar were
recently described.19
There are limitations to our results and the conclusions
derived from them. The first limitation is the retrospective
1609
nature of the study and the inherent biases that accompany

this type of study. The second limitation is that no validated
questionnaire was used to assess patient satisfaction with
the postoperative appearance. The concept, beauty is in the
eye of the beholder, can account for the different perceptions
of subjective variables, such as cosmesis and esthetics,
which can occur among individuals. Disparate opinions of
postoperative outcomes could have been attained by the
surgeon and patients with the surgeon thinking that the
outcome was superior to what the patient believed. However, the notion that a better outcome would be perceived by
the surgeon rather than the patient is not uniformly substantiated by the literature. Patient and surgeon opinions 1
year postoperatively following cosmetic reduction mammaplasty revealed that the majority of patients rated the esthetic outcomes of surgery significantly higher than surgeons.20
CONCLUSIONS
External genitalia abnormalities are a hallmark of congenital malformations in patients with EEC and they are the
cause of great psychological distress in some male patients.
Attention to cosmesis during reconstruction helps improve
the perception of body image and self-esteem in these patients. We describe various surgical techniques to perform
genitoplasty with excellent esthetic outcomes according to
our patients and families. In select cases careful and partial
mobilization of the crura from the pubic rami allows penile
lengthening. In secondary reconstructive efforts subperiosteal dissection provides a virgin surgical field, while protecting neurovascular and erectile structures.

EEC exstrophy-epispadias complex
REFERENCES
1.
2.
3.
4.
5.
6.
7.
8.
Lakshmanan Y, Gearhart JP and Jeffs RD. Review of longterm results of staged functional reconstruction and bladder neck plasty in bladder exstrophy patients followed from
birth at the Johns Hopkins Hospital. In: The ExstrophyEpispadias Complex: Research Concepts and Clinical Applications. Edited by JP Gearhart and R Mathews. London:
Kluwer Academic/Plenum Publishers 1999; pp 149 163.
Grady RW and Mitchell ME: Complete primary repair of exstrophy. J Urol 1999; 162: 1415.
Meyer KF, Frietas Filho LG, Martins DMS, Vaccari M and
Carnevale J: The exstrophy-epispadias complex: is aesthetic appearance important? BJU Int 2004; 93: 1062.
VanderBrink BA, Stock JA and Hanna MK: Aesthetic aspects
of bladder exstrophy: results of puboplasty. J Urol 2006;
159: 1684.
Pinto PA, Stock JA and Hanna MK: Results of umbilicoplasty
for bladder exstrophy. J Urol 2000; 164: 2055.
Cantwell FV: Operative treatment of epispadias by transplantation of the urethra. Ann Surg 1895; 22: 689.
Ransley PG, Duffy PG and Wollin M: Bladder exstrophy closure and epispadias repair. In: Operative Surgery-Paediatric Surgery, 4th ed. Edinburgh: Butterworths 1989; p 620.
Duckett JW: Use of paraexstrophy skin pedicle grafts for correction of exstrophy and epispadias repair. Birth Defects
1977; 13: 175.
1610
9.
10.
11.
12.
13.
14.
15.
16.
17.
18.
19.
20.
Johnston JH: Lengthening of the congenital or acquired short

penis. Br J Urol 1974; 46: 685.
Kelley JH and Eraklis AJ: A procedure for lengthening the
phallus in boys with exstrophy of the bladder. J Ped Surg
1971; 6: 645.
Mitchell ME and Bagli DJ: Complete penile disassembly for
epispadias repair: the Mitchell technique. J Urol 1996; 155:
300.
Silver RI, Yang A, Ben-Chaim J, Jeffs RD and Gearhart JP:
Penile length in adulthood after exstrophy reconstruction.
J Urol 1997; 157: 999.
Schillinger JF and Wiley MJ: Bladder exstrophy: penile
lengthening procedure. Urology 1984; 24: 434.
Gearhart JP, Peppas DS and Jeffs RD: Complications of paraexstrophy skin flaps in the reconstruction of classical bladder exstrophy. J Urol 1993; 150: 627.
Surer I, Baker LA, Jeffs RD and Gearhart JP: The modified
Cantwell-Ransley repair for exstrophy and epispadias: 10year experience. J Urol 2000; 164: 1040.
Hurwitz RS, Woodhouse CRJ and Ransley P: The anatomical
course of the neurovascular bundles in epispadias. J Urol
1986; 139: 68.
Koff SA and Eakins M: The treatment of penile chordee using
corporeal rotation. J Urol 1984; 131: 931.
Lindgren BW, Reda EF, Levitt SB, Brock WA and Franco I:
Single and multiple dermal grafts for the management of
severe penile curvature. J Urol 1998; 160: 1128.
Khoury AE, Papanikolau F, Afshar K and Zuker R: A novel
approach to skin coverage for epispadias repair. J Urol
2005; 173: 1332.
Godwin Y, Wood SH and ONeill TJ: A comparison of the patient
and surgeon opinion on the long-term aesthetic outcome of
reduction mammaplasty. Br J Plast Surg 1998; 51: 444.
EDITORIAL COMMENTS
These authors outlined the evolution of the technique of
genitoplasty in children born with EEC. There is probably
no area in pediatric urology where more mistakes have been
made and more poor outcomes have been seen than in attempts at genitoplasty in this group of patients.
The authors reviewed their experience with 70 male children born with EEC and outlined the major problems noted
with genitalia repair during a number of years. They stress
the importance of dermal grafting for correcting corporeal
disproportion, the importance of mobilizing the corporeal
bodies from the pubis for penile lengthening, the importance
of wider mobilization and the release of midline dorsal scarring.

Is this a perfect study? No. Does this report give us a
single uniform method for dealing with genitoplasty in children born with EEC? No. Does this report give us any
long-term information on erectile or sexual function in this
group of individuals? No. However, the authors give us some
important guidelines on dealing not only with primary genitoplasty, but more importantly with complications of previously done genitoplasty. The principles and techniques outlined by the authors are sound and rational.
Terry W. Hensle
Columbia University
New York, New York
This is a retrospective review of aesthetic outcomes after
genitoplasty in 65 boys with exstrophy (43) and epispadias
(22). Of the patients 60 (92%) were satisfied with the appearance of the penis. The report details various surgical
maneuvers used by the senior author to achieve such excellent results. Does form lead to function?
Surgical reconstruction of bladder EEC has evolved in the
last 4 decades. Good surgical outcomes with primary closure,
continence and genitoplasty are now the expected standard
of care. Whether this translates into reasonable quality of
life can only be addressed by prospective patient reported
outcome data on adult patients with bladder EEC. Several
validated questionnaires, such as the Incontinence Symptom Index-Pediatric,1 the International Inventory of Erectile
Function-15 and the Female Sexual Function Inventory are
available off the shelf for this purpose. Such standardized
studies will lead to meaningful comparisons of surgical techniques in these complex cases and they may guide therapy in
the future.
Yegappan Lakshmanan
Division of Pediatric Urology
The Brady Urological Institute
The Johns Hopkins Medical Institutions
Baltimore, Maryland
1.
Nelson CP, North A, Gearhart JP and Lakshmanan Y: Patientreported urinary function and incontinence among adults
born with classic bladder exstrophy. J Urol, suppl., 2007;
177: 123, abstract 372.
Sexual Function in Adult Patients With

Spina Bifida and its Impact on Quality of Life
Jenny Lassmann, Francisco Garibay Gonzalez, Jeanne B. Melchionni, Patrick S. Pasquariello, Jr.
and Howard M. Snyder, III*
From the Division of Urology, Department of General Pediatrics (JBM, PSP), Childrens Hospital of Philadelphia,
Purpose: We evaluated sexual function in adult patients with spina bifida and its impact on quality of life.
Materials and Methods: Between March 2005 and May 2006, 76 adult patients with spina bifida, including 34 women and
42 men, with a mean age of 24.4 years (range 18 to 37) were seen for followup at our institution. Following institutional review
board approval data were collected from medical records and validated questionnaires were completed, including the Watts
Sexual Function Questionnaire to assess sexual function and SF-36 to assess quality of life.
Results: Of the 76 patients 18 (24%), including 9 women and 9 men, achieved sexual intercourse at least once in the last 2
months. There was no difference regarding gender distribution and mean age SD in sexually active vs not sexually active
patients (25.8 4.2 vs 24 4.5 years, p 0.13). All levels (thoracic to sacral) of myelomeningocele were seen in the 2 groups
with significant higher lesions of neurological impairment in not sexually active than in sexually active patients. No
difference was seen in relation to ambulatory status and urinary incontinence. Overall total Watts Sexual Function
Questionnaire scores in sexually active patients were similar in men and women. When comparing the 4 domain scores on
desire, arousal, orgasm and satisfaction, women scored similar to men. Only 2 men tried to achieve paternity but failed and
1 woman gave birth. When evaluating SF-36 for quality of life, there was no significant difference in physical health (42.4
11.9 vs 38.7 7.2, p 0.11) and mental health (54.1 11.3 vs 58.6 10.7, p 0.12) subscores in sexually active vs not
sexually active patients.
Conclusions: In our cohort 24% of adult patients with spina bifida were sexually active. Sexual activity was not related to
gender, level of urinary incontinence or extent of physical disability but it was more likely in patients with more caudal levels
of neurological impairment. Sexual function seems not to affect health related quality of life in these patients.
Key Words: spina bifida in adults; sexual function; quality of life
pina bifida is the most common disabling birth defect

in the United States with a reported prevalence of
4.6/10,000 births and a female-to-male ratio of 1.2.1
Major advancements in surgical and multidisciplinary treatment of these children have markedly increased life expectancy with 50% to 75% surviving into adulthood.2,3 One of the
greatest challenges today is to establish a transition of care
for this growing population of adult patients from pediatric
to adult centers. To better understand and define their needs
the assessment of factors contributing to health related QOL
is necessary. QOL instruments are designed to measure the
subjective perception of the patient regarding physical, functional emotional and social impact of a medical condition.
However, only a small number of studies to assess QOL in
patients with SB have been done with only 1 providing data
on adults.4,5 The studies suggest that QOL is to a higher
degree associated with social factors, such as school, peers,
family and job skill development, than with physical deficits.
Study received institutional review board approval.

* Correspondence: Division of Pediatric Urology, Childrens Hospital of Philadelphia, 3rd Floor Wood Building, Childrens Hospital
of Philadelphia, 34th St. and Civic Center Blvd., Philadelphia,
Pennsylvania 19104-4399 (telephone: 215-590-2769; FAX: 215-5903985).
0022-5347/07/1784-1611/0
Sexual activity is an important aspect of adult life. Physical limitations and the lack of independence can interfere
with normal psychosexual development, resulting in limited
relationships and rare dating in these patients.6 Few studies
have focused on this topic, mostly investigating males,7,8
with only 1 using a validated questionnaire.9 We performed
a cross-sectional study to asses sexual function in a cohort of
adult patients with SB and its impact on QOL by applying
validated questionnaires.

Study Population
Between March 2005 and May 2006, 89 consecutive patients
with SB who were 18 years old or older and followed since
childhood were seen at our institution, which is a tertiary
pediatric center. Of the patients 34 women and 42 men with
a mean age of 24.4 years (range 18 to 37, median 24) were
included in the study. A total of 13 patients with impaired
cognitive function were unable to complete validated questionnaires and they were excluded. Data were collected from
medical records, including the documented level of neurological impairment, ambulatory status and associated hydrocephalus, as determined by the presence or absence of a
ventriculoperitoneal or ventriculojugular shunt. Each pa-
1611
Vol. 178, 1611-1614, October 2007

Printed in U.S.A.
DOI:10.1016/j.juro.2007.03.162
1612
SEXUAL FUNCTION AND SPINA BIFIDA
tient was interviewed to specify methods of bladder emptying and urinary incontinence by determining the frequency
of urinary leakage (never, dry longer than 4 hours, dry 2 to
3 hours or always wet). Employment was assessed and defined as having a job on the open market with no inclusion of
sheltered employment. Validated questionnaires to assess
sexual function and QOL were completed.
WSFQ
WSFQ is a 17-item, self-reported instrument assessing the 4
domain scores of sexual experience, including desire (6
items), arousal (4 items), orgasm (4 items) and satisfaction
(3 items) with gender specific versions for arousal and orgasm. Responses are measured on a 5-point Likert scale,
ranging from always to never. When the overall WSFQ score
is calculated, the possible range is 17 to 85 points with high
scores reflecting better sexual function. Content validity,
internal consistency and test-retest reliability are established.10,11 The survey was applied to all sexually active
patients who stated that they had achieved sexual intercourse in the past since the content of the questions requires
this condition. Patients not sexually active were asked to
complete particular items to assess desire and arousal.
SF-36
General health related QOL was measured using the validated SF-36, a 36-item questionnaire that assesses health in
the 8 domains of physical function, limitations due to physical problems, limitations because of emotional problems
(role emotional), social function, mental health, energy, pain
(role physical) and general health perception. Scores are
transformed in 2 main scores, that is physical health and
mental health. All scores are expected to be a mean SD of
50 10 in the general population in the United States with
a lower score indicating a greater effect of disease on QOL.
Statistical Analysis
The chi-square test for categorical variables and the Student
t test for quantitative variables were applied with p 0.05
considered statistically significant. The study was approved
by the institutional review board and prior consent was
obtained.
RESULTS
Sexual Activity and Related Findings
Of the 76 patients 18 (24%), including 9 women and 9 men,
were sexually active, defined as having achieved sexual intercourse at least once in the last 2 months. Table 1 lists
sociodemographic, neurological, orthopedic and urological
findings, and QOL scores in sexually active and not sexually
active patients. There was no difference regarding age in
sexually active vs not sexually active patients. All levels
(thoracic to sacral) of myelomeningocele were seen in the 2
groups with significant higher lesions of neurological impairment and more common hydrocephalus in those not
sexually active. The level of urinary incontinence and ambulatory status as a reflection of physical disability were
equal in the 2 groups. In patients not sexually active no
difference was seen regarding urinary incontinence between
men and women, whereas in sexually active patients women
had significantly better urinary continence than men. In
TABLE 1. Sociodemographic, neurological, orthopedic and

urological findings, and QOL scores in sexually active vs not
sexually active adult patients
No. pts
Mean SD age
No. sex (%):
M
F
No. impairment level (%):
T
L1
L3
L45
S
No. shunt (%):
With
Without
No. ambulatory status
(%):
No aid
Walks with walking
appliance
Wheelchair outside
walking aid at home
Wheelchair only
No. living (%):
Independent
In chronic care facility
With close relatives
No. urinary incontinence
(%):
Never
Dry more than 4 hrs
Dry 23 hrs
Always wet
No. gainful employment
(%):
Yes
No
Mean SD QOL:
Physical health
Mental health
Sexually Active
Not Sexually
Active
18
25.8 4.2
58
24 4.5
0.5
9 (21)
9 (27)
33 (79)
25 (73)
0.5
4 (22)
7 (39)
3 (17)
4 (22)
24 (42)
2 (3)
14 (24)
16 (28)
2 (3)
0.05
11 (61)
7 (39)
52 (90)
6 (10)
0.05
4 (22)
5 (28)
2 (3)
20 (25)
2 (11)
7 (12)
7 (39)
29 (50)
0.1
6 (33)
12 (67)
2 (3)
5 (9)
51 (88)
0.001
3 (17)
5 (28)
8 (44)
2 (11)
10 (17)
14 (24)
18 (31)
16 (28)
0.5
7 (39)
11 (61)
17 (29)
41 (71)
0.5
42.4 11.9
54.1 11.3
38.7 7.2
58.6 10.7
0.1
0.1
p Value
evaluation of gainful employment no significant difference

was seen between the sexually active and not sexually active
groups (table 1), whereas in sexually active patients significantly more men than women were employed (table 2).
When evaluating SF-36 for QOL, there was no difference in
physical and mental health subscores between the sexually
active and not sexually active groups (table 1).
Sexual Function Scores in Sexually Active Patients
WSFQ scores in 18 sexually active patients were similar in
men and women with no difference in the 4 domain scores of
desire, arousal, orgasm and satisfaction. The level of impairment was equally distributed between men and women. Two
men had tried to achieve paternity but failed and 1 woman
gave birth. All patients except 2 (1 female and 1 male) wish
to have children in the future (table 2).
Desire and Arousal in Not Sexually Active Patients
Of 58 not sexually active patients, including 25 females and
33 males, 22 (38%) refused to complete particular WSFQ
items to assess desire and arousal. When evaluating sexual
desire and arousal in the remaining 33 not sexually active
patients, 10 (17%) denied any desire, 6 (11%) experience
desire but were not aware of arousal and 20 (35%) reported
desire along with arousal. Of 58 not sexually active patients
37 (64%) wished to have children in the future (table 3).

TABLE 2. Sexual function in sexually active female and male
adults with SB, and impairment level, urinary incontinence,
maternity/paternity rate and wish to have children in future
Mean SD WSFQ (18

pts):
Desire
Arousal
Orgasm
Satisfaction
No. impairment level (%):
T
L3
L5
S
No. urinary leakage (%):
Dry longer than 4 hrs
Dry 23 hrs
Always wet
No. gainful employment
(%):
Yes
No
No. maternity/paternity
(%):
No attempt
Failed
Achieved
No. wish to have children
in future (%):
Yes
No
Females
Males
61.6 8.4
63.9 8.9
0.5
20.2 3.1
16.4 2.6
12.4 2.6
12.4 2.8
22 3
13.8 3.5
14.8 3.5
13.3 1.3
0.2
0.08
0.1
0.4
2 (22)
4 (45)
1 (11)
2 (22)
2 (22)
3 (34)
2 (22)
2 (22)
0.5
7
1
1
1
7
1
0.025
1 (11)
8 (89)
6 (67)
3 (33)
0.025
7
2
1
8
1
p Value
8
1
Nine patients per group.
Overall 53 patients (70%) in the cohort wished to have

children in the future, of whom 37 (70%) were not sexually
active.
DISCUSSION
To our knowledge this is the first study to assess sexual
function in a cohort of adult patients with SB and its impact
of QOL using validated questionnaires. WSFQ was used to
obtain comparable data using a validated instrument for
male and female patients to assess sexual function. Published data on healthy controls in different age groups are
lacking but WSFQ scores in sexually active adult patients
with SB appear slightly higher than in female breast cancer
survivors and hypertensive or kidney cancer populations
and lower than in provided control data (table 4).10 13
In our study WSFQ scores in men and women were equal.
Game et al pointed out that erectile function was disturbed
in sexually active men with SB who were evaluated with the
International Index of Erectile Function9 and women seem
to experience dysfunction to the same degree. This raises the
question of whether a validated instrument such as WSFQ
or the International Index of Erectile Function, which was
developed in anatomically normal adults, could be applied
and transferred to those with myelomeningocele. For better
insight a disease specific survey might be warranted and it
could be compared to generic sexual function questionnaires.
Of our cohort of adult patients with spina bifida 24% were
sexually active and, thus, they were delayed compared to a
contemporary population of high school students in the
United States, where the rate was reported to be 55%.14 At
least 35% of not sexually active patients with SB and, therefore, 55% of the study group experienced desire and arousal.
Compared to other reported data, in which sexual inter-
1613
course and arousal were observed in 25% to 50% and 72% to

75%, respectively, of adults with myelomeningocele,7,9,15,16
sexual function in our group seems to be at the lower end.
Possibly our study population was disabled to a higher degree, requiring more continued orthopedic, neurological and
urological followup than other study cohorts. In accordance
with Decter et al,7 a distinct neurological level could not be
defined to predict the likelihood of better sexual function but
lower and less severe lesions are more favorable. As observed by Verhoef et al, sexually active patients significantly
less often experience hydrocephalus.8 Furthermore, sexually
activity in our study cohort was unrelated to the level of
urinary incontinence and ambulatory status as a measurement of physical disability and dependence of care.
There was no difference in gender distribution among
sexually active patients, suggesting that sexual maturation
and discovery for the female adolescent with physical impairment is as complex as for young men, although the
ability to be aroused seems to be less important to accomplish sexual intercourse in women. Others reported that
female gender is more likely to be associated with sexual
intercourse but their study population included about 50%
patients with SB occulta.8
Of our cohort 70% had the wish to achieve maternity/
paternity but 70% of them were not sexually active. With
increasing age more patients might become sexually active.
However, although young people with SB receive an acceptable level of general sex education, a lack of specific information in regard to SB, in regard to being aware about
decreased infertility in males and other conditions, was
noted in a recent study,16 which might account for the gap in
our observation.
No significant difference was seen in physical and mental
health scores in sexually active and not sexually active patients (table 1). In fact, despite a lower physical health score,
patients who were not sexually active scored higher in mental health compared to sexually active patients and the
general population. In the general 18 to 34-year-old population in the United States mean physical health and mental
health scores were reported to be 53.7 and 48, respectively.17
The lower mental health score in sexually active patients
seems to reflect rather higher psychological and functional
status, closer to the general population. The use of general
TABLE 3. Reported desire, arousal and wish to have children in

future in patients not sexually active
Overall
No answer
No desire
Desire, no
arousal
Desire arousal
Impairment
level:
T
L3
L45
S
Wish to have
children in
future:
Yes
No
Total No.
(%)
No. Females
(%)
No. Males
(%)
58
22 (38)
10 (16)
6 (11)
25
12 (48)
6 (24)
3 (12)
33
10 (30)
4 (12)
3 (9)
20 (35)
4 (16)
16 (49)
Not available
25 (43)
15 (26)
16 (28)
2 (4)
14 (56)
3 (12)
7 (28)
1 (4)
11 (33)
12 (36)
9 (28)
1 (3)
0.1
37 (64)
21 (36)
14 (56)
11 (44)
23 (70)
10 (30)
0.5
p Value
1614

TABLE 4. WSFQ scores in adult patients with SB vs published data
References
Group
11
Watts
Ganz et al12
Watts10
Anastasiadis et al13
Watts11
Watts10
Present series
Women with diabetes

Women with breast Ca
Pts with hypertension
Pts with renal Ca
Female controls
Male female
controls
Adults with SB
Mean SD WSFQ Score
No.
Pts
Mean
Age
Desire
Arousal
Orgasm
30
494
84
84
33
84
42.4
56.2
47.5
57.7
39.4
47.5
18.4 2.9
18.2 3.6
20.57
17.6
20.2 2.9
22.21
15.5 2.9
13.3 3.9
14.57
10.6
15.6 2.7
16.11
14.1 2.9
13 3.2
13.97
10.2
14.8 2.6
14.98
11.3 2.2
11.9 2
11.17
6.6
12.1 1.4
12.25
59.3 8.0
56.5 9.3
60.17
44.8
62.8 6.4
64.95
18
25.8
21.1 3.1
15.1 3.3
13.6 3.2
12.9 2.2
62.7 8.5
health related QOL instruments in cognitively impaired patients might be questioned and validation studies are rare.
In a recent study performed in cognitively impaired patients
with multiple sclerosis it was shown that the reliability and
validity of SF-36 was not decreased,18 leading us to think
that the use of these tools to assess our population was
acceptable.
Our study has several limitations. Patients in our study
population undergo continued multidisciplinary followup at
a tertiary center for pediatrics and, therefore, our cohort
might have more severe disease. Because of cognitive problems, instead of using a self-reported questionnaire we
rather chose to interview these patients, which may have
influenced the answers. On the other hand, questions could
be explained when required and, furthermore, the influence
by parent opinion was limited. Patients in the study showed
a wide range of cognitive function. Since patients with hydrocephalus show a discrepancy between verbal and functional performance,19 limited function might not always
have been recognized as such. Formal IQ measurement as a
more objective assessment was not performed. Finally, it
should be acknowledged that patients who were unable to
answer the validated surveys were excluded from study.
CONCLUSIONS
Our data suggest that in young adult patients with SB sexual
activity is delayed. It is not related to gender, the level of
urinary incontinence or the extent of physical disability but it
more likely occurs in patients with a lower level of neurological
impairment and those without hydrocephalus. In sexually active patients sexual function is disturbed, affecting men and
women to the same degree. Sexual function seems not to affect
health related QOL in these patients.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
14.

QOL quality of life
SB spina bifida
WSFQ Watts Sexual Function Questionnaire
15.
16.
17.
REFERENCES
1.
2.
Lary JM and Edmonds LD: Prevalence of spina bifida at

birthUnited States, 19831990: a comparison of two surveillance systems. MMWR CDC Surveill Summ 1996;
45: 15.
Oakeshott P and Hunt GM: Long-term outcome in open spina
bifida. Br J Gen Pract 2003; 53: 632.
18.
19.
Satisfaction
Totals
Bowman RM, McLone DG, Grant JA, Tomita T and Ito JA:
Spina bifida outcome: a 25-year prospective. Pediatr Neurosurg 2001; 34: 114.
Leger RR: Severity of illness, functional status, and HRQOL in
youth with spina bifida. Rehabil Nurs 2005; 30: 180.
Lemelle JL, Guillemin F, Aubert D, Guys JM, Lottmann H,
Lortat-Jacob S et al: Quality of life and continence in patients with spina bifida. Qual Life Res 2006; 15: 1481.
Blum RW, Resnick MD, Nelson R and St Germaine A: Family
and peer issues among adolescents with spina bifida and
cerebral palsy. Pediatrics 1991; 88: 280.
Decter RM, Furness PD 3rd, Nguyen TA, McGowan M, Laudermilch C and Telenko A: Reproductive understanding, sexual functioning and testosterone levels in men with spina
bifida. J Urol 1997; 157: 1466.
Verhoef M, Barf HA, Vroege JA, Post MW, Van Asbeck FW and
Gooskens RH: Sex education, relationships, and sexuality
in young adults with spina bifida. Arch Phys Med Rehabil
2005; 86: 979.
Game X, Moscovici J, Game L, Sarramon JP, Rischmann P and
Malavaud B: Evaluation of sexual function in young men
with spina bifida and myelomeningocele using the International Index of Erectile Function. Urology 2006; 67: 566.
Watts RJ: Sexual functioning, health beliefs, and compliance
with high blood pressure medications. Nurs Res 1982; 31:
278.
Watts RJ: Sexual function of diabetic and nondiabetic African
American women: a pilot study. J Natl Black Nurses Assoc
1994; 7: 50.
Ganz PA, Rowland JH, Desmond K, Meyerowitz BE and Wyatt
GE: Life after breast cancer: understanding womens
health-related quality of life and sexual functioning. J Clin
Oncol 1998; 16: 501.
Anastasiadis AG, Davis AR, Sawczuk IS, Fleming M, Perelman MA and Burchardt M: Quality of life aspects in kidney
cancer patients: data from a national registry. Support
Care Cancer 2003; 11: 700.
Warren CW, Santelli JS, Everett SA, Kann L, Collins JL and
Cassell C: Sexual behavior among U. S. high school students, 1990 1995. Fam Plann Perspect 1998; 30: 170.
Cass AS, Bloom BA and Luxenberg M: Sexual function in
adults with myelomeningocele. J Urol 1986; 136: 425.
Sawyer SM and Roberts KV: Sexual and reproductive health in
young people with spina bifida. Dev Med Child Neurol
1999; 41: 671.
Ware JE Jr and Kosinski M: SF-36 Physical and Mental
Health Scales: A Manual for Users, 2nd ed. Lincoln, Rhode
Island: Quality Metric 2001.
Marrie RA, Miller DM, Chelune GJ and Cohen JA: Validity
and reliability of the MSQLI in cognitively impaired patients with multiple sclerosis. Mult Scler 2003; 9: 621.
Wills KE, Holmbeck GN, Dillon K and McLone DG: Intelligence and achievement in children with myelomeningocele.
J Pediatr Psychol 1990; 15: 161.
Bladder I
Urachal Anomalies: A Longitudinal
Study of Urachal Remnants in Children and Adults
Richard A. Ashley, Brant A. Inman, Jonathan C. Routh, Audrey L. Rohlinger,
Douglas A. Husmann* and Stephen A. Kramer
From the Department of Urology and Division of Biostatistics (ALR), Mayo Clinic, Rochester, Minnesota
Purpose: Urachal anomalies are often recognized in children but they may persist into adulthood and cause considerable
morbidity. We reviewed the clinical findings associated with these embryological structures and evaluated risk factors for
cancer to define the optimal timing for intervention.
Materials and Methods: We examined the medical records of 176 patients diagnosed with a urachal anomaly between 1951
and 2005. Association of clinicopathological features with malignancy for adults was evaluated using univariate and
multivariate logistic regression studies.
Results: Urachal remnants were identified in 46 children and 130 adults. Of the 46 children 20 (43%) presented with
umbilical drainage and 23 (50%) could be diagnosed by physical examination. A total of 34 children (74%) underwent simple
excision. In contrast, 65 of the 130 adults (50%) presented with hematuria, 78 (60%) required cystoscopy and 53 (41%)
required computerized tomography. A total of 66 adults (51%) required more extensive surgical intervention with partial or
radical cystectomy. Of adult urachal anomalies 66 (51%) were malignant. Patient age 55 years or older and hematuria were
the strongest predictors of urachal malignancy (each p 0.001). Cancer risk increased with advancing adult age.
Conclusions: Urachal anomalies present and progress differently in pediatric and adult populations. Adults are at high risk
for urachal cancer but children usually incur lower morbidity. Urachal lesions noted early in childhood should be excised to
prevent problems in adulthood.
Key Words: urachus, abnormalities, neoplasms, urachal cyst
he urachus is a 3-layered canal that connects the

allantois to the fetal bladder. Descent of the bladder in
month 5 of development stretches the urachus, causing its lumen to obliterate and become the median umbilical
ligament.1 Occasionally this process may be incomplete and
an epithelialized urachal canal may persist into adulthood.2
This leaves the potential for various urachal anomalies,
including cysts, sinus tracts, diverticula and malignancies.2 4 Although some remnants may resolve spontaneously, others lead to umbilical drainage, hematuria or
pain.510 Because most urachal anomalies are presumed to
be benign, they are underrepresented in the literature and
management recommendations are sparse. If urachal remnants are not removed in childhood and they are observed,
exposure to chronic urinary stasis, infection and inflammation in the remnant may predispose the patient to potentially lethal UrC.1113 We identified the clinical and pathological differences in pediatric and adult patients with
urachal processes to define the optimal timing for surgical
intervention.
Study received Mayo Foundation institutional review board approval.

* Financial interest and/or other relationship with Qmed.
Correspondence and requests for reprints: Department of Urology, Mayo Clinic, 200 First St. Southwest, Rochester, Minnesota
55905 (telephone: 507-284-3249; FAX: 507-284-4987; e-mail:
kramer.stephen@mayo.edu).
0022-5347/07/1784-1615/0

Patient Cohort
After receiving approval from the Mayo Foundation Institutional Review Board we reviewed all medical records of
patients seen at our institution between 1951 and 2005 with
a pathological and/or radiological diagnosis of a urachal
remnant. We examined clinical, laboratory, radiographic,
treatment and pathological data. Urachal anomalies were
defined based on pathological assessment of resected specimens and the imaging tests used to delineate the anatomy.
Omphalomesenteric ducts and cutaneous sinus tracts not
associated with the urachal lumen were excluded.
Statistical Analyses
As previously described, univariate and multivariate logistic
regression models were constructed to predict urachal malignancy from the clinical and imaging features that were
present at diagnosis. The variables analyzed were patient
age, patient gender, abdominal pain, voiding dysfunction,
hematuria, lesion size and structure.13 The associations of
these features with the pathological diagnosis (benign vs
malignant) were summarized using the OR and 95% CI.
Patient age was analyzed as a categorical variable and as a
continuous variable in the logistic regression model. This
was done to evaluate changes in cancer risk with increasing
age. Statistical analyses were performed using the SAS
software package. All tests were 2-sided and p 0.05 was
considered statistically significant.
1615
Vol. 178, 1615-1618, October 2007

Printed in U.S.A.
DOI:10.1016/j.juro.2007.03.194
1616
URACHAL ANOMALIES
RESULTS
Study Cohort
A total of 180 patients with a urachal anomaly were identified. Four patients denied permission to use their records.
Our final study group consisted of 46 pediatric patients (25
boys and 21 girls) and 130 adults (86 men and 44 women)
(table 1). Overall median followup was 2.8 years (range 6
months to 44 years). For adults with malignancy median
followup was 3 years (range 3 months to 41 years) and for
adults with benign lesions median followup was 2.8 years
(range 1 month to 44 years). Pediatric patients had a median
followup of 2.3 years (range 1 month to 27 years). Children
presented at a median age of 2.0 years (range newborn to
16.0 years) and 9% were born prematurely, including 3 and
1 at 37 and 32 weeks of gestation, respectively). A simultaneous congenital anomaly was identified in 17 of the 46
children (37%), of which the most common anomaly was
vesicoureteral reflux in 6 (35%). Adults presented at a median age of 54.5 years (range 17 to 89). Two adults had
genitourinary anomalies, including 1 with surgically treated
vesicoureteral reflux and 1 with a duplicated renal collecting
system.
Pediatric Urachal Anomalies
Children often presented with umbilical drainage (20 of 46
or 43%) or were diagnosed incidentally during abdominal
surgery (13 of 46 or 28%) (table 1). These various operations
TABLE 1. Clinical characteristics of 176 patients with urachal

anomalies from 1951 to 2005
Characteristic
Overall
Sex:
M
F
Symptoms:*
Hematuria
Pain
Irritative voiding
Palpable mass
Urinary tract infection
Cellulitis
Umbilical drainage
Incidental finding during surgery
Mucusuria
Diagnosis method:
Ultrasonography
Physical examination
Sonogram
Voiding cystourethrogram
Excretory urogram
Cystoscopy
CT
Magnetic resonance imaging
Treatment:
Observation
Endoscopic procedure
Excision
Partial cystectomy
Cystectomy
Radiation
Incision drainage
Pathological findings:
Patent urachus
Vesico-urachal diverticulum
Urachal sinus
Urachal cyst
UrC
No. Children (%)

46
No. Adults (%)

130
25 (54)
21 (46)
86 (66)
44 (34)
1 (2)
10 (22)
1 (2)
7 (15)
2 (4)
7 (15)
20 (43)
13 (28)
0
64 (49)
35 (27)
16 (12)
6 (5)
6 (5)
4 (3)
10 (8)
23 (18)
6 (5)
10 (22)
23 (50)
5 (11)
5 (11)
0
3 (7)
5 (11)
1 (2)
6 (5)
20 (15)
1 (1)
0
7 (5)
78 (60)
53 (41)
2 (2)
7 (15)
1 (2)
34 (74)
0
0
0
4 (9)
12 (9)
6 (5)
39 (30)
54 (42)
12 (9)
6 (5)
1 (1)
6 (13)
1 (2)
14 (30)
25 (54)
0
0
6 (5)
13 (10)
45 (35)
66 (51)
* Several patients presented with more than 1 symptom.

Several patients underwent more than 1 diagnostic examination.
TABLE 2. Morbidity associated with urachal remnants in adults

and children
Morbidity
No. Children (%)
No. Adults (%)
p Value
Overall
Hematuria*
Irritative voiding
Umbilical drainage
Skin cellulitis
Infected cyst
Pain
Urinary tract infection
46
3 (10)
1 (2)
20 (43)
7 (15)
9 (36)
10 (22)
2 (4)
130
80 (66)
16 (12)
10 (8)
4 (3)
7 (16)
35 (27)
6 (5)
0.001
0.05
0.001
0.02
0.08
0.56
0.99
* Data available on 31 children and 122 adults.

Data available on 25 children and 45 adults.
included gastrointestinal surgery in 8, augmentation cystoplasty in 3, ureteroneocystostomy in 1 and ovarian cyst

excision in 1. In patients who presented with urinary tract
infection Escherichia coli, Klebsiella species and Proteus
species were the most common pathogens. Ten children
(22%) were diagnosed by physical examination alone with
umbilical drainage and/or palpable suprapubic masses leading to the diagnosis. Of the 21 patients who underwent
imaging 10 (48%) were diagnosed by abdominal ultrasonography. The most common urachal anomalies in children
were urachal cyst in 25 of 46 (54%) and urachal sinus in 14
(30%). Nine of the 25 patients (36%) with urachal remnants
had infection at presentation (table 2). Eight of the 9 patients with infection (89%) underwent initial management
with antibiotic therapy and/or percutaneous drainage. No
child presented with urachal torsion, rupture or stone. Of
the 46 children with urachal anomalies 34 (74%) were
treated with simple excision of the remnant along with a
small bladder cuff. No child required umbilectomy. Mean
size of the resected lesion was 2.2 cm (range 0.5 to 5.0).
No evidence of pediatric urachal cancer was identified
after histopathological analysis. Culture evaluation of the
9 infected remnants showed Staphylococcus aureus in 6,
Escherichia coli in 1, Streptococcus viridans in 1 and mixed
growth in 1. Postoperative wound infection occurred in 1 of
34 children (3%). No deaths or complications were immediately attributable to the urachal remnants but 2 of the 7
children who were only observed died of congenital heart
defects.
Adult Urachal Anomalies
Several distinct differences in patient morbidity were evident between adults and children (table 2). Of the 130 adults
64 (49%) and 35 (27%) presented with hematuria and pain,
respectively. Mucosuria was an adult specific symptom in
our cohort (6 of 130 patients or 5%). Unlike the evaluation in
children 78 adults (60%) required cystoscopy, which revealed an obvious mass at the bladder dome in 64 (82%).
Urine cytology was collected from 48 adults and 14 samples
(29%) were positive for cancer.
In the adult cohort 53 patients (41%) underwent CT for a
urachal process. Of the 48 radiographic studies available for
review 21 (44%) showed lesions that were extravesical and
solid, 13 (27%) showed extravesical and cystic lesions, 7
(15%) showed only bladder dome thickening, 3 (6%) showed
solid and cystic lesions, and 3 (6%) showed a solid extravesical component with bladder dome thickening. The urachal
remnant wall was calcified in 10 of the 48 patients (21%). Of
URACHAL ANOMALIES
the 48 cyst cases radiographic evidence of urinary calculi
and vesico-urachal diverticula was present in 1 (2%) and 3
(6%), respectively. One urachal cyst contained a 1.3 cm
calcium oxalate monohydrate stone and 3 of the 6 vesicourachal diverticula (50%) contained calculi of varied composition, including 2 of 50% calcium phosphate and 50% calcium oxalate dihydrate, and 1 of 100% uric acid. No calculi
were associated with concurrent infection.
The pathological outcome in the adult cohort was distinctly different from that in the pediatric group. The most
common urachal lesion identified in the 130 adults was UrC
in 66 (51%). All 66 patients had histological evidence of
adenocarcinoma but 13 (20%) also had concurrent secondary
pathological subtypes.13 High grade cancer was identified in
38 of the 66 patients (58%), while 13 (20%) presented with
metastases at diagnosis. The overall 5-year cancer specific
survival rate for the UrC cohort was 49%.13
Urachal cysts were also commonly seen in adults (45 of
130 or 35%) and the rate of loculated infection was similar to
that in children (table 2). One adult presented with an acute
abdomen because of a torsed urachal cyst and another had
chronic abdominal pain due to a stone in a vesico-urachal
diverticulum.
In contrast to the simple excisions performed in children,
extended partial cystectomy in 52 of the 130 adults (42%)
and radical cystectomy in 12 (9%) were required to treat
cancer or locally destructive infections. Umbilectomy was
performed in 40 adults, including 8 with benign lesions and
32 with urachal cancer. Mean size of the resected lesion was
4.6 cm (range 0.5 to 25.0). Multiple bacterial species were
isolated from the infected cysts of adults, including Staphylococcus aureus, Streptococcus viridans, coagulase-negative
staphylococcus, Bacteroides species, Corynebacterium species, Proteus species and mixed species in 1 each. All patients treated for benign urachal disorders had resolution of
symptoms. Postoperative complications in this group included pulmonary embolism and drug overdose in 1 patient
each. Nine adults died of concurrent comorbid conditions.
Assessing Risk of Urachal Malignancy
UrC occurred only in the adult cohort and, therefore, we
used only adults in our analysis. Urachal lesions with calcifications shown by CT (OR 11.5, p 0.03), patient age 55
years or older (OR 3.6, p 0.001), lesion size greater than
4 cm (OR 2.3, p 0.03) and hematuria at diagnosis (OR 18.3,
p 0.001) were significantly associated with cancer in a
univariate model. After multivariate analysis only 2 factors
(age 55 years or greater and hematuria) remained statistically significant. Close attention was given to the impact of
age on the risk of UrC. Patients with cancer were 22 to 89
years old (median age 61). We noted that for every 1-year
increase in age the risk of cancer increased by 5% (OR 1.05,
95% CI, 1.031.07, p 0.001). Furthermore, for every
10-year increase in age the risk of cancer increased by 61%
(OR 1.61, 95% CI 1.28 2.03, p 0.001).
Urachal Remnants Managed
by Conservative Measures Alone
A total of 24 patients (11 children and 13 adults) were
diagnosed with a urachal remnant and were treated with
observation alone (19) or incision and drainage followed by a
course of antibiotics and observation (5) (table 1). Conserva-
1617
tive management was used if the patient had other comorbidities requiring more attention or the patient/guardian
elected this. We followed this group of patients for a median
of 1.2 years (range 2 months to 21 years). However, despite
the recommendation for annual followup patients were
routinely noncompliant with only 4 of 24 (17%) receiving
more than 5 years of followup. Furthermore, 2 children
(18%) on the observation protocol showed a complication
(urachal abscess) and required surgical treatment. Their
symptoms resolved after complete surgical excision.
DISCUSSION
The natural disease course of urachal remnants is unknown
and rarely studied. Prior groups reviewed the incidence of
various types of remnants and diagnostic modalities.14,15
We present a large series of pediatric and adult patients
with varying types of urachal processes. Urachal anomalies
in children often presented incidentally or with relatively
benign symptoms, whereas a considerable portion of adults
presented with UrC. Treatments for adults and children
were different because they were dictated primarily by the
severity of the underlying pathological condition. Our series
shows that advancing age is associated with an increased
UrC risk. These data suggest that untreated pediatric urachal remnants may under certain circumstances transform
into aggressive urachal malignancies in adulthood. Whether
observation of pediatric urachal remnants is safe must be
established but the rare occurrence of congenital urachal
anomalies makes prospective studies difficult. Given the
poor prognosis associated with UrC, the increasing risk for
cancer with age and the relatively minor morbidity of pediatric urachal excision consideration should be given to removal of the congenital remnants at diagnosis.
In this series male patients outnumbered female patients
in the pediatric and adult cohorts with a ratio of boys to girls
of 1.2:1 and of men to women of 2.0:1. No routine association
between persistent urachal remnants and other medical disorders was identified. Children were more likely to present
with umbilical drainage or have incidental lesions that were
found during surgery for other reasons. Adults often had
hematuria or abdominal pain. Children did not require
costly procedures for diagnosis but adults usually were diagnosed by cystoscopy and/or CT. We identified a considerable number of urachal cysts in the 2 cohorts but the true
patent urachus was unique to children. This unusual entity
was present in 13% of the pediatric cohort but it was absent
in adults. In contrast, vesico-urachal diverticula were rare
in our pediatric cohort, as in other large series.14 However,
6 adults (5%) had this structure, of whom 3 presented with
irritative voiding symptoms. Thus, children with unidentified vesico-urachal diverticula potentially may have symptoms but to our knowledge this possibility has not been
evaluated formally.
Morbidity in children with urachal remnants consisted
largely of umbilical drainage and cutaneous or urinary tract
infection. However, adults were also treated for infected
urachal processes, which suggests that the risk of abscess
formation is not age dependent. As in prior studies, native
skin flora were the usual pathogens.8,14 16 It is important to
note that bacterial culture results of urine and the urachal
remnant were sometimes discordant. This finding warrants
broad-spectrum antibiotic therapy until culture results are
1618
URACHAL ANOMALIES
known from the 2 sites. Interestingly stone formation

seemed to be adult specific. This may be due to long-term
urinary stasis in conjunction with crystallization promoters
such as mucus or desquamated epithelium in the urachal
remnant. Early surgical intervention in childhood at initial
diagnosis could prevent stone formation and abscess development in these patients.
Although pediatric UrC was reported previously,1,1719 it
did not occur in our cohort. In contrast, many adults presented with UrC and the 5-year cancer specific survival rate
was 49%.13 This worrisome finding raises the question of
whether it is safe to only observe pediatric urachal remnants. If completely ignored, urachal remnants may become
a silent, chronic source of inflammation and infection that
ultimately leads to carcinogenesis. This path to malignancy
has been observed with other genitourinary tumors.11,12
Patients with a urachal mass and hematuria who are 55
years or older are at high risk for malignancy.13 Our studies
also suggest that adults with a urachal anomaly should
undergo CT because a solid or calcified supravesical lesion
may be indicative of cancer. It is important to note that 25%
of cystic urachal lesions in adults were malignant. Thus,
cystic urachal masses may not always be benign.20 In this
study we also noted the critical impact of increased age on
increased cancer risk. The data suggest that UrC is a progressive phenomenon. The longer the urachal remnant remains in situ, the higher the risk that it will undergo malignant transformation. Waiting for symptoms to occur is
dangerous because 20% of our patients with UrC initially
presented with metastatic disease.13 Based on our identification of risk factors for cancer and after reviewing the
aggressive nature of this disease our practice is changing.
We believe that early surgical intervention will prevent future life threatening disease.
We acknowledge the inherent weaknesses of a retrospective study. In addition, we recognize that urachal
remnants are rare and our center is a tertiary referral
facility. Referral bias may have influenced the prevalence
of malignancy in the urachal remnants of our adult population. However, because of the rarity of urachal processes, it would be difficult to prospectively study enough
patients to identify the groups at highest risk for malignant transformation.
CONCLUSIONS
Patients with persistent urachal remnants present to the
urologist for evaluation and recommendations. Urachal remnants in children and adults may cause considerable morbidity. Children with urachal anomalies did not present
with cancer in our series. In contrast, adults are at considerable progressive risk for cancer and they may require
more invasive surgery. Given the poor prognosis for adults
with UrC, the lack of criteria to predict which children are at
risk for future carcinogenesis and the relative ease of pediatric urachal excision, we believe that early removal of urachal remnants at first diagnosis is key for preventing future
morbidities.

CT computerized tomography
UrC urachal cancer
REFERENCES
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
15.
16.
17.
18.
19.
20.
Cornil C, Reynolds CT and Kickham CJ: Carcinoma of the

urachus. J Urol 1967; 98: 93.
Sheldon CA, Clayman RV, Gonzalez R, Williams RD and
Fraley EE: Malignant urachal lesions. J Urol 1984;
131: 1.
Herr HW: Urachal carcinoma: the case for extended partial
cystectomy. J Urol 1994; 151: 365.
Siefker-Radtke AO, Gee J, Shen Y, Wen S, Daliani D, Millikan
RE et al: Multimodality management of urachal carcinoma:
the M. D. Anderson Cancer Center experience. J Urol 2003;
169: 1295.
Agatstein EH and Stabile BE: Peritonitis due to intraperitoneal perforation of infected urachal cysts. Arch Surg 1984;
119: 1269.
Goldman IL, Caldamone AA, Gauderer M, Hampel N,
Wesselhoeft CW and Elder JS: Infected urachal cysts: a
review of 10 cases. J Urol 1988; 140: 375.
Guarnaccia SP, Mullins TL and Sant GR: Infected urachal
cysts. Urology 1990; 36: 61.
Horgan PG, Johnson S and Courtney D: Intraperitoneal rupture of an infected urachus. Br J Urol 1994; 73: 216.
MacNeily AE, Koleilat N, Kiruluta HG and Homsy YL: Urachal abscesses: protean manifestations, their recognition,
and management. Urology 1992; 40: 530.
Upadhyay V and Kukkady A: Urachal remnants: an enigma.
Eur J Pediatr Surg 2003; 13: 372.
Chow WH, Lindblad P, Gridley G, Nyren O, McLaughlin JK,
Linet MS et al: Risk of urinary tract cancers following
kidney or ureter stones. J Natl Cancer Inst 1997; 89: 1453.
West DA, Cummings JM, Longo WE, Virgo KS, Johnson FE
and Parra RO: Role of chronic catheterization in the development of bladder cancer in patients with spinal cord injury. Urology 1999; 53: 292.
Ashley RA, Inman BA, Sebo TJ, Leibovich BC, Blute ML,
Kwon ED et al: Urachal carcinoma: clinicopathologic features and long-term outcomes of an aggressive malignancy.
Cancer 2006; 107: 712.
Cilento BG Jr, Bauer SB, Retik AB, Peters CA and Atala A:
Urachal anomalies: defining the best diagnostic modality.
Urology 1998; 52: 120.
Mesrobian HG, Zacharias A, Balcom AH and Cohen RD: Ten
years of experience with isolated urachal anomalies in children. J Urol 1997; 158: 1316.
Berman SM, Tolia BM, Laor E, Reid RE, Schweizerhof SP
and Freed SZ: Urachal remnants in adults. Urology 1988;
31: 17.
Clapuyt P, Saint-Martin C, De Batselier P, Brichard B, Wese
FX and Gosseye S: Urachal neuroblastoma: first case report. Pediatr Radiol 1999; 29: 320.
Huang HY, Ko SF, Chuang JH, Jeng YM, Sung MT and Chen
WJ: Primary yolk sac tumor of the urachus. Arch Pathol
Lab Med 2002; 126: 1106.
Nascimento AF, Dal Cin P, Cilento BG, Perez-Atayde AR,
Kozakewich HP and Nose V: Urachal inflammatory myofibroblastic tumor with ALK gene rearrangement: a study
of urachal remnants. Urology 2004; 64: 140.
Thali-Schwab CM, Woodward PJ and Wagner BJ: Computed
tomographic appearance of urachal adenocarcinomas: review of 25 cases. Eur Radiol 2005; 15: 79.
Results of Bladder Neck Reconstruction

After Newborn Complete Primary Repair of Exstrophy
John P. Gearhart,* Andrew Baird and Caleb P. Nelson
From the Division of Pediatric Urology, Department of Urology, The James Buchanan Brady Urological Institute, The Johns Hopkins
University School of Medicine, Baltimore, Maryland and Department of Urology (CPN), Childrens Hospital Boston, Harvard Medical
School, Boston, Massachusetts
Purpose: We describe the results of modified Young-Dees-Leadbetter bladder neck reconstruction to achieve continence in
patients who underwent complete primary repair of bladder exstrophy as newborns.
Materials and Methods: Using the Johns Hopkins Exstrophy Database we identified patients who underwent bladder neck
reconstruction after having undergone complete primary repair of bladder exstrophy as newborns. We determined patient
characteristics and surgical outcomes.
Results: A total of 30 males and 3 females were referred after complete primary repair of bladder exstrophy, of whom 26
underwent bladder neck reconstruction. Results of complete primary repair of bladder exstrophy were strongly associated
with osteotomy use. Of the patients 19 (58%), including 16 males and 3 females, who underwent complete primary repair of
bladder exstrophy without osteotomy had complications (dehiscence and bladder prolapse), while none of 14 male patients
who underwent complete primary repair of bladder exstrophy with osteotomy had complications, although none were
subsequently continent. Of the 19 patients who had complications after complete primary repair of bladder exstrophy 12
underwent bladder neck reconstruction and total continence was achieved in only 3 (25%). Of the 14 patients with successful
complete primary repair of bladder exstrophy 8 (57%) are dry day and night, 4 (28%) are dry during the day and wet at night,
and 2 (14%) are completely incontinent. Continent children underwent successful complete primary repair of bladder
exstrophy with pelvic osteotomy, all underwent hypospadias repair before age 1 year and none required ureteral reimplantation before bladder neck reconstruction.
Conclusions: Many patients who undergo newborn complete primary repair of bladder exstrophy will require bladder neck
reconstruction. Bladder neck reconstruction is more successful in those in whom complete primary repair of bladder
exstrophy was successful. As in all types of repair, failed initial closure usually results in a bladder that is unsuitable for
bladder neck reconstruction. These patients often require bladder augmentation and a continent stoma to be dry.
Key Words: abnormalities, bladder exstrophy, reconstructive surgical procedures, osteotomy, urinary incontinence
n the last several years there has been interest in CPRE

in the newborn period.1 Various series have been published about outcomes, including the need for ureteral
reimplantation in the first year of life, the number of patients requiring hypospadias repair after this procedure and
the complication rates of this operation.2 6 However, little
has been published about long-term continence results or
the need for BNR in this group of patients. Borer et al
reported that ultimately almost 65% to 70% of patients
undergoing CPRE in the newborn period will require a bladder neck outlet procedure.7
All exstrophy repair methods have merit in experienced
hands. However, the quality and size of the bladder template, the size of the penis, the depth of the urethral grove,
the width of the pubic diastasis, and the experience of the
surgeon and exstrophy team still determine the outcome no
matter what method of repair is chosen at primary closure.
We report the outcomes of patients who underwent CPRE in
the newborn period and who required a bladder neck proce-

* Correspondence: Division of Pediatric Urology, The Brady Urological Institute, The Johns Hopkins Hospital, Marburg 149, 600
North Wolfe St., Baltimore, Maryland 20287.
0022-5347/07/1784-1619/0
dure to be dry. This report comes from a center where CPRE

is not typically performed but where significant experience
with the exstrophy complex exists.

Patients were identified from the institutional review board
approved exstrophy/epispadias database at our institution
(JPG), which is regularly updated and contains data on more
than 858 patients with the exstrophy/epispadias complex
seen since 1975. Medical records were reviewed to determine
patient characteristics, results of CPRE and results of subsequent BNR.
Subjects who presented with failure of newborn CPRE
(dehiscence, prolapse, etc) underwent repeat closure with
anterior innominate and vertical iliac osteotomy at this institution and all remained closed. Urethral reconstruction
was performed when needed for hypospadiac or epispadiac
defects.
All patients were monitored yearly with gravity cystograms under anesthesia to assess increases in bladder
growth. Criteria for BNR were achieving a minimum bladder capacity of 85 cc on cystometry with anesthesia, a stable
1619
Vol. 178, 1619-1622, October 2007

Printed in U.S.A.
DOI:10.1016/j.juro.2007.03.175
1620
BLADDER NECK RECONSTRUCTION AFTER NEWBORN EXSTROPHY REPAIR
bladder on urodynamics and being maturationally ready for

a postoperative timed voiding regimen.
RESULTS
A total of 33 patients, including 30 males and 3 females, who
underwent newborn CPRE were referred to our center since
1996. Of these patients 26 had undergone BNR at the time
of this report.
Results of CPRE
All 16 males and 3 females who underwent CPRE without
osteotomy in the newborn period had complications, including complete dehiscence in 7, bladder prolapse in 7, major
pubic separation in 1, impassable strictures in 2, and total
glans and corporeal loss in 2. They are the subject of a prior
communication.6 All patients were closed in the first week of
life and 9 were closed in the first 48 hours of life. The
patients were immobilized with a spica cast (13) and
mummy wrap (6). No external fixators or traction devices
were used after reconstruction. A total of 13 patients were
closed by fellowship trained pediatric urologists and the
remainder were closed by nonfellowship trained urologists
and pediatric surgeons. All 7 patients with dehiscence, including 6 males and 1 female, 7 with bladder prolapse,
including 6 males and 1 female, and 1 male with major pubic
separation subsequently underwent repeat closure with osteotomy. These patients presented to our unit at a mean age
of 9.6 months.
None of the 14 patients who underwent CPRE with osteotomy had major complications or failure. Two patients
were closed in the first 48 hours of life but they needed
osteotomy because of diastasis exceeding 5 cm. Ten patients
were closed during the first week of life but after age 72
hours and 2 were closed at age 2 weeks. Six patients underwent anterior innominate and vertical iliac osteotomy with
an external fixator device and modified Bucks traction, 4
underwent only anterior innominate osteotomy with external fixators and modified Bucks traction, and 4 underwent
posterior iliac osteotomy with spica cast immobilization. All
were closed by fellowship trained pediatric urologists. None
showed continence after primary CPRE and all were at least
4 years after newborn closure. Seven of 14 patients had a dry
interval of 1 to 2 hours but they still leaked urine. All were
made hypospadiac at primary repair. Six patients required
ureteral reimplantation in the first year of life for febrile
urinary tract infections despite suppressive antibiotics and
confirmation that they did not have outlet obstruction.
These patients presented to our unit at a mean age of 22
months.
Urethral Repair After CPRE
Hypospadias repair was performed in all 14 patients with
successful newborn CPRE at birth, including 11 before and
3 (under testosterone stimulation) after referral. The urethra was located at the penoscrotal junction in all patients.
Fistulas occurred in 5 patients (40%) and all were repaired
successfully, although 1 male who underwent hypospadias
repair at our institution required 3 attempts. Full-thickness
skin graft and/or buccal mucosa was required to replace all
or part of the urethra in 14 patients. In addition, 5 patients
required placement of tissue expanders before Cantwell-
Ransley repair. A total of 14 patients have undergone

Cantwell-Ransley epispadias repair and 2 are being observed. There were 2 fistulas and 1 patient had meatal
stenosis. Two females required tubed bladder grafts to replace the urethra at repeat closure.
BNR After CPRE
Of the 19 patients with failed CPRE 12 underwent BNR at
a mean age of 5.9 years with a mean bladder capacity of
88 cc. All were male and no females have undergone BNR at
this time. Total daytime and nighttime dryness was
achieved in only 3 patients (25%) with 2 requiring DDAVP
and 9 who remained incontinent. In the successful CPRE
group mean age at BNR was 4.9 years with a mean bladder
capacity of 115 cc. Eight of these patients (57%) were dry for
3 hours during the day and dry at night, including 3 on
desmopressin, 1 (7%) is dry during the day and wet at night,
3 (21%) are dry for 2 hours during the day and wet at night
at less than 1 year after BNR and 2 (14%) are completely
incontinent. Of those who are dry day and night all underwent successful CPRE in the newborn period, all underwent
osteotomy, none required ureteral reimplantation before
BNR and all underwent hypospadias repair before age 1
year.
DISCUSSION
As mentioned, the first subset of 19 patients was described
in a prior communication.5,6 However, this current study
offers a unique comparison between 2 sets of patients with
CPRE, that is those who did well at primary closure and
those who did not. Only 3 of the 12 patients (25%) who
underwent BNR became dry day and night. This coincides
with previous data showing that with failed primary closure
only 60% of patients have adequate capacity for a bladder
outlet procedure and less than half of them become dry.8
Regardless of the type of failure (dehiscence, prolapse or
outlet obstruction causing upper tract changes) multiple
procedures decrease the ability of the bladder to grow, not
only before, but also after an outlet procedure.9
Irrespective of the type of repair chosen primary soft
tissue and bony pelvic closure must be protected in the
postoperative state to ensure the best chance of success. At
our institution after MSRE or combined closure (bladder
closure with Cantwell-Ransley epispadias repair) immobilization is ensured with modified Bryants traction or, if osteotomy was performed, with external fixation and Bucks
traction. A recent report by Meldrum et al shows that the
success rate of primary closures is almost 98% if appropriate
immobilization techniques and traction are used as an adjunct to primary repair.10 In addition, careful control of
postoperative movement and pain is performed with injectable agents and a tunneled epidural catheter left in place for
almost 2 weeks in all patients. This helps the newborn adapt
to its situation and allows rest without excessive movement.
In addition, these adjuncts markedly decrease bladder
spasms after closure. Nonetheless, in some series of successful CPRE in the newborn period spica casts were used as the
preferred technique of immobilization even when osteotomy
was not done.3,4
In this select group of 19 failed cases all repeat closures
were successful. All 3 patients who were dry after an outlet
procedure had undergone transverse innominate and verti-

cal iliac osteotomy at repeat closure and they were in traction for 6 weeks. Thus, even when treatment fails, osteotomy
can salvage some of these patients and allow them to continue on the reconstructive pathway.11
All patients in the successful CPRE group underwent
osteotomy, including an anterior and posterior approach in
10 and 4, respectively. Often with an early closure (less than
48 hours) the pelvis is malleable and does not require osteotomy. However, in 2 patients with early closure the surgeons believed that a 5 cm diastasis was extreme and elected
osteotomy at this early stage with a successful result. The 2
patients who were closed at 2 weeks because of medical
conditions underwent transverse innominate osteotomy and
vertical iliac osteotomy with 4 weeks of fixation and traction
without complications. Again, even delayed closure can have
a successful result with appropriate postoperative care. Of
the 10 patients who were closed after 72 hours of life but at
age younger than 1 week 4 had a posterior approach with a
spica cast and 6 had an anterior innominate approach with
fixators and traction. Thus, osteotomy and immobilization
have a positive effect regardless of the method of repair
chosen and often they can enhance the chance of primary
closure success.12
Ureteral reimplantation after newborn repair has been
reported in up to 50% of patients undergoing CPRE.1 Recurrent breakthrough infections typically prompt this intervention along with outlet obstruction and upper tract deterioration. Regardless of the reason none of the patients in this
series who became continent after BNR underwent this procedure before BNR. Although prior publications show that
reimplantation can be successfully performed before BNR
with acceptable outcomes and low complications,13 it is our
opinion that each time the bladder is violated there is another insult to smooth muscle cells and ultimate capacity
could be decreased. After primary closure regardless of technique patients must be followed carefully to detect the appearance of significant reflux. When reflux develops, endoscopic treatment may provide an alternative to open
reimplantation. One of us (JPG) has used dextranomer/
hyaluronic acid bulking agent in 5 patients with breakthrough febrile urinary tract infections with good success,
that is in 2 of the 5 reflux was cured, and reflux decreased to
grade 1 in 2 and to grade 2 from grade 3 in 1. This substance
did not hinder BNR or reimplantation in the 3 patients who
subsequently underwent BNR and it obviated another open
bladder procedure. Additionally, some groups have recommended ureteral reimplantation at primary closure.14 Great
care should be taken if this is attempted because the ureteral reimplants may need revision and may even need to be
moved in the event that BNR is needed as the child ages,
bladder capacity becomes larger and the anatomical relationships of the trigone change.
In addition to osteotomy and the lack of ureteral reimplantation, all patients with successful CPRE who were
referred to our center underwent hypospadias repair before
age 1 year. In the 4 patients repaired at our institution a flap
of testosterone stimulated penile skin was used in the manner described by Hafez and El-Sherbiny3 but the glans was
split and not tunneled. We remain totally unconvinced that
sacrificing the urethral plate is required to get the posterior
vesicourethral unit into the pelvis. It does not increase ultimate penile length and it effectively sacrifices straight
forward Cantwell-Ransley repair for a difficult penoscrotal
1621
hypospadias repair with a paucity of skin. Our 40% fistula

rate in our successful CPRE group attests to this fact. No
matter what method of exstrophy or epispadias repair is
chosen, early urethral repair is best. In a recent study by
Kufner et al in a group of patients with MSRE bladder
capacity increased more and at a more rapid rate in those
undergoing epispadias repair by ages 6 to 12 months.15 This
likely also holds true in patients with exstrophy who are
made hypospadiac after CPRE. Additional outlet resistance
allows more urine to pool in the bladder and generates an
increase in capacity.
Long-term outcomes data after CPRE are difficult to find.
Several groups have reported complete continence in some
patients and dry intervals in others.1,2,4,7 What is clear from
our select, referred group of patients with successful CPRE
closure is that none of them became continent of urine after
newborn repair. This supports data from a recent thorough
study by Borer et al, in which almost 65% of patients required BNR after newborn CPRE.7 The results of the 2
studies suggest that even after well executed newborn
CPRE without complications and with careful followup BNR
is needed in a significant number of patients.
Gearhart et al recently reported a series of 95 patients
with exstrophy with long-term followup who underwent
MSRE with abdominal wall, bladder, posterior urethral closure and anterior urethral closure well upon the penile
shaft.16 Of the newborns 18 underwent osteotomy at primary closure. All underwent Cantwell-Ransley epispadias
repair in early life and BNR when capacity was adequate
and the child was ready to become dry. The records of 67
patients with a minimum of 5 years of followup were reviewed, of whom 47 (70%) became dry for 3 hours during the
day and dry at night, while 7 (10%) were dry for 3 hours
during the day but wet at night, 7 were incontinent and 6
had failed BNR and underwent continent diversion. Although comparisons of exstrophy techniques are difficult,
certain conclusions can be drawn. In this current series of
patients with successful CPRE those who had a bladder
capacity of greater than 100 cc at BNR became continent
more often and time to dryness was also more rapid than in
those with a capacity of less than 100 cc. This was also the
case for the mentioned series of MSRE repairs done by a
single surgeon.6
Interestingly the current group of successful CPRE repairs done by various surgeons yielded acceptable continence rates after referral for BNR. Certainly as time goes by
we think that some patients with CPRE who underwent
BNR less than 1 year ago will become dry, increasing the
overall continence rate. If only 2 of 3 patients become dry,
the overall continence rate would approach 70%. This continence experience with patients with CPRE referred for
BNR parallels that of patients with successful MSRE referred for BNR, in whom overall continence rates are good.17
By following basic tenets during primary closure and followup either technique can yield a bladder with the potential to achieve continence after BNR.
CONCLUSIONS
Many patients who undergo CPRE in the newborn period
will require a bladder outlet procedure. Continence rates
after BNR are better in those in whom primary repairs are
successful, as in the MSRE group. Certain criteria help
1622
predict improved chances of success after BNR in this select

group of referred patients with CPRE. Surgeons who care for
patients with exstrophy should keep this in mind when
confronted with a patient with CPRE requiring continence
surgery.
11.
14.
BNR bladder neck reconstruction

CPRE complete primary repair of exstrophy
MSRE modern staged repair of exstrophy
12.
13.
15.
REFERENCES
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
Grady RW and Mitchell ME: Complete primary repair of exstrophy. J Urol 1999; 162: 1415.
Hammouda HM and Kotb H: Complete primary repair of bladder exstrophy: initial experience with 33 cases. J Urol 2004;
172: 1441.
Hafez AT and El-Sherbiny MT: Complete repair of bladder
exstrophy: management of resultant hypospadias. J Urol
2005; 173: 958.
Hafez AT, El-Sherbiny MT and Goni MA: Complete repair of
bladder exstrophy: preliminary experience with neonates
and children with failed initial closure. J Urol 2001; 165:
2428.
Gearhart JP: Complete repair of bladder exstrophy in the
newborn: complications and management. J Urol 2001;
165: 2431.
Gearhart JP and Baird AD: Failed complete repair of bladder
exstrophy: insights and outcomes. J Urol 2005; 174: 1669.
Borer JG, Gargollo PC, Hendren WH, Diamond DA, Peters CA,
Atala A et al: Early outcome following complete primary
repair of bladder exstrophy in the newborn. J Urol 2005;
174: 1674.
Gearhart JP, Ben-Chaim J, Scortino C, Sponseller PD and
Jeffs RD: Multiple reoperative bladder exstrophy closures:
what affects the potential of the bladder? Urology 1996; 47:
240.
Peppas DS, Tchetgen M-B, Lee BR, Jeffs RD and Gearhart JP:
A quantitative histological analysis of the bladder in classical bladder exstrophy in various stages of reconstruction
utilizing color morphometry. In: The Exstrophy-Epispadias
Complex: Research Concepts and Clinical Applications. Edited by JP Gearhart and R Mathews. New York: Kluwer
Academic/Plenum Publishers 1999; p 41.
Meldrum KK, Baird AD and Gearhart JP: Methods of pelvic
and extremity immobilization following bladder exstrophy
closure: complications and impact on success. Urology 2003;
62: 1109.
16.
17.
Gearhart JP: Failed bladder exstrophy closure: evaluation and

management. Urol Clin North Am 1991; 18: 687.
Gearhart JP, Peppas DS and Jeffs RD: The failed exstrophy
closure: strategy for management. Br J Urol 1993; 71: 217.
Mathews R, Hubbard J and Gearhart JP: Ureteral reimplantation before bladder neck plasty in the reconstruction of
bladder exstrophy: indications and outcomes. Urology 2003;
61: 821.
Garat GM, Dellapena E, Caffarati Y and Villavicencio: Prevention of vesicoureteral reflux at the time of complete primary
repair of the exstrophy-epispadias complex. Int Urol Nephrol 2004; 36: 211.
Kufner M, Mathews RI and Gearhart JP: Impact of epispadias
repair on bladder capacity in boys with classic bladder
exstrophy. Presented at Joint Meeting of the European
Association of Urology/Section of Urology, American Academy of Pediatrics, Uppsala, Sweden, June 1518, 2005.
Gearhart JP and Baird A: The modern staged repair (MSRE) of
bladder exstrophy: contemporary series. Presented at annual congress of European Society for Paediatric Urology,
Athens, Greece, April 2729, 2006.
Surer I, Baker LA, Jeffs RD and Gearhart JP: Modified Young
Dees Leadbetter bladder neck reconstruction in patients
with successful primary closure elsewhere: a single institutional experience. J Urol 2001; 65: 2438.
EDITORIAL COMMENT
The experience of these authors with patients referred after
exstrophy repair using the Mitchell technique demonstrates
that these patients have a viable bladder. Furthermore,
many of them can achieve continence with formal BNR even
if they are incontinent after initial repair. Whether continence can be achieved in a similar group of patients using
endoscopic techniques, biofeedback or other reported methods remains to be seen.
The association of successful initial closure with osteotomies deserves emphasis, as does the association of successful initial closure with formal pediatric urology fellowship training. At our institution we use osteotomies liberally
in conjunction with primary exstrophy closure. As a group,
we should encourage the treatment of these patients by
fellowship trained pediatric urologists since, as this study
suggests, experience matters.
Richard Grady
Pediatric Urology
Childrens Hospital and Regional Medical Center
Division of Urology
University of Washington School of Medicine
Seattle, Washington
A Comparison of the Monti and Casale (Spiral Monti) Procedures

Jeffrey A. Leslie, Mark P. Cain,*, Martin Kaefer, Kirstan K. Meldrum, Andrew M. Dussinger,
Richard C. Rink and Anthony J. Casale
From the Division of Pediatric Urology, Indiana University School of Medicine, James Whitcomb Riley Hospital for Children,
Indianapolis, Indiana
Purpose: We present our long-term followup and comparison of outcomes between the Monti and Casale (spiral Monti)
procedures in a large group of children and young adults.
Materials and Methods: A retrospective chart review was done, including all patients undergoing the Monti or Casale
procedure at our institution with a minimum followup of 6 months. Age at surgery, the bowel segment used, stomal location,
the number and type of revisions or endoscopic procedures required after channel creation, problems catheterizing and
channel continence were documented, and a database was created.
Results: Of 188 patients identified with at least 6 months of followup 109 underwent a Monti procedure, while 79 underwent
a Casale procedure. Patient age at surgery was 10 months to 31 years (mean 10.2 years). Mean followup was 43 months for
the entire cohort, and 47.2 and 37.2 months for the Monti and Casale groups, respectively. A total of 43 open revisions were
required in 36 patients (19.1%). Stomal revisions accounted for 18 procedures, while subfascial revisions accounted for 25 in
17 (9.0%) and 21 (11.2%) patients, respectively. A total of 21 endoscopic procedures requiring anesthesia were performed in
17 patients (9.0%). In the Monti group stomal revision was required in 11 patients (10.1%), while subfascial revisions were
required in 9 (8.3%). In the Casale group stomal revision was required in 6 patients (7.6%), while subfascial revisions were
required in 12 (15.2%). Of the channels 98% were completely continent at the stoma.
Conclusions: In a large population of children and young adults we report durable and reliable long-term results with the
Monti and Casale procedures, including continence at the stoma. The only significant difference noted between the 2
procedures was a higher incidence of subfascial revisions for umbilical stomas in each group. The need for subfascial revision
is highest in spiral Monti channels placed in the umbilicus.
Key Words: urinary bladder, neurogenic; stomas; urinary catheterization; appendix; urinary diversion
he Mitrofanoff principle for creating a continent catheterizable stoma using appendix has been a mainstay
in the armamentarium of pediatric urologists and reconstructive surgeons since it was originally described in
1980.1 Applications of this technique have expanded to include the use of ureter, tapered ileum, stomach, tubularized
bladder flaps and transverse tubularized bowel, as originally
described by Yang2 and Monti et al,3 and later modified by
Casale.4 Interest in these other tissues stems from the occasional lack of a suitable appendix or more commonly the
desire to use the appendix for the MACE procedure.5 The
Yang-Monti and Casale (spiral Monti) techniques for creating a continent catheterizable channel using a transversely
tubularized bowel segment have been used extensively in
pediatric reconstruction for several years, especially when
the appendix is not available or usable. Although it is generally accepted among pediatric urologists that the appendix
is the ideal channel for Mitrofanoff creation, increasing use
of the MACE procedure has expanded the need for alternative tissues for the Mitrofanoff channel.
* Correspondence and requests for reprints: Division of Pediatric

Urology, Indiana University School of Medicine, James Whitcomb
Riley Hospital for Children, 702 Barnhill Dr., No. 4230, Indianapolis, Indiana 46202 (telephone: 317-278-7446; FAX: 317-274-7481;
e-mail: mpcain@iupui.edu).
Financial interest and/or other relationship with BoehringerIngelheim.
0022-5347/07/1784-1623/0
When the Monti channel does not provide adequate channel

length to traverse the abdominal wall without tension, the
Casale procedure has become the best alternative. To our
knowledge there are no published comparisons to date of the
classic Monti tube to the Casale method with respect to longterm outcomes. We have routinely performed the Yang-Monti
or Casale procedure as an isolated procedure in select cases but
more commonly in conjunction with the MACE procedure
and/or bladder augmentation. We report long-term followup
and a comparison of outcomes between the Monti and Casale
procedures in a large cohort of young patients.
A retrospective chart review was done, including all patients
undergoing the Monti or Casale procedures at our institution with a minimum followup of 6 months. Age at surgery,
the bowel segment used, stomal location, the number and
type of revisions or endoscopic procedures required after
channel creation, problems catheterizing and continence of
the channel were documented and a database was created.
All postoperative procedures related to the channel that
required anesthesia were included in the analysis.
Open revision procedures were classified as stomal or
subfascial, of which the latter included revisions for angulation leading to catheterization problems or false passage
as well as bladder level problems such as inadequate tunnel
1623
Vol. 178, 1623-1627, October 2007

Printed in U.S.A.
DOI:10.1016/j.juro.2007.03.168
1624
COMPARISON OF MONTI AND CASALE (SPIRAL MONTI) PROCEDURES
length. Open revisions for stomal stenosis or granulation

tissue that remained extraperitoneal were classified as
stomal. Endoscopic procedures were classified as diagnostic
if the channel was easily catheterizable intraoperatively and
cystoscopically appeared normal. Endoscopy was classified
as therapeutic if it required dilation, endoscopic incision of a
stricture or tissue web, catheter placement over a wire or
difficult catheterization with an indwelling catheter secured
for several days postoperatively to achieve passive dilation.
Statistical analysis was performed using SAS. Chisquare analysis was used for comparisons of complications
between Monti and Casale groups, and between umbilical
and RLQ stomal sites. Fishers exact test was used for subgroup comparisons.
RESULTS
Between 1997 and 2004, 201 patients underwent a Monti or
Casale procedure for creation of a catheterizable urinary
channel. Of the 188 patients identified with at least 6
months of followup postoperatively 109 (58%) underwent a
single Monti procedure, while 79 (42%) underwent a Casale
procedure. No double Monti channels were created and
when possible the channel was implanted into the bladder
rather than into an intestinal segment. The channel was
constructed from ileum in all except 3 patients, in whom
sigmoid colon was used. Patient age at surgery was 10
months to 31 years (mean 10.2 years). Mean followup was 43
months (range 6.5 to 97) for the entire group, and 47.2
(range 6.7 to 97) and 37.2 months (range 6.5 to 86.3) for the
Monti and Casale subgroups, respectively.
In the entire group a total of 43 open revisions were
required in 36 patients, yielding an open revision rate of
19.1%. There were a total of 18 stomal revisions in 17 patients (9.0%) and 25 subfascial revisions in 21 (11.2%). One
of these patients required subfascial revision and 2 stomal
revisions but is now completely continent at the stoma and
catheterizes without difficulty. One patient required stomal
revision and subsequent subfascial revision. A total of 21
endoscopic procedures requiring anesthesia were performed
in 17 patients (9.0%). Six of these procedures were purely
diagnostic for reported catheterization difficulties, while 15
were therapeutic. One of these patients required a diagnostic and therapeutic endoscopic procedure, while 1 required 1
diagnostic and 3 therapeutic procedures. These 2 patients
are now completely continent at the stoma and have no
difficulty catheterizing. A total of 143 (76.1%) patients required no channel related surgery during followup and 152
(80.9%) required no open revisions.
A total of 22 open revisions in 20 patients (18.3%) were
required in the 109 undergoing the Monti procedure compared to 21 revisions in 16 (20.3%) of the 79 who underwent
creation of a spiral Monti. In the Monti group 11 patients
(10.1%) required stomal revision and 9 (8.3%) required subfascial revisions with 2 of the latter requiring a second
revision. Nine Monti cases (8.3%) required endoscopy postoperatively. One of these patients required a therapeutic
and a diagnostic endoscopic procedure, while 5 others required a single therapeutic endoscopy and 3 underwent
purely diagnostic cystoscopy via the channel. Of all 109
patients with a Monti channel 83 (76.1%) required no postoperative procedures related to the channel and 89 (81.7%)
required no open revisions.
In the Casale group stomal revision was required in 6 patients (7.6%), of whom 1 required 2 revisions. Subfascial revisions were required in 12 patients (15.2%), of whom 2 required
a second revision. Endoscopy was performed in 8 patients
(10.1%), of which 1 was purely diagnostic. Six patients underwent a single therapeutic endoscopy, of whom all currently
catheterize successfully. One patient required 1 diagnostic and
3 therapeutic endoscopies, and is now catheterizing successfully. Of the 79 patients with a Casale channel 60 (75.9%)
required no postoperative procedures related to the channel
and 63 (79.7%) required no open revisions.
Stomas were placed at the umbilicus in 95 patients
(50.5%) and in the RLQ in 80 (42.6%). Stomal location could
not be determined from the operative report in 9 patients
and it was the left lower quadrant in 4. Of the 95 umbilical
channels a total of 27 open revisions were required in 21
(22.1%) patients, including only 8 at the stomal level in 7
(7.4%) patients and the other 19 in 16 patients (16.8%) at the
subfascial level. Three of these 16 patients required 2 subfascial revisions, while the other 12 required 1 revision each.
One of the 7 patients requiring stomal revision required a
second revision. A total of 13 endoscopic procedures were
required in this group of patients, of which 10 were therapeutic and 3 were diagnostic.
Of the 80 RLQ stomas a total of 14 open revisions were
required in 13 patients (16.3%). Eight patients (10.0%) required a single stomal revision, while 6 subfascial revisions
were required in 5 (6.3%). Seven endoscopic procedures were
required in this group, of which 4 were therapeutic and 3
were diagnostic.
At last followup 169 of the 188 patients (89.9%) reported no
significant difficulty catheterizing the urinary channel, while
14 continued to report some difficulty but were able to successfully catheterize as directed. Seven of these patients required
no surgical intervention postoperatively and all were continent
at the stoma. The other 7 patients each required 1 postoperative intervention, including subfascial revision in 1, and stomal
revision, diagnostic cystoscopy and therapeutic cystoscopy in 2
each. One of these patients had occasional leakage at the
stoma, while the remaining 6 were completely continent. Four
channels became obliterated or could not be catheterized, of
which 1 was recently revised. An additional channel was abandoned due to poor patient compliance, obesity and subsequent
lack of use. Therefore, at last followup 183 patients (97.3%)
were able to catheterize successfully.
Continence via the channel could be assessed from
records in all except 2 patients. Four patients (3.2%) reported rare or occasional leakage at the stoma but none had
frank leakage. Two of these patients have a Casale channel
to the umbilicus, while the other 2 have a Casale channel to
the left lower quadrant and a Monti channel to the umbili-
TABLE 1. Complications of Monti vs Casale procedures

No. Monti (%)
Overall
Difficult or unable to
catheterize
Endoscopy with anesthesia
Stomal revision
Subfascial revision
Open revision
Any surgery
No. Casale
p Value
(%)
(chi-square test)
109
10 (9.2)
79
9 (11.4)
0.62
9 (8.3)
11 (10.1)
9 (8.3)
20 (18.3)
26 (23.9)
8 (10.1)
6 (7.6)
12 (15.2)
16 (20.3)
19 (24.1)
0.66
0.56
0.14
0.74
0.98

TABLE 2. Complications of umbilical vs RLQ stomal sites
Overall
Endoscopy with
anesthesia
Stomal revision
Subfascial revision
Open revision
Any channel
revision surgery
No. Umbilical
(%)
No. RLQ (%)
p Value
(chi-square test)
95
10 (10.5)
30
6 (7.5)
0.49
7 (7.4)
16 (16.8)
21 (22.1)
25 (26.3)
8 (10.0)
5 (6.3)
13 (16.3)
17 (21.3)
0.54
0.05
0.33
0.43
cus. Including all patients with functional channels at last

followup in whom continence could be determined, 177 of
181 (97.8%) were completely continent at the stoma.
DISCUSSION
The use of transversely tubularized intestinal segments in
pediatric and adult reconstructive surgery has blossomed in
recent years. The focus of this study was to document the
long-term success rate and compare the traditional Monti
channel to the Casale channel. To our knowledge this is the
largest such series in the literature and the only to compare
the Casale technique to the traditional Monti method.
The ureteral Mitrofanoff procedure has been used successfully, primarily in patients with 1 poorly/nonfunctioning
kidney, but it is fraught with higher complication rates of
incontinence6 and stomal stenosis.7 Tubularized bladder
flaps, ie continent vesicostomy, have also been used successfully in children, as described by Rink et al.8 However, the
stomal stenosis rate reported with this procedure is
high.8 10 In addition, this procedure requires sacrificing a
portion of the bladder to form the channel, which can be
problematic if bladder capacity is small or the bladder wall
is thickened and noncompliant, as is often the case when a
catheterizable abdominal wall channel is desired.
As described independently by Yang2 and Monti et al,3
transversely tubularized bowel segments provide the advantage of the mobility and availability of a small segment of
bowel (usually ileum), while eliminating some inherent
problems attributable to tapered segments. Transverse reconfiguration of the bowel segment allows catheterization to
occur parallel to the valvulae conniventes, rather than perpendicularly, as in the case of tapered bowel segments,
thereby decreasing the incidence of catheterization difficulties. In addition, this technique can be used simultaneously
with augmentation enterocystoplasty or intestinal neobladder creation without the need for additional bowel anastomosis.
In the largest series in the literature involving the YangMonti procedure primarily as a continent urinary stoma
1625
Monti et al reported followup on 55 channels, of which 9

were used for MACE and 46 were used for urinary drainage.11 Seven of these tubes were double Monti channels and
the other 48 were single Monti channels. Average followup
in this series was only 4 to 7 months with continence reported in 91% of patients. Only 1 patient required revision
for stomal stenosis. One of the 5 patients with incontinence
was rendered dry by adjusting the catheterization routine,
while the other 4 required 2 open revisions and 2 endoscopic
procedures, respectively.11
In another large series Castellan et al reported a comparison among different types of channels for urinary and fecal
incontinence, including 45 Monti urinary channels, with a
mean followup of 38 months (minimum 7).12 Four of these
channels were double Monti channels, while the others were
single Monti channels. Three patients (7%) required channel
replacement due to complete fibrosis and 3 (7%) had stomal
incontinence. Two of those patients with incontinence had sigmoid single Monti channels and 1 had a double Monti channel.
The investigators reported problematic catheterization in 4
patients (9%), of which only 1 required stomal revision. Including all patients in this series who received a Monti channel for
bowel or bladder incontinence, stoma related problems occurred in 22%, including complete channel fibrosis in 6%, revision surgery using anesthesia in 8%, and 8% in whom problems resolved with prolonged catheterization.
Narayanaswamy et al reported their results with 94 Mitrofanoff procedures, of which 25 were Monti channels.13 Mean
followup for the Monti group was 25 months (minimum 6). Of
note 17 of these 25 urinary channels were double Monti channels. Of the 25 patients 15 (60%) had problems with catheterization with conduit stenosis and/or diverticular pouch
formation in 13. Interestingly 6 of 7 patients with pouch
formation had a double Monti channel. Overall 23 of 25
patients successfully catheterized at the time of the report
and only 3 of 25 had stomal leakage. The group reported no
difference in stomal stenosis rates between appendiceal
channels and Monti channels on final analysis.
To our knowledge no data exist on the Casale channel as
a urinary conduit for intermittent catheterization beyond
the initial description of the procedure by Casale.4 In that
initial report at a maximum followup of 1 year all 8 patients
were dry and catheterizing easily, although 1 required 1
open revision due to peristomal abscess complications, leading to distortion of the stoma.
Table 1 lists postoperative complications in our patient
cohort, comparing the Casale technique to the standard
Monti procedure. Notably none of the complication rates
were significantly different between the 2 procedures and
the rates were similar to those reported in other smaller,
earlier series except a slightly higher subfascial revision
rate in the spiral group, which did not attain statistical
TABLE 3. Complications of Monti vs spiral Monti procedures based on channel location
Overall
Endoscopy
Stomal revision
Subfascial revision
Open revision
Any surgery
No. Monti to
RLQ (%)
No. Monti
to Umbilicus (%)
No. Casale to
RLQ (%)
No. Casale to
Umbilicus (%)
p Value
(Fishers exact test)
55
4 (7.3)
6 (10.9)
4 (7.3)
10 (18.2)
13 (23.6)
44
5 (11.4)
3 (6.8)
5 (11.4)
8 (18.2)
11 (25.0)
25
2 (8.0)
2 (8.0)
1 (4.0)
3 (12.0)
4 (16.0)
51
5 (9.8)
4 (7.8)
11 (21.6)
13 (25.5)
14 (27.5)
0.92
0.94
0.096
0.58
0.77
1626
A, longer extravesical segment of a Monti or Casale channel, which is usually required to reach umbilicus, can lead to catheterization
problems such as false passage and diverticuli, especially near detrusor hiatus. B, minimizing extravesical portion of the channel decreases
subfascial complications.
significance (p 0.14). Interestingly when comparing the 2

most common stomal sites (umbilicus and RLQ), the only
significant difference was a higher subfascial revision rate in
the umbilical group (p 0.05, table 2). This may be a confounding issue with the higher rate of subfascial revisions in
the Casale group, although 1 of us performed most of these
procedures due to a preference for placing the urinary channel at the umbilicus, allowing easier access for catheterization, especially for wheelchair bound patients. Indeed, 51 of
the 95 umbilical stomas (53.7%) in these patients were
Casale channels compared with only 25 of the 80 RLQ channels (31.3%).
Table 3 shows subgroup analysis based on stomal location
and channel type. Although differences in the subfascial
revision rates did not attain statistical significance, only 1 of
25 Casale channels (4%) to the RLQ required revision compared with 11 of 51 Casale channels (21.6%) to the umbilicus
(Fishers exact test p 0.090) The lack of statistical significance was likely due to the small number of patients in each
group. Thus, it appears that a Casale channel to the umbilicus is associated with a 5-fold higher deep or subfascial
revision rate, perhaps due to the longer channel distance
required to reach the umbilicus, especially in the small,
contracted bladder (part A of figure). This complication can
be minimized by placing the stoma in the RLQ in patients
with a small bladder, thereby minimizing the nonfixed portion of the channel between the bladder hiatus and fascia
(part B of figure). This must be weighed against the benefit
of an umbilical continent urinary stoma, which is easier to
access for catheterization. This can be an important factor to
consider for many patients with neurogenic disorders, who
are often wheelchair bound or immobile.
With regard to the current study one should note that the
mean followup in the Casale group is almost 1 year less than
that in the standard Monti group, as would be expected since
the Casale technique was not described until 1999. This
difference in followup may impact the incidence of complications with a greater long-term risk for subfascial reoperative procedures in the Casale group.
Continence via abdominal wall stoma at last followup in
this series was almost 98% and has also been reported to be
greater than 90% in most other major series. It is worthwhile to note that in many smaller series revision surgery
for incontinence was often required for channels implanted
into a bowel segment. Although on occasion channel placement into an intestinal segment cannot be avoided, we have
traditionally made every attempt to implant the conduit into
the native bladder. If detrusor is not accessible or available,

such as in a neobladder or secondary augmentation, colon
should be the next choice because creation of a submucosal
tunnel is easier in colon than in small bowel. Additionally,
we have noted thinning of the intestinal segments with time,
which may increase the risk of catheterization problems,
perforation or incontinence. Because of this, we advocate
securing the submucosal tunnel with permanent sutures to
the posterior fascial wall to prevent loss of the continence
mechanism or angulation.
Approximately 10% of patients with Casale or single
Monti tubes experience some difficulty with catheterization
or they have obliterated channels. Of all patients in this
series 183 (97.3%) successfully catheterized at last followup,
highlighting the durability of the 2 techniques.
CONCLUSIONS
In a large group of children and young adults we observed
durable and reliable results with the Casale and the traditional single Monti procedure. Continence rates at long-term
followup approach 98% with 97% of patients successfully
catheterizing. The choice of a Casale or Monti procedure as
well as the stomal location should be based on intraoperative
anatomy, surgeon preference and stomal location preference, considering the comorbidities in each individual. However, the umbilical stomal location appears to be associated
with higher rates of subsequent subfascial revision. Regardless of the channel type we firmly believe that making the
extravesical portion of the channel as short and straight as
possible, avoiding redundancy of the channel between the
bladder hiatus and fascia, provides optimal results with
these techniques.

MACE Malone antegrade continence enema
RLQ right lower quadrant
REFERENCES
1.
2.
Mitrofanoff P: Trans-appendicular continent cystostomy in the

management of the neurogenic bladder. Chir Pediatr 1980;
21: 297.
Yang WH: Yang needle tunneling technique in creating antireflux and continent mechanisms. J Urol 1993; 150: 830.

3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
Casale AJ: A long continent ileovesicostomy using a single

piece of bowel. J Urol 1999; 162: 1743.
Monti PR, Lara RC, Dutra MA and de Carvalho JR: New
techniques for construction of efferent conduits based on
the Mitrofanoff principle. Urology 1997; 49: 112.
Malone PS, Ransley PG and Kiely EM: Preliminary report: the
antegrade continence enema. Lancet 1990; 336: 1217.
Van Savage JG, Khoury AE, McLorie GA and Churchill BM:
Outcome analysis of Mitrofanoff principle applications using appendix and ureter to umbilical and lower quadrant
stomal sites. J Urol 1996; 156: 1794.
Mor Y, Kajbafzadeh AM, German K, Mouriquand PD, Duffy
PG and Ransley PG: The role of ureter in the creation of
Mitrofanoff channels in children. J Urol 1997; 157: 635.
Rink RC, McLaughlin KP, Adams MC and Keating MA: Modification of the Casale vesicostomy: continent diversion
without the use of bowel. J Urol, suppl., 1995; 153: 339A,
abstract 442.
Cain MP, Casale AJ, King SJ and Rink RC: Appendicovesicostomy and newer alternatives for the Mitrofanoff procedure: results in the last 100 patients at Riley Childrens
Hospital. J Urol 1999; 162: 1749.
Cain MP, Rink RC, Yerkes EB, Kaefer M and Casale AJ:
Long-term followup and outcome of continent catheterizable vesicostomy using the Rink modification. J Urol 2002;
168: 2583.
Monti PR, de Carvalho JR and Arap S: The Monti procedure:
applications and complications. Urology 2000; 55: 616.
Castellan MA, Gosalbez R, Labbie A, Ibrahim E and Disandro
M: Outcomes of continent catheterizable stomas for urinary
and fecal incontinence: comparison among different tissue
options. BJU Int 2005; 95: 1053.
Narayanaswamy B, Wilcox DT, Cuckow PM, Duffy PG and
Ransley PG: The Yang-Monti ileovesicostomy: a problematic channel? BJU Int 2001; 87: 861.
EDITORIAL COMMENT
These authors report the largest single center experience
with catheterizable channels used for lower urinary reconstruction according to the Mitrofanoff principle. They presented 201 patients treated with a Monti (109) or a Casale
(79) tube. The latter is a modification of the first procedure
with the advantage of providing a double length Monti channel without a circular anastomosis between them, which is a
theoretical disadvantage of the classical double Monti tube.
The aim of the study was to assess the durability of the
procedures (long-term results) and evaluate the complica-
1627
tions rate between these 2 approaches. They defined revision

procedures as open (stomal or subfascial) and endoscopic,
which enabled them to compare more objectively the severity of the complication.
The authors found that none of the complication rates
were significantly different between the 2 techniques except
a higher subfascial revision in stomas placed in the umbilical area with twice the complications of the Casale group
(21.6% vs 11.4%). The authors justify this finding by arguing
that this might have been caused by the personal preference
of 1 author to almost always reach the umbilicus for placing
the stoma.
Interestingly this finding opens a fine reflection about the
real necessity of the double channels in general (double
Montis or Casales) since the authors present a 5-fold higher
revision rate for the Casale channel in the umbilicus compared to the RLQ. This finding corresponds with our personal experience with 107 urinary catheterizable channels,
of which reservoirs for bladder substitution (mainly rhabdomyosarcoma cases) have stomas placed in the umbilicus. It
is a well-known technical step that the outlet channel should
be as short as possible, if possible leaving the valve mechanism and reaching the inverted skin tube with the anterior
reservoir wall sutured to the posterior fascial wall to keep
the channel angulation stable. We have previously observed
in an experimental model that angulation is even more
important than the embedded length of the tube or the
submucosal tunnel. Only a final lower quadrant stoma can
always enable this end result. Although the authors prefer
the RLQ, we personally advocate the midline as the preferred stomal site because of the absence of musculature and
ease of mobilizing skin flaps to the cavity to reach the distal
part of the tube.
The most important conclusion of this study is that the
Casale or probably the double Monti tubes are really not
necessary in the majority of cases because, if the stoma is to
be placed in the RLQ, a simple Monti channel is probably
enough.
Antonio Macedo
Department of Pediatric Urology
Federal University of Sao Paulo
So Paulo, Brazil
Endoscopic Injection of
Bulking Agents in Children With
Incontinence: Dextranomer/Hyaluronic
Acid Copolymer Versus Polytetrafluoroethylene
Lori Dyer, Israel Franco, Casimir F. Firlit, Edward F. Reda, Selwyn B. Levitt and Lane S. Palmer*
From the Schneider Childrens Hospital, North Shore-Long Island Jewish Health System (LD, SBL, LSP), New Hyde Park
and Westchester Medical Center (IF, EFR), Valhalla, New York, and Cardinal Glennon Childrens Hospital (CFF), St. Louis, Missouri
Purpose: Treating pediatric incontinence can be challenging. Many surgical procedures are available with variable success
and complications. Endoscopic injection of bulking agents into an incompetent bladder neck was first described using Teflon
and most currently using Deflux. We compared the results of bladder neck injection using Deflux and Teflon to manage
urinary incontinence in children.
Materials and Methods: A retrospective study was performed in children who underwent primary endoscopic injection of
Teflon or Deflux to the bladder neck. Patients with prior bladder neck surgery were excluded. Data were collected on
underlying anatomical pathology, preoperative bladder capacity, endoscopic approach, quantity of bulking agent and outcome. Dry was defined as 3 hours or greater awake without wetting. Wet was defined as the need for a pad or diaper.
Results: A total of 34 patients underwent primary injection of bulking agents, including 32 via a retrograde approach and
2 via a combined antegrade-retrograde approach. A total of 20 children with a mean age of 2.7 years were injected with Teflon
and 11 were injected with Deflux. One of the 20 Teflon injected patients was dry more than 6 months. One of the 14 Deflux
injected patients was dry at 3 months in the daytime, another 2 improved at 3 months (antegrade/retrograde in 1) but
worsened at 6 months and another was dry after a second injection. All 4 improved children had age appropriate bladder
capacity. The volume of injected agent was 1 to 10 cc. No complications were seen. Five children per group underwent
subsequent open continence surgery.
Conclusions: Bladder neck injection of bulking agents is a generally ineffective therapy for incontinence. While neither the
number of injections nor the bulking agent used affected the results, adequate bladder capacity and antegrade injection with
Deflux heralded short-term improvement. Patients with exstrophy consistently did poorly.
Key Words: bladder, urinary incontinence, endoscopy, polytetrafluoroethylene, dextranomer-hyaluronic acid copolymer
he treatment of urinary incontinence secondary to

sphincter and/or bladder neck incompetence in children can be challenging. The underlying causes of
incontinence include anatomical and neurogenic etiologies
that may be associated with a high or low pressure bladder.
The armamentarium of medical and surgical therapies is
aimed at increasing bladder outlet resistance. Many surgical
procedures, including bladder neck reconstruction, slings
and artificial urinary sphincters, carry variable success and
are associated with considerable morbidity and long recovery periods. Pharmacotherapy is less successful and it also
has associated untoward effects. The concept of a minimally
invasive approach to the vexing problem of incontinence in
this complex population is attractive for obvious reasons,
that is shorter convalescence and less discomfort for the
child.
The injection of bulking agents into an incompetent bladder neck was first described in children in the 1970s using
Teflon and it has been reevaluated every decade with
new agents. In chronological order they include Teflon
* Correspondence: 1999 Marcus Ave., M 18, Lake Success, New

York 11042 (telephone: 516-466-6953; FAX: 516-4665608).
0022-5347/07/1784-1628/0
(polytetrafluoroethylene)1 and Contigen (glutaraldehyde

cross-linked bovine collagen)2 in the late 1970s and 1980s, followed by silicone (polydimethylsiloxane)3 and Durasphere (carbon coated beads)4 in the 1990s, and currently Deflux (dextranomer/hyaluronic acid copolymer).5 Issues such as low
success rates and undesirable safety profiles have limited
each agent. These issues have led to the use of these agents
in a select population in which they may be used as primary
treatment or to complement another of the treatments
noted. Of all agents Teflon has generally been considered to
have the highest success rate. However, its use was abandoned after the implication of particle migration.6 8 The
most recently used injectable bulking agent, Deflux, does not
appear to carry this risk and its efficacy for treating incontinence in children is early in its evaluation. We compared
the results of bladder neck injection using Deflux and Teflon
to manage urinary incontinence.
A retrospective study was performed in children who underwent endoscopic injection of Teflon or Deflux to the bladder
neck. Patients were excluded if prior bladder neck surgery
had been performed, such as a sling, artificial urinary
1628
Vol. 178, 1628-1631, October 2007

Printed in U.S.A.
DOI:10.1016/j.juro.2007.05.092
ENDOSCOPIC INJECTION OF BULKING AGENTS FOR INCONTINENCE

sphincter or bladder neck reconstruction. Data were collected on the underlying anatomical pathology, eg bladder
exstrophy, myelomeningocele, etc, as well as other demographic and medical information, such as medications and
measured preoperative bladder capacity at voiding cystourethrogram or urodynamic evaluation. Additional data were
collected with respect to the specific endoscopic approach as
described, quantity of bulking agent used and outcome. The
outcome was determined by parental and patient interviews
at subsequent followup visits. Patients were considered dry
if there were at least 3 hours awake without wetting. Wet
was defined as the need for a pad or diaper. All patients were
wet preoperatively. Postoperative evaluations were guided
using these criteria.
Endoscopic Procedure
All bladder neck injections were performed with the patient
under general anesthesia. Latex precautions were taken
when appropriate. The same endoscopic procedure was used
irrespective of patient gender. In the retrograde approach a
needle inserted through the cystoscopic channel was placed
into the submucosa of the bladder neck. Care was taken to
identify a supple nonfibrotic area of the bladder neck for
injection. The bolus of the bulking agent was injected to
raise a mound. All bladder necks were approached circumferentially with the intent of luminal occlusion with injections made at the 12, 3, 6 and 9 oclock positions. In the
combined antegrade-retrograde approach access to the bladder was achieved percutaneously under the guidance of a
cystoscope passed into the bladder via the urethra. The
access tract was dilated until it was able to accommodate a
10Fr offset cystoscope. The needle was placed through the
working channel of the bladder cystoscope and submucosal
injections were made into nonfibrotic areas, again with the
goal of luminal occlusion. These injections were made with
an assistant observing through the urethral cystoscope to
ascertain that the needle did not traverse the full thickness
of the bladder neck and observe the effectiveness of luminal
occlusion. Urethral catheters were not left indwelling postoperatively. Patients who underwent combined antegraderetrograde procedures were maintained for about 1 week
with a suprapubic cystostomy tube left through the incision,
through which the access sheath was placed.
RESULTS
There were 34 patients with urinary incontinence who underwent primary injection of bulking agents. Patients were
excluded if there was prior surgery of the bladder neck, such
as a sling, artificial urinary sphincter or bladder neck reconstruction. A total of 20 children with a mean age of 2.7 years
(range 5 months to 22 years) were injected with Teflon
between 1988 and 1993, and 14 with a mean age of 14 years
(range 9 months to 20 years) were injected between 2002
and 2006 with Deflux. In each group the largest number of
patients had exstrophy/epispadias or myelodysplasia (see
table). All except 1 child in the Teflon group and 2 in the
Deflux group were on anticholinergic agents. An additional
child in the Deflux cohort was taking pseudoephedrine. A
total of 32 patients were injected via a retrograde approach
and 2 in the Deflux cohort were injected via a combined
retrograde and antegrade approach through a suprapubic
cystostomy.
1629
Causes of incontinence
Myelodysplasia
Bladder exstrophy
Isolated epispadias
Urethral duplication
Other
No. Teflon Group
No. Deflux Group
5
13
2
0
0
7
2
1
2
2
The 20 patients injected with Teflon required 1 to 5 injection procedures each. Only 1 of these children (5%), who
had epispadias, was considered dry after 6 months. This
patient had a somewhat decreased bladder capacity of 100 cc
(estimated 164 cc.) In all children with a primary diagnosis
of classic bladder exstrophy Teflon injection failed (0% improvement). Bladder capacity preoperatively and postoperatively in these patients was uniformly unchanged.
The 14 patients injected with 1 to 10 cc Deflux required 1
to 4 injection procedures each. The average volume per injection in the Deflux group was 4.3 cc. One of these children,
who had a short urogenital sinus, was dry at 6 months
(actual and estimated bladder capacity 300 cc). She was
injected antegrade with 10 vials of Deflux. Another 2 children, including 1 injected antegrade and 1 injected retrograde, showed improvement at 3 months but had symptomatic worsening at 6 months. One child with sacral agenesis
in whom bladder neck injection initially failed maintained
daytime continence for 6 months following a second injection
via a combined retrograde-antegrade approach. All 4 children showing improvement had age appropriate bladder
capacity. Three of these 4 children had myelomeningocele
and 1 had epispadias without exstrophy.
Average bladder capacity in children with failure vs some
improvement was 250 vs 350 cc. Children with exstrophy
(study criteria mandated that they did not undergo a prior
continence procedure) uniformly had no improvement in
continence or bladder capacity following Deflux injection.
Increasing the amount of Deflux injected correlated positively with symptomatic improvement. Children who were
considered improved or cured received between 4 and 10 cc
Deflux per injection.
Complications were uncommon in the 2 groups. There
were no complications in the Teflon treated group. In the
Deflux treated group 1 suprapubic tube was dislodged and
replaced in the retrograde-antegrade group. Five children
per treatment group underwent subsequent open continence
surgery. No significant increase in fibrotic tissue or difficult
surgical dissection was noted in any of the 10 patients who
progressed to definitive bladder neck reconstruction. Further long-term therapy is required for recent failures.
DISCUSSION
An ideal bulking agent for the bladder neck in children with
incontinence would be safe and provide reliable and durable
results. Teflon was used by Politano in the 1970s with good
results in 8 of the 11 patients in his report.1 The controversy
regarding particle migration6 8 truncated further use of this
agent. Cross-linked glutaraldehyde bovine collagen was associated with allergic reactions, thus, requiring pre-injection
skin testing for a hypersensitivity reaction. Moreover, its
use was curtailed primarily because of the routine need for
re-injection due to absorption (it is biodegradable) and com-
1630
mon extrusion of the implant. Guys3 and Halachmi9 et al

reported complete continence in 20% to 35% and some improvement in 50% of children injected with polydimethylsiloxane. However, the recent fear of intracorporeal silicone
limits the popularity of this agent. Deflux is currently approved by the Food and Drug Administration for vesicoureteral reflux. This agent is reported to be nonimmunogenic and nonmigratory. Its use for treating incontinence
has not been widely reported. These issues led to the use of
these agents for incontinence in a select population in which
they may be used as a primary treatment or complement 1 of
the other treatments noted.
In the mid 1970s Politano first described Teflon to treat
incontinence.1 In 1985 Vorstman and Politano reported on
11 patients, of whom 8 were injected retrograde and 5 were
injected perineally, with a followup of 5.3 years.10 Of that
cohort 45% were considered dry and an additional 27% were
considered improved at last followup. Our experience does
not mirror that series. That study was small and included
patients who previously underwent Young-Dees-Leadbetter
bladder neck reconstruction. We noted no complications
with the endoscopic approach, while Vorstman and Politano
described 1 patient who had an abscess after injection via a
perineal approach. Patients underwent between 1 and 7
injections with 5 to 21 cc Teflon. Vorstman and Politano
maintained that none of their 300 patients experienced clinical manifestations of particle migration or embolization.
However, they conceded that particle migration was documented in the lymph nodes in autopsy studies.
To date ours is the third series of patients in which Deflux
was used as a bladder neck bulking agent for urinary incontinence in children. In the series by Caione and Capozza 19
patients, of whom 11 previously underwent bladder neck
surgery, were treated via a retrograde approach with an
average of 2.8 cc Deflux.11 Patients required between 1 and
3 injections. A total of 13 patients had exstrophy/epispadias
and 3 had a congenitally neuropathic bladder. At 1 year of
followup 56.3% of patients showed improvement and 18.7%
were cured, defined as 2.5 hours of continence. The prospective series by Lottman et al included up to 7 years of followup in patients injected in retrograde fashion.5 Of the 33
initially treated children 24 were injected primarily. That
cohort was more evenly divided between patients with exstrophy and myelomeningocele (13 and 16, respectively). At
3 years 30% of patients were dry and 20% were improved of
the 20 with followup, while 58% of those who previously
underwent bladder neck reconstruction were dry. Of the 5
patients with 7 years of followup 40% were dry. Four of our
14 patients injected with Deflux demonstrated improved
continence. It is difficult to compare the series because our
study only included patients in whom Deflux was used primarily, while the previous series contained a mixed cohort of
patients in whom Deflux was used primarily or secondarily
following bladder neck reconstruction. Our followup is too
short to compare with the long-term data of Lottman et al
but it compares favorably with the short-term success of the
2 previously published series.
In this series the specific endoscopic approach appeared
to impact on the outcome. The appeal of an endoscopic approach is its minimally invasive nature and rapid convalescence, and the rarity of complications compared to those of
bladder neck reconstruction, sling procedures or artificial
urinary sphincters. In the most recent cases we used a
combined antegrade-retrograde approach, as first described

by Dean et al.12 Our success rate improved when we compared these Deflux cases to Deflux cases in which a purely
retrograde approach was used. The benefits of this approach
include better visualization and access to the incompetent
bladder neck as well as confirmation of accurate needle
placement and injection adequacy. To limit the morbidity of
this procedure several precautions were taken, including the
use of a prior drain site to avoid bowel injury, needle confirmation of bladder access and retrograde cystoscopic confirmation of access. We used the site of the peel away sheath to
place the suprapubic cystostomy tube. This tube allowed
continuous bladder drainage and the avoidance of bladder
neck instrumentation while the Deflux hardened.
An important caveat regarding the benefit of primary
injection therapy is its impact on potential future bladder
neck surgeries. In our series 10 children who received primary injection treatment for incontinence underwent subsequent definitive open anti-incontinence surgery. These operations were not made more difficult by Teflon in 5 children
or by Deflux in 5. Vorstman and Politano found that surgical
dissection was more difficult in previously injected patients
with post-injection failure,10 while the experience of BenChaim et al with bladder neck reconstruction after collagen
injection2 was similar to ours with Teflon and Deflux. Since
the injections are submucosal, dissection to isolate the full
thickness of the bladder neck for a sling or an artificial
urinary sphincter should not be more difficult. Certainly in
the exstrophy population scar from the prior surgery makes
this dissection more difficult but not any more difficult than
after a bulking agent was also used. With respect to bladder
neck reconstruction the bolus of injectable agent can be
removed by incising the roof of the bolus and the surgery
may then be completed.
There are several limitations to our series. 1) Our followup of patients was short, particularly in those in whom
Deflux was used. Longer term followup is needed to better
assess the efficacy of our surgical approach as well as the
durability of the agents that were used. Currently available
long-term studies demonstrate decreasing durability of continence with most agents, including Deflux. We anticipate
this also to be the case in our patients. 2) We do not have
consistent urodynamic data on our patients. Videourodynamics were not done to document open bladder necks. This
information would be helpful for deriving potential predictors of success.13 It would allow better patient selection and
perhaps maximize the efficacy of primary injection therapy.
3) Our series included only patients in whom bladder neck
injection was performed primarily. To our knowledge this
makes our series unique because the cohort was better defined. However, we still had small subpopulations of children with different causes of incontinence, which makes
comparisons and, therefore, conclusions difficult to draw.
CONCLUSIONS
Endoscopic injection of bulking agents to the bladder neck as
primary treatment for incontinence in children is often ineffective and expensive. This appears to be true irrespective
of the bulking agent used and it is particularly true in
children with exstrophy. Short-term results seem to be better in children with myelomeningocele or after the injection
of larger volumes of bulking agent. A combined antegrade-

retrograde endoscopic approach is a promising technique for
improving success in select patients.
7.
REFERENCES
8.
1.
Politano VA: Periurethral Teflon injection for urinary incontinence. Urol Clin North Am 1978; 5: 415.
2. Ben-Chaim J, Jeffs R, Peppas D and Gearhart J: Submucosal
bladder neck injections of glutaraldehyde cross-linked bovine collagen for the treatment of urinary incontinence in
patients with the exstrophy/epispadias complex. J Urol
1995; 154: 862.
3. Guys J, Simeoni-Alias A and Delarue A: Use of polymethylsiloxane for endoscopic treatment of neurogenic incontinence in children. J Urol 1999; 162: 2133.
4. Chrouser KL, Fick F, Goel A, Itano NB, Sweat SD and Lightner DJ:
Carbon coated zirconium beads in B-glucan gel and bovine glutaraldehyde cross-linked collagen injections for intrinsic sphincter deficiency: continence and satisfaction after extended followup. J Urol 2004; 174: 1152.
5. Lottmann HB, Margaryan M, Lortat-Jacob S, Bernuy M and
Lackgren G: Long-term effects of dextranomer endoscopic
injections for treatment of urinary incontinence: an update of a prospective study of 31 patients. J Urol 2006;
175: 1485.
6. Aaronson IA, Rames RA, Greene WB, Walsh LG, Hasal UA
and Garen PD: Treatment of reflux: migration of Teflon to
the lungs and brain. Eur Urol 1993; 23: 394.
9.
10.
11.
12.
13.
1631
Miyakita H and Puri P: Particles found in lung and brain

following subureteral injection of PTFE paste are not
Teflon particles. J Urol 1994; 152: 636.
Malizia AA Jr, Reiman HM, Myers RP, Sande JR, Barham SS,
Benson RC Jr et al: Migration and granulomatous reaction
after periurethral injection of polytef (Teflon). JAMA 1984;
251: 3277.
Halachmi S, Farhat W, Metcalfe P, Bagli DJ, McLorie GA and
Khoury AE: Efficacy of polydimethylsiloxane injection to
the bladder neck and leaking diverting stoma for urinary
continence. J Urol 2004; 171: 1287.
Vorstman B and Politano V: Polytetrafluoroethylene injection
for urinary incontinence in children. J Urol 1985; 133: 248.
Caione P and Capozza N: Endoscopic treatment of urinary
incontinence in pediatric patients: 2-year experience with
dextranomer/hyaluronic acid copolymer. J Urol 2002; 168:
1868.
Dean G, Kirsch AJ, Packer MG, Scherz HC and Zaontz MR:
Antegrade and retrograde endoscopic Deflux bladder
neck bulking in the pediatric incontinence. Presented at
meeting of European Society for Paediatric UrologyAmerican Academy of Pediatrics, Uppsala, Sweden, June
1518, 2005, abstract 106.
Chernoff A, Horowitz M, Combs A, Libretti D, Nitti V and
Glassberg KI: Periurethral collagen injection for the treatment of urinary incontinence in children. J Urol 1997; 157:
2303.
First Stage Approximation of the Exstrophic

Bladder in Patients With Cloacal ExstrophyShould
This be the Initial Surgical Approach in all Patients?
J. C. Thomas,* R. T. DeMarco, J. C. Pope, IV, M. C. Adams and J. W. Brock, III
From the Division of Pediatric Urology, Monroe Carrell Jr. Vanderbilt Childrens Hospital, Nashville, Tennessee
Purpose: Cloacal exstrophy is rare and it represents a reconstructive challenge. Options for managing the urinary tract
include primary closure or approximation of the bladder halves in the midline with later closure. We present our observations
and evolving thoughts concerning optimal treatment in these patients.
Materials and Methods: We retrospectively reviewed the records of patients with cloacal exstrophy seen in the last 5 years.
Initial management was examined, including complete primary closure vs a staged approach. We noted midline defects,
spinal cord abnormalities or other anatomical reasons that precluded primary closure.
Results: Seven patients, including 5 females and 2 males, were identified. An omphalocele noted in all 7 patients was closed
in 5 at initial operation. All underwent preservation of the hindgut in the fecal stream. Spinal cord tethering was noted in
7 of 7 cases. Complete primary bladder closure was performed in 3 of the 7 patients, while the size of the bladder plates or
a large abdominal wall defect precluded closure in the remainder. Continence was not achieved in the 3 cases closed
primarily. All patients achieving urinary continence underwent bladder neck closure and augmentation cystoplasty with a
continent catheterizable channel.
Conclusions: Patients with cloacal exstrophy have anatomical issues that can prevent complete primary bladder closure or
preclude the achievement of urinary continence. The high incidence of tethered cord places these patients at risk for upper
tract changes and bladder decompensation during followup. Despite successful primary closure in 3 of 7 patients all have a
tiny bladder and require secondary procedures to become continent. Extensive dissection during the first operation can
contribute to more difficult dissection with potential increased morbidity during subsequent surgeries. Therefore, the best
initial approach for the typical patient may be closure of the abdominal wall and approximation of the exstrophied bladder
halves in the midline. Secondary closure with continent diversion and reconstruction of the external genitalia can be
performed at ages 18 to 24 months.
Key Words: bladder, bladder exstrophy, cloaca, stomas, urinary incontinence
loacal exstrophy is a rare occurrence with an incidence of 1/200,000 to 1/400,000 live births.1 It represents one of the most challenging reconstructive endeavors faced by pediatric surgeons and urologists. Aside
from the genitourinary defects, there are other associated
anomalies of the gastrointestinal, musculoskeletal and neurological systems that require a multidisciplinary approach
when counseling anxious parents.
Since the first survivor was reported in 1960 by Rickham,2
advances in total parenteral nutrition and perinatal care have
increased the once dismal survival rate of these individuals
to more than 90%.3 The wide spectrum of anatomical abnormalities necessitates an individualized approach to reconstructing the urinary tract. Because most of these patients
have some type of spinal dysraphism combined with a severe
anatomical insult, reconstruction of the bladder and outlet
rarely achieves continence and volitional voiding. Most of
these patients require intermittent catheterization and, although successful functional bladder closure has been re-
* Correspondence: Division of Pediatric Urology, Monroe Carrell Jr.

Vanderbilt Childrens Hospital, 4102 Doctors Office Tower, 2200 Childrens Way, Nashville, Tennessee 37232-9820 (telephone: 615-3431895; FAX: 615-936-1061; e-mail: john.thomas@vanderbilt.edu).
0022-5347/07/1784-1632/0
ported, most need eventual augmentation cystoplasty with

or without construction of a continent abdominal wall stoma
and bladder neck closure or reconstruction to remain continent.
Depending on overall patient status and associated anatomical features initial bladder management consists of primary or delayed bladder closure. We have attempted these 2
approaches and in reflecting on our experience we favor
approximation of the exstrophied bladder halves in the midline for most patients. Secondary closure along with reconstruction of the external genitalia can be performed at
around ages 18 to 24 months. Such an approach may provide
the most efficient and least morbid means to achieve urinary
continence for these patients.
METHODS
We retrospectively reviewed the records of patients with
cloacal exstrophy seen in the last 5 years. Initial management consisted of a multidisciplinary team evaluation with
subsequent operative intervention when the patient was
deemed stable. Chromosome analysis was performed and it
revealed 46, XX in 4 and 46, XY in 2 patients. One of these
genetic males was gender converted at birth due to inade-
1632
Vol. 178, 1632-1636, October 2007

Printed in U.S.A.
DOI:10.1016/j.juro.2007.03.164
FIRST STAGE APPROXIMATION OF EXSTROPHIC BLADDER IN CLOACAL EXSTROPHY

quate phallic size. Another female patient was referred at
age 11 months and genetic information was not available.
Separation and tubularization of the cecal plate from the
exstrophied bladder halves and colostomy construction were
performed in conjunction with pediatric surgeons. Closure of
an associated omphalocele was performed when possible and
there was no concern for respiratory compromise. The bladder was closed primarily or approximated in the midline,
creating the appearance of classic bladder exstrophy. In
addition to the surgical approach, we noted midline defects
or other anatomical reasons that precluded primary closure.
Renal anomalies were documented by abdominal ultrasound, while spinal abnormalities were evaluated with spinal ultrasound at birth and 4 of 7 patients underwent subsequent lumbosacral magnetic resonance imaging.
RESULTS
Seven patients, including 5 females and 2 male, of whom 1
had been gender converted, with cloacal exstrophy were
identified. Initial radiological assessment revealed spinal
cord tethering in 7 of 7 patients (100%) with associated
closed myelomeningocele, anterior sacral meningocele, lipomyelocele and caudal regression syndrome in 4, respectively. Vertebral anomalies were noted in the 3 other patients (50%). Five patients had renal anomalies, consisting
of unilateral renal agenesis in 2, multicystic dysplastic kidney in 1, ureteropelvic junction obstruction in an ectopic
kidney in 1 and a supernumerary kidney in 1.
Complete primary bladder closure was performed in 2
female patients within the first 48 hours of life. One patient
had undergone primary closure elsewhere and was referred
to us several months later with subsequent medical issues.
An omphalocele was noted in all cases and closed in 5 of 7 at
initial operation. In 1 patient the omphalocele was allowed
to re-epithelialize as a ventral hernia and another patient
required a silicone silo, followed by a Gore-Tex patch in an
attempt to close the abdominal wall. The 4 remaining patients underwent abdominal wall closure with the creation
of bladder exstrophy at the initial operation by approximation of the 2 bladder halves. The reasons for not closing the
bladder primarily were small size or agenesis of 1 hemibladder, a large cecal fissure, inability to close the abdominal
wall defect, and overall patient medical and nutritional status. The hindgut was preserved in all patients operated on
at our institution (6 of 7) as a colostomy in 4, a mucous
fistula in 1 and a cecostomy in 1. The patient referred after
primary closure was done elsewhere had an ileostomy. She
experienced severe short gut syndrome and no preservation
of the hindgut.
Construction of a continent urinary diversion with a catheterizable channel and bladder neck closure was performed
at ages 19 and 24 months in 2 of the 4 patients, respectively,
who underwent initial approximation of the bladder plates.
These 2 patients required iliac osteotomies and external
fixation devices. At 9 and 19 months of followup, respectively, these 2 patients are dry and doing well. Of the remaining 2 patients 1 died of respiratory arrest following
repeated fungal and bacterial septicemia, while the other
was doing well and awaited continent diversion.
Three patients underwent primary bladder closure at
birth, including the female referred to us at age 11 months
for further management of the short gut and failure to
1633
thrive. She was subsequently lost to followup after being

sent elsewhere for liver and small bowel transplant evaluation. Mean followup in the remaining 2 patients was 46 and
57 months, respectively. Those 2 patients were incontinent
of urine and awaiting future reconstruction. Urodynamic
data were available on 1 patient and revealed a bladder
capacity of 56 cc with an incompetent bladder outlet. It is
anticipated that each patient will require augmentation cystoplasty, creation of a continent abdominal wall stoma and
bladder neck repair or closure.
DISCUSSION
Survival is no longer the main issue for patients with cloacal
exstrophy. Numerous advances in parenteral nutrition, neonatal intensive care and surgical management have shifted
our focus as caretakers to improving the overall quality of
life for these challenging patients.4 Because a wide spectrum
of physical anomalies is associated with cloacal exstrophy,
each patient deserves a multidisciplinary, individualized approach in terms of surgical reconstruction.
The evolution of the surgical approach to these patients is
well documented in the literature.311 Basically after the
condition of the child is stabilized operative management
consists of separating and tubularizing the cecal plate from
the exstrophied bladder plates, closing the omphalocele
when possible, creating a colostomy or ileostomy and closing
the bladder in primary fashion or reapproximating the bladder halves in the midline to aid in abdominal wall closure.
Preservation of the hindgut is critical since it will grow and
function to minimize fluid and electrolyte problems.9 If the
hindgut is not used for bowel reconstruction, it should be left
as a mucous fistula for use in later vaginal reconstruction or
bladder augmentation.10
In contrast to patients with classic bladder exstrophy,
those with cloacal exstrophy rarely if ever have an adequate
bladder following initial closure. Most also require intermittent catheterization via the urethra or abdominal wall
stoma. Husmann et al reported continence in a group of
patients with cloacal vs bladder exstrophy following staged
bladder reconstruction.11 Overall 22% of the patients (5 of
23) with cloacal exstrophy vs 82% of those with classic bladder exstrophy achieved continence. Of the patients with
cloacal exstrophy who achieved continence 4 of 5 did not
have a coexisting neurological abnormality and all underwent bladder neck reconstruction. Eventual continence was
achieved in 11 of 23 patients with 6 requiring bladder neck
closure and a Mitrofanoff channel to become dry.
These data underscore the importance of coexisting
neurological abnormalities found in patients with cloacal
exstrophy. They have been reported in 50% to 100% of
patients and they contribute a neurogenic component to
the anatomical insult seen in patients with cloacal exstrophy, thus, making continence rare following primary bladder closure and/or bladder neck reconstruction.12 Our series had a 100% incidence of spinal cord tethering, and 4
of 7 patients (57%) had caudal regression syndrome, anterior meningocele, lipomyelocele and skin covered myelomeningocele, respectively. Repair of myelomeningocele,
anterior meningocele and spinal cord detethering was
performed in 3 patients, respectively. One patient died
and 3 with a tethered cord are being followed closely by
our neurosurgery colleagues. A detailed description of the
1634
neuroanatomy of patients with cloacal exstrophy showed

that the nerve supply to the hemibladders originates on
the serosal surface of the hindgut and extends laterally,
making them vulnerable to the extensive surgical dissection required when separating the bowel from each hemibladder.11,13
The complex nature of cloacal exstrophy necessitates an
individualized approach when reconstructing the urinary
tract. Various algorithms have been presented in the literature. All approaches stress initial separation of the cecum
from the exstrophied hemibladders, closure of the omphalocele when possible and preservation of the hindgut, preferably as a colostomy.311 Alternatives to bladder reconstruction are primary closure with a later staged continence
procedure or reapproximation of the exstrophied halves and
delayed closure when the patient has grown or requires
further stabilization. Despite the approach taken most patients require multiple procedures to become continent and
the majority will depend on CIC to empty the bladder.8,9,11
Husmann et al reported persistent incontinence in 18 patients following staged bladder reconstruction.11 The cause
was a noncompliant bladder in 6 of the 18 cases and inadequate outlet resistance in 12. Management consisted of bladder neck closure with a simultaneous Mitrofanoff procedure,
artificial sphincter placement and periurethral collagen or
construction of a noncontinent urinary diversion. Lund and
Hendren reported urinary reconstruction in 40 patients with
cloacal exstrophy, representing the largest series in the literature.9 They noted that 78% of the patients were dry and
35 of 40 required augmentation cystoplasty, while all underwent a continence procedure, including bladder neck reconstruction, intestinal nipple valve or a Mitrofanoff procedure.
Of note, all except 3 patients required CIC to empty the
bladders. The need for subsequent procedures and requirement for intermittent catheterization was also noted by Mitchell et al.8 In their experience with 8 patients treated from birth,
excluding 2 who did not want continent diversions, the remaining 6 required secondary bladder reconstruction and a continence procedure performed with the majority needing CIC.
The results of 1-stage closure were recently reported in a small
series of patients with cloacal exstrophy using the complete
primary exstrophy repair technique following initial approximation of the bladder halves.14
These data suggest that it is possible to close the bladder
primarily in patients with cloacal exstrophy. However,
hopes for a functional bladder with volitional voiding are
poor. Closing the bladder primarily must be a carefully
weighed decision. Certain factors that might preclude primary closure are the size of the bladder plates and omphalocele, the degree of pubic diastasis and overall patient condition. Osteotomy was shown to be important for decreasing
the potential complications following primary bladder closure, especially with a wide pubic diastasis. In fact, staging
pelvic closure during 1 to 2 weeks was shown to be especially
useful for failed initial closure or extreme diastasis.15 Despite these measures most patients require multiple surgical
procedures to correct other anomalies and avoiding the potential complications of a tenuous bladder closure would be
of great benefit for these patients.
However, there is a clear benefit to closing the bladder in
patients with classic bladder exstrophy. These patients typically do not have a coexisting neurological abnormality and
the procedure allows normal bladder cycling.11 A significant
number of these patients become continent and void spontaneously following staged or complete primary repair. This
is in stark contrast to patients with cloacal exstrophy, who
almost assuredly require CIC to empty the bladder.
It is important to remember the significant advances in
urological reconstructive procedures that have given us the
ability to make these patients dry, although at the expense
of multiple surgical procedures. In addition, it would seem
intuitive that, although this is possible following the extensive dissection required to close the bladder primarily, it
would make the inevitable augmentation or catheterizable
channel construction more difficult and perhaps increase the
rate of postoperative complications. To our knowledge no
direct, randomized comparison of delayed vs primary bladder closure supports this issue and, therefore, this point
would be based only on individual surgeon experience with
the 2 techniques. An apparent benefit to early bladder closure in the properly selected patient could be to protect the
bladder mucosa from developing squamous metaplasia. This
is purely speculative since to our knowledge the implication
of this finding in terms of future malignancy when combined
with an intestinal augment is not currently known. Another
benefit may be parental psychological well-being because
the child would look closer to normal without the bladder
everted on the lower abdominal wall.
Fortunately cloacal exstrophy is rare. Our small experience has led us to raise certain issues in the care of these
difficult cases. Although reasons for delaying bladder closure are outlined, we are not against primary closure in the
carefully selected patient. Not surprisingly in our series all
such patients are incontinent and await further reconstruction. In 1 patient despite favorable anatomy, and following
extensive dissection and closure of the bladder we noted that
the bladder was not much bigger than the Malecot catheter
left as a drainage tube. In the other primarily closed bladder
urodynamics revealed a capacity of 56 cc 4 years after the
initial procedure. Of the patients in whom we delayed closure reconstruction at 19 and 24 months, respectively, went
well with minimal postoperative complications, although
our followup is short. Division of the bladder neck was relatively easy in these patients since they underwent no prior
urethral reconstruction. It has been our approach to recommend construction of a continent reservoir with closure of
the bladder neck for the typical patient with cloacal exstrophy.
CONCLUSIONS
Quality of life issues now predominate in children born with
cloacal exstrophy. Parents must be prepared to handle the
multitude of surgical procedures required to repair the gastrointestinal and urinary tract as well as the orthopedic
anomalies encountered. Despite our ability to close the bladder primarily and reconstruct the bladder neck a coexisting
neurological defect is present in almost 100% of patients,
making volitional voiding and future continence unlikely
with complete primary repair or a 1-stage approach. In our
experience and that of others most of these patients require
construction of a continent reservoir to become dry. Approximating the bladder halves in the midline at birth may be a
viable option to consider in all patients. Subsequent dissection would seem easier and the risk of postoperative complications may be decreased. However, we have not yet per-

formed reconstruction in our second group of patients who
underwent primary bladder closure at birth to determine
whether we should adopt this approach in all patients. We
believe that this approach may allow most of these patients
to achieve continence with the least number of surgeries and
lesser morbidity.

CIC clean intermittent catheterization
REFERENCES
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
15.
Hurwitz RS, Manzoni GA, Ransley PS and Stephens FD: Cloacal exstrophy: a report of 34 cases. J Urol 1987; 138: 1060.
Rickham PP: Vesico-intestinal fissure. Arch Dis Child 1960;
35: 97.
Mitchell ME and Plaire C: Management of cloacal exstrophy.
In: Pediatric Gender Assignment: A Critical Appraisal. Edited by S Zderic. Amsterdam: Kluwer Academic/Plenum
Publishers 2002; pp 267273.
Matthews R, Jeffs RD, Reiner WG, Docimo SG and Gearhart
JP: Cloacal exstrophy: improving the quality of lifethe
Johns Hopkins experience. J Urol 1998; 160: 2452.
Ricketts RR, Woodard JR, Zwiren GT, Andrews HG and
Broecker BH: Modern treatment of cloacal exstrophy. J Pediatr Surg 1991; 26: 444.
Stolar CJH, Randolph JG and Flanigan LP: Cloacal exstrophy:
individualized management through a staged surgical approach. J Pediatr Surg 1990; 25: 505.
Smith EA, Woodard JR, Broecker BH, Gosalbez R and
Rickettes RR: Current urologic management of cloacal
exstrophy: experience with 11 patients. J Pediatr Surg
1997; 32: 256.
Mitchell ME, Brito CG and Rink RC: Cloacal exstrophy reconstruction for urinary continence. J Urol 1990; 144: 554.
Lund DP and Hendren WH: Cloacal exstrophy: a 25-year experience with 50 cases. J Pediatr Surg 2001; 36: 68.
Diamond DA and Jeffs RD: Cloacal exstrophy: a 22-year experience. J Urol 1985; 133: 779.
Husmann DA, Vandersteen DR, McLorie GA and Churchill
BM: Urinary continence after staged bladder reconstruction for cloacal exstrophy: the effect of coexisting neurological abnormalities on urinary continence. J Urol 1999; 161:
1598.
McLaughlin KP, Rink RC, Kalsbeck JE, Keating MA, Adams
MC, King SJ et al: Cloacal exstrophy: the neurological
implications. J Urol 1995; 154: 782.
Schlegel PN and Gearhart JP: Neuroanatomy of the pelvis in
an infant with cloacal exstrophy: A detailed microdissection
with histology. J Urol 1989; 141: 583.
Lee RS, Grady R, Joyner B, Casale P and Mitchell M: Can a
complete primary repair approach be applied to cloacal
exstrophy? J Urol 2006; 176: 2643.
Mathews R, Gearhart JP, Bhatnagar R and Sponseller P:
Staged pelvic closure of extreme pubic diastasis in the
exstrophy-epispadias complex. J Urol 2006; 176: 2196.
EDITORIAL COMMENTS
These authors reinforce time honored precepts taught by our
late colleague Dr. Robert Jeffs. They describe these extreme
birth defects and whether one closes the bladder along with
the omphalocele depends on multiple factors. I disagree with
1635
the authors on the timing of continent diversion in an 18 to

24-month-old patient. The fact remains that most patients
require it. In our approved database, which contains more
than 65 patients with the OEIS (omphalocele-exstrophyimperforate anus-spinal defects) complex, only 1 is voiding
and dry. I would like more information on the exact type of
osteotomy and immobilization. In our series of 15 cloacal
reclosures many patients had a posterior iliac osteotomy
with inadequate immobilization. In that series we combined
a vertical iliac osteotomy lateral to the sacroiliac joint and a
transverse innominate osteotomy just above the hip. This
combination restores the pelvic girdle, decreases tension on
the midline, restores more normal pelvic angles and allows
better apposition of the pubis with or without an intrasymphyseal bar.1 In these patients with a large omphalocele and
extreme diastasis it is better to close the abdominal wall,
establish a colostomy and join the bladder halves, as suggested. Since most patients require a continent pouch, we
wait 10 to 12 months until the bones are stronger, do an
osteotomy, place a fixator and immobilize, gradually reduce
the diastasis during 2 to 3 weeks and then perform the soft
tissue closures (reference 15 in article). These data were
reinforced by Ben-Chaim et al, who found an 89% complication rate in cases closed without an osteotomy and a 17%
complication rate in those with an osteotomy.2 These authors describe these bladders as small and, if used for later
augmentation, the template must be successfully closed and
hopefully grow after closure. We agree with their time honored approach with a few caveats for this complex. The
newborn with OEIS is formidable and careful surgical judgment using all surgical adjuncts, as the authors point out, is
the true secret to success.
John P. Gearhart
Jeffs Pediatric Urology Service
The Johns Hopkins Hospital
Baltimore, Maryland
1.
2.
Schartz R, King J, Purves J, Sponseller P and Gearhart J:

Repeat pelvic osteotomy in cloacal exstrophy: applications
and outcomes. Unpublished data.
Ben-Chaim J, Peppas D, Sponseller P, Jeffs R and Gearhart J:
Applications of osteotomy in the cloacal exstrophy patient.
J Urol 1995; 154: 865.
This rare abnormality occurs in a spectrum and, therefore,

no surgeon or institution has enough cases to be able to
discuss the management authoritatively. When I reflect on
my personal experience with 12 cases, including 3 in which
I performed the initial surgery and 9 that were referred after
initial management elsewhere, I can state from the outset
that no single unified approach could have been applied.
In some cases the abdominal wall defect is relatively small,
allowing omphalocele repair, separation of the hindgut, colostomy, closure of the hemibladders and primary abdominal wall
repair. In severe pelvic diastasis gradual approximation during
a 2-week period, as proposed by Mathew et al (reference 15 in
article), appears to be more appropriate.
Similar to the experience of the authors, all of my 12
cases required bladder augmentation and continent diver-
1636
sion. A composite of stomach and small bowel provided a

satisfactory urinary reservoir with a continent channel and
stoma for intermittent catheterization. In 3 of the 9 referred
cases the cecum was incorporated in the initial bladder
closure, which is inadvisable, and an ileostomy was performed. The ileostomy with its watery stool was converted to
a continent Kock fecal reservoir, thus, eliminating the ileostomy bag. In other cases in which the cecum was preserved
pull-through of the colostomy through the perineum was
feasible.
Finally, the aim of treatment in children born with cloacal
exstrophy is to give them the best life possible wherever they
live in this world. Therefore, it is my belief that the surgical
strategy and management should be individualized and based
not only on anatomical, but also on social and cultural factors.
Moneer K. Hanna
New York Hospital-Weill Cornell Medical Center
New York, New York
REPLY BY AUTHORS
This article was meant to be conceptual and to raise questions regarding the optimal management of this difficult and
rare anomaly based on our experience. We agree that treatment should be individualized based on multiple factors.
However, the fact remains that almost all of these patients
will likely require construction of a continent reservoir. The
timing of this reconstruction was between 18 and 24 months
in our small series, and we do not disagree with reconstruction at an earlier age. We did use transverse innominate
pelvic osteotomies followed by stabilization with an external
fixation device. Although we have not applied the gradual
diastasis reduction method, it has clearly been shown to be
useful in patients with severe pelvic diastasis. Because continence is unlikely following primary bladder closure, initial
approximation of the bladder halves allows these patients to
grow, may reduce the number of future operations and
makes construction of a continent reservoir easier.
Basic Science Research Finalists

In Utero Exposure to Benzophenone-2 Causes
Hypospadias Through an Estrogen Receptor Dependent Mechanism
Michael H. Hsieh,* Erin C. Grantham, Benchun Liu, Reginald Macapagal, Emily Willingham
and Laurence S. Baskin
From the Department of Urology, University of California-San Francisco, San Francisco, California, and Department of Biology, Baylor
University (EW), Waco, Texas
Purpose: Additives such as benzophenone-2 are commonly used in cosmetic products and food container plastics to filter out
ultraviolet light. In pregnant women exposure may result in transplacental transfer of benzophenone-2 to fetuses. Benzophenone-2 is estrogenic in vitro and in the rat uterotropic assay. Estradiol causes hypospadias in mice and estrogen-like
compounds are also postulated to cause hypospadias. We determined whether hypospadias would develop in male mice
exposed to benzophenone-2 in utero and whether this outcome depended on estrogen receptor pathways.
Materials and Methods: Timed pregnant C57BL/6 mice were administered benzophenone-2 (6.25 mg) or control vehicle by
oral gavage from gestational days 12 through 17 and they were sacrificed on day 18. Fetuses were weighed and sexed,
anogenital distance was measured and genital tubercles were harvested for paraffin sections or quantitative reverse
transcriptase-polymerase chain reaction analysis of genes purportedly involved in genital tubercle development.
Results: Eight of 57 benzophenone-2 treated male fetuses (14%) whose genital tubercles were examined histologically had
hypospadias (p 0.0064). Co-administration of benzophenone-2 with the estrogen receptor antagonist EM-800 resulted in
normal genital tubercles, ie no hypospadias, in 26 of 26 mice. Likewise no EM-800 or control treated male genital tubercles
showed hypospadias. Benzophenone-2 treated male mice had no changes in body mass adjusted anogenital distance relative
to controls. Reverse transcriptase-polymerase chain reaction revealed that genital tubercles of benzophenone-2 treated male
mice expressed higher levels of estrogen receptor- relative to male controls (p 0.04).
Conclusions: These findings suggest that benzophenone-2 may cause hypospadias via signaling through the estrogen
receptor. Further study of human benzophenone-2 exposure and its effects is needed to support this hypothesis.
Key Words: urethra; hypospadias; endocrine disruptors; benzophenone; receptors, estrogen
enzophenone-2 is added to food container plastics

and over 200 cosmetic products, including lipsticks
and perfumes, to filter out ultraviolet light.1 In 1
report transdermal absorption of BP2 by pregnant women
was postulated to result in transplacental transfer of this
compound to fetuses.2 This possibility is alarming, given
that BP2 is estrogenic in the rat uterotropic assay and in
multiple in vitro models, including ER binding assays.3,4
However, to our knowledge there are no published data on
transplacental transfer of BP2 in animals or humans.
Approximately 1/250 males is born with hypospadias,
an abnormality in which the urethra opens onto the ventral aspect of the penis rather than at the tip.5 Fetal
exposure to estrogen-like compounds has been postulated
as a cause of hypospadias.5 According to this hypothesis
prenatal exposure to estrogen-like compounds may prevent normal fusion of the urethral folds and, hence, result
in hypospadias. We determined whether mice exposed to
BP2 in utero would show hypospadias. We also measured
secondary indicators of feminization, namely body weight,
AGD and GT expression patterns of transcription factors,
* Correspondence: 400 Parnassus Ave., UFP, 6th Floor Crede Ambulatory Care Center, Box 0738, San Francisco, California 94143
(telephone: 415-415-1617; FAX: 415-476-8849; e-mail: mhsieh@
urology.ucsf.edu).
Supported by National Institutes of Health Grant R01 DK058105-03.
0022-5347/07/1784-1637/0
and steroid and growth receptors. Finally, we ascertained

whether any BP2 induced hypospadias depended on ER
signaling.

Animals and Treatment
Timed pregnant C57BL/6NCr mice (Animal Production
Program, National Cancer Institute at Frederick, Frederick, Maryland) were received on GD 5. From GDs 12
through 17 they were administered by subcutaneous injection 1) 10 g EM-800, an antiestrogen that blocks ER-
and ,6 dissolved in 100 l 10% ethanol/90% corn oil or 2)
100 l ethanol/corn oil as a control vehicle. Mice were also
administered by oral gavage 1) 6.25 mg BP2, an intermediate rodent dose based on previous studies,15 in 100 l
10% ethanol/90% corn oil or 2) 100 l ethanol/corn oil as a
control vehicle.7 A minimum of 5 dams were used for each
treatment group. All animals were housed 1 per 20 25
47 cm cage with laboratory grade pine shavings heat
treated to remove resins as bedding. They were acclimated to 68F to 74F and 40% to 50% relative humidity on
a reversed light schedule of 14 hours of light and 10 hours
of darkness, and they had free access to mouse chow and
tap water. At GD 18, when urethral closure at the tip of
the GT has normally occurred,7 the dams were sacrificed
1637
Vol. 178, 1637-1642, October 2007

Printed in U.S.A.
DOI:10.1016/j.juro.2007.03.190
1638
IN UTERO EXPOSURE TO BENZOPHENONE-2 AND HYPOSPADIAS
and the fetuses were harvested. After AGD measurement,

gonadal identification of sex and assessment of body
weight, GT were harvested by microdissection for histological analysis or inclusion in RT-PCR analysis.
Measurement of Body Weight
Intact fetuses were dried after removal from the placenta
and weighed.
Measurement of AGD
Using a dissecting microscope fetal pelvises were digitally
photographed under 2.5 magnification. A ruler placed next
to each GT was measured in pixels using Adobe Photoshop and used as a reference to measure the AGD of each
fetus. One observer (EC) performed all AGD measurements
while blinded to treatment group. Another observer (MH),
who had previous experience with measuring AGD, corroborated the technique of measurement. AGDI was calculated
for each fetus by dividing AGD by the cube root of body mass
since AGD is related to body mass.8
Quantitative PCR Analysis
Total RNA from individual GTs was extracted using a
QIAgen RNeasy Micro Kit. Optical density at 260/280
nm was measured for all samples. Those with a ratio of
greater than 1.6 were considered high purity RNA and
reverse transcribed to cDNA. Quantitative real-time PCR
analysis (TaqMan) was performed elsewhere on high
purity samples. Primers for ATF3 and estrogen and ,
androgen, progesterone, epidermal growth factor and insulin-like growth factor-1 receptors were obtained from
Applied Biosystems. Primer identification numbers were
Mm00476032_m1, Mm00433149_m1, Mm00599819_m1,
Mm00442688_m1, Mm00435625_m1, Mm00433023_m1
and Mm00802831_m1.
Assessment of Hypospadias
All GTs not used for RNA expression analyses were fixed in
formalin, embedded in paraffin, cut with a microtome (4 m
sections) and stained with hematoxylin and eosin. The criterion for hypospadias was the coexistence of an open urethra and a visible corpus cavernosum, which is a structure
that occurs in the GT shaft.
Reconstruction in 3D
Anatomical computer reconstruction of serial sections was created using a digital camera (Nikon, Melville, New York) and
WinSurf 4.3 software. Every fifth to tenth section was digitized. The corporeal bodies, glans and urethra were manually
outlined and assessed against the original histological sections.
We performed 3D analysis of the urethral lumen in comparison
to the spongiosum, erectile tissue and other surrounding structures in the x and y axes as an animated motion picture with
views of interest captured as static images.7
Weight AGD and AGDI for fetuses from each treatment
group were averaged and compared using ANOVA. If
p 0.05, we then proceeded with pairwise analysis using the
2-tailed t test. These analyses were performed in 2 ways,
including 1) by combining all fetuses of the same gender in
each treatment group regardless of litter of origin (combined

analysis) and 2) by combining fetuses of the same gender in
each litter before calculating average values for each treatment group (nested analysis). This was done to detect any
litter specific confounding effects.
RT-PCR data were analyzed using ANOVA, followed by
the 2-tailed t test if p 0.05. The Grubbs outlier test was
applied to all RT-PCR data and values were excluded as
outliers if p 0.05. Fishers exact 1-tailed analysis of 2 2
contingency tables was performed on hypospadias frequency
data. One-tailed tests were used because hypospadias is a
unidirectional end point.
RESULTS
Body Weight
By combined analysis using ANOVA the body weight of
fetuses did not differ significantly among the treatment
groups (p 0.085, fig. 1). Nested analysis of body weight
using ANOVA also did not demonstrate any differences
among the treatment groups (p 0.94).
AGD
The AGD of male fetuses born to dams exposed to BP2 were
greater than the AGD of male mice born to dams given
control vehicle (1.25 vs 1.09 mm, fig. 1, p 0.037). The AGD
of female fetuses did not differ significantly whether their
mothers received BP2 or control vehicle (p 0.16).
There were no statistically significant differences in
AGDI between males born to BP2 vs control treated dams
(p 0.069). The AGDI for female progeny of BP2 treated
dams were also not significantly different than the AGDI of
female offspring of control treated dams (p 0.059).
Detection of Hypospadias
We found that 8 of 57 male offspring (14%) (6 of 20 litters)
born to dams given 6.25 mg doses of BP2 showed hypospadias (p 0.0064). In contrast, all 47 male progeny of control
treated dams showed normal GTs (see table). The 13 male
mice whose mothers were administered EM-800 alone did
not show hypospadias or any other gross abnormality of the
GTs. All 26 male mice whose mothers were exposed to BP2
plus EM-800 also had normal GTs.
Figure 2 shows the typical morphology of GTs from male
offspring of dams exposed to BP2, the antiestrogen EM-800
and/or control vehicles. The 3D reconstruction of select individuals from each group confirmed histological assessments (fig. 3).
Gene Expression in GTs
Real-time quantitative RT-PCR is a highly sensitive method
of measuring gene expression differences in tissues. Unfortunately the sensitivity of this technique introduces significant variability to results.9 Hence, before performing RTPCR we decided to exclude all outlier data using the Grubbs
test and a threshold of p 0.05. This resulted in the exclusion of 2 of 124 values, including 1 high value for ATF3
expression in the GT of a female pup from a BP2 treated
dam and 1 high value for ER- expression in another female
pup from a BP2 treated dam.
With outlier data excluded ANOVA analysis of RT-PCR
data demonstrated significant differences among the treat-
1639
FIG. 1. Body weight of BP2 treated mice (combined vs nested analysis ANOVA p 0.085 vs 0.94). Anogenital distance of BP2 treated mice
(combined vs nested analysis ANOVA p 0.0000057 vs 0.00027), BP2 treated vs control male mice (combined vs nested analysis 2-tailed
t test p 0.037 vs 0.057) and BP2 treated vs control female mice (combined vs nested analysis p 0.16 vs 0.104). Anogenital distance index
(combined vs nested analysis ANOVA p 0.0000086 vs 0.00023), BP2 treated vs control male mice (combined vs nested analysis 2-tailed
t test p 0.069 vs 0.060) and BP2 treated vs control female mice (combined vs nested analysis 2-tailed t test p 0.059 vs 0.053).
ment groups for ER- (p 0.022, fig. 4). Including outlier

data yielded p 0.091. The 2-tailed t test revealed that male
progeny from BP2 treated dams expressed higher levels of
ER- relative to male offspring of control dams (p 0.04).
However, RT-PCR revealed no statistically significant
differences in GTs from the male progeny of BP2 vs control
treated dams in expression levels of multiple steroid and
growth factor receptors, and transcription factors, including
androgen receptor, progesterone receptor, ER- and activated transcription factor-3 (data not shown).
DISCUSSION
Estrogenic compounds are thought to contribute to the increasing prevalence of hypospadias in humans. This is particularly relevant for compounds such as BP2, which feature
estrogenic activity in the rat uterotropic assay and in multiple in vitro models.3,4 These prior studies showed that BP2
binds to mouse as well as human ER and it promotes uterine
growth in ovariectomized rats and the proliferation of estrogen dependent human breast cancer cell lines. Other sub-
1640
Hypospadias frequency in mice treated with BP2 and EM-800
Controls only
BP2 only
EM-800 only
EM-800 BP2
No.
Dams
No. Hypospadiac
Pups
Total No. Pups

(% hypospadiac)
16
20
5
7
0
8
0
0
47
(0)
57 (14.0)
13
(0)
26
(0)
BP2 only vs controls only p 0.0064, and vs EM-800 and BP2 p 0.05.
stances with estrogenic activity, such as diethylstilbestrol,

cause hypospadias in rats.11 Furthermore, 17--ethinyl estradiol induces hypospadias in mice.7
Prenatal exposure to the ultraviolet stabilizer BP2
caused hypospadias in 14% of male mice. Although to our
knowledge the level of human exposure to BP2 is unknown,
related compounds such as benzophenone-3 are readily absorbed through human skin.10 We administered BP2 by oral
gavage because human exposure is likely in part from ingestion of lipsticks and food product packaging containing
BP-2. Transdermal absorption, which may be an important
route of exposure, was not examined because mice often
lick off dermally applied compounds. BP2 titration to 100fold lower doses (62.5 g per dose) resulted in a low but
detectable rate of hypospadias (data not shown). The preliminary results of exposure models using Environmental
Protection Agency software suggest that these levels are
within 3-fold of human exposure levels. Regardless, clinical transdermal absorption studies should be performed
in the future.
None of the 47 control treated male mice had hypospadias
and co-administration of the ER antagonist EM-800 with
BP2 resulted in no hypospadias in 26 exposed male mice.
Although the sample size may not appear large, the hypospadias rate in BP2 exposed mice was statistically significant and the rate of spontaneous hypospadias in mice is
essentially zero.6 These outcomes suggest that BP2, like
estradiol, may act through the ER to induce hypospadias.
Unfortunately reliable identification of hypospadias in a
given mouse requires use of the entire GT for histology,
which precludes concomitant RT-PCR analysis of ER expression. Hence, we were unable to identify increased ER-
expression in mice with hypospadias. Rather, we noted overall increased ER- expression in the male progeny of dams
exposed to BP2. Although it is possible that BP2 indirectly
interacts with ER, for example by up-regulating estrogen
production, this compound binds to the and forms of ER
with high affinity.7 Not all male mice in this study showed
hypospadias. This result is likely because hypospadias may
have multifactorial causes and some individuals are predisposed to environmental influence.5 In addition, although
BP2 has estrogenic activity, it is weak in comparison to
potent, naturally occurring ER ligands such as estradiol.12
Interestingly BP2 exposure was not associated with significant changes in fetal body mass or AGDI. AGD is sexually dimorphic in mice7 with males having a longer AGD
than females. Hence, we analyzed whether AGD of male
progeny of BP2 exposed dams was similar to that of female
progeny. We chose not to stratify the analysis of body mass
or AGDI by the presence or absence of hypospadias because
of the small number of hypospadiac mice (8 of 57). An association between hypospadias and male AGD shortening in
animal models was described for putative endocrine disruptors such as phthalates.13 One group identified a relationship between in utero phthalate exposure and the AGD
distance in human male infants.14 However, to our knowledge there are no data on the relationship of AGD to hypospadias in humans. In utero exposure of male mammals to
androgens is thought to result in greater birth weight.15
When sheep are exposed to prenatal, exogenous androgens,
female offspring are larger and other morphometric indexes
are increased in each sex.16 Overall these studies suggest
that in mammals larger male offspring are a result of differences in androgen concentrations between male and female embryonic milieus. Conversely inhibiting male hormone activity by direct antagonism or by increasing
exposure to female hormones may result in smaller male
offspring. Indeed, this was the case in rodent exposures to
high doses of estrogens.17 Several groups have reported an
association between low birth weight and hypospadias in
boys.18 These reports did not identify the causes of this
possible association and they may have been subject to selection bias (patients in neonatal intensive care units)18 or
founder effect (single country studies with relatively homogeneous populations).19 Hence, the nature of the relationship between body weight and hypospadias remains to be
established.
RT-PCR revealed that BP2 treatment induced up-regulation of ER- in male mice. In fact, ER- levels in BP2 treated
males were similar to those in BP2 treated females. This
finding is intriguing in light of our hypothesis that BP2
induces hypospadias through ER signaling. Although we
believe that BP2 binds ER, the expression profile of GTs
from male mice exposed to this compound is dissimilar to the
profile associated with ethinyl estradiol exposure (unpublished data). We speculate that the degree of binding between BP2 and ER may account for its differing effects on
growth receptor and hormone receptor expression compared
to those of ethinyl estradiol. Through RT-PCR we also examined the expression of androgen receptor, progesterone
receptor, ER- and activated transcription factor-3 because
they are implicated in other models of hypospadias (unpublished data). The expression of these molecules was unchanged.
FIG. 2. Morphology of hypospadias in BP2 treated mice. GT of

control treated male shows section from male mouse whose mother
received control vehicle by oral gavage. GT of BP2 treated male
shows section from male mouse whose mother received BP2 by oral
gavage. H & E, reduced from 40.
1641
FIG. 3. GT 3D reconstruction in control (A) and BP2 treated (B) male fetuses confirmed to be hypospadiac based on histology. Green areas
indicate glans. Gray areas indicate corporeal bodies. Yellow areas indicate open urethra. Blue areas indicate closed urethra.
To our knowledge our findings represent the first report

that a nonestrogen endocrine disruptor can induce hypospadias in ER dependent fashion. A previous study showed that
BP2 induces the proliferation of MCF-7 cells, an estrogen
sensitive breast cancer cell line.12 This activity is abrogated
by the ER antagonist ICI 182, 780. In our study we found
that EM-800, another ER antagonist, abolished the ability of
BP2 to induce hypospadias.
Unlike endocrine disruptors such as diethylstilbestrol,

avoidance of exposure to BP2 during pregnancy may be
difficult since this compound is an ingredient in food packaging plastics and more than 200 cosmetic products.1 The
hydroxyl groups of BP2 may confer on it its estrogenic
activity.4 Our data and previous studies suggest that it
may be prudent to avoid the design and use of future
ultraviolet stabilizers featuring multiple hydroxyl groups.
FIG. 4. Box and whiskers graph of ER- expression in BP2 treated GTs. Y axis indicates percent ER- expression relative to housekeeping
gene GAPDH. Upper and lower whiskers indicate range. Upper and lower extents of box indicate IQR. Cont, control. ANOVA p 0.022,
control vs BP2 treated males p 0.04 and BP2 treated males vs BP2 treated females p 0.66.
1642
CONCLUSIONS
BP2 induced hypospadias in a significant proportion of male
mice exposed in utero. Interestingly secondary features of
feminization, such as decreased body weight and decreased
AGD, were not present in BP2 treated male mice. ER- was
up-regulated in BP2 treated male mice, suggesting a role for
ER in BP2 induced GT signaling. Future studies are required to determine human exposure levels to BP2 and the
relationship, if any, between human hypospadias, BP2
and ER.
ACKNOWLEDGMENTS
EM-800 was provided by F. Labrie, Laval University, Quebec, Canada. Quantitative real-time PCR was performed at
the University of California San Francisco Cancer Center
Comprehensive Genome Analysis Core.
5.
6.
7.
8.
9.
10.

3D
AGD
AGDI
BP2
ER
GD
GT
PCR
RT-PCR
3-dimensional
anogenital distance
AGD index
benzophenone-2
estrogen receptor
gestational day
genital tubercle
polymerase chain reaction
reverse transcriptase-PCR
11.
12.
13.
14.
REFERENCES
1.
2.
3.
4.
Gray S, Houlihan J, Subramanian H, Kropp T, Markey K and

Gouldin C: Benzophenone-2. Environmental Working
Group 2006.
Communication from the Commission to the Council and European Parliament on the implementation of the Community Strategy for Endocrine Disruptersa range of substances suspected of interfering with the hormone systems
of humans and wildlife. Commission of the European Communities 2001; p 706.
Yamasaki K, Takeyoshi M, Sawaki M, Imatanaka N, Shinoda
K and Takatsuki M: Immature rat uterotropic assay of 18
chemicals and Hershberger assay of 30 chemicals. Toxicology 2003; 183: 93.
Morohoshi K, Yamamoto H, Kamata R, Shiraishi F, Koda T
and Morita M: Estrogenic activity of 37 components of
commercial sunscreen lotions evaluated by in vitro assays.
Toxicol In Vitro 2005; 19: 457.
15.
16.
17.
18.
19.
Baskin LS, Himes K and Colborn T: Hypospadias and endocrine disruption: is there a connection? Environ Health
Perspect 2001; 109: 1175.
Gauthier S, Caron B, Cloutier J, Dory YL, Favre A,
Larouche D et al: (S)-()-4-[7-(2,2-dimethyl-1-oxopropoxy)4-methyl-2-[4-[2-(1-piperidinyl)- ethoxy]phenyl]-2H-1-benzopyran-3-yl]-phenyl 2,2-dimethylpropanoate (EM-800): a
highly potent, specific, and orally active nonsteroidal antiestrogen. J Med Chem 1997; 40: 2117.
Kim KS, Torres CR Jr, Yucel S, Raimondo K, Cunha GR and
Baskin LS: Induction of hypospadias in a murine model by
maternal exposure to synthetic estrogens. Environ Res
2004; 94: 267.
Vandenbergh JG and Huggett CL: The anogenital distance
index, a predictor of the intrauterine position effects on
reproduction in female house mice. Lab Anim Sci 1995; 45:
567.
Bustin SA and Nolan T: Pitfalls of quantitative real-time reverse-transcription polymerase chain reaction. J Biomol
Tech 2004; 15: 155.
Jiang R, Roberts MS, Collins DM and Benson HA: Absorption
of sunscreens across human skin: an evaluation of commercial products for children and adults. Br J Clin Pharmacol
1999; 48: 635.
Henry EC, Miller RK and Baggs RB: Direct fetal injections of
diethylstilbestrol and 17 beta-estradiol: a method for investigating their teratogenicity. Teratology 1984; 29: 297.
Matsumoto H, Adachi S and Suzuki Y: Estrogenic activity of
ultraviolet absorbers and the related compounds. Yakugaku Zasshi 2005; 125: 643.
Mylchreest E, Cattley RC and Foster PM: Male reproductive
tract malformations in rats following gestational and lactational exposure to Di(n-butyl) phthalate: an antiandrogenic mechanism? Toxicol Sci 1998; 43: 47.
Swan SH, Main KM, Liu F, Stewart SL, Kruse RL, Calafat AM
et al: Decrease in anogenital distance among male infants
with prenatal phthalate exposure. Environ Health Perspect
2005; 113: 1056.
de Zegher F, Francois I, Boehmer AL, Saggese G, Muller J,
Hiort O et al: Androgens and fetal growth. Horm Res 1998;
50: 243.
Gill JW and Hosking BJ: Acute prenatal androgen treatment
increases birth weights and growth rates in lambs. J Anim
Sci 1995; 73: 2600.
Yasuda Y, Kihara T and Nishimura H: Effect of ethinyl estradiol on development of mouse fetuses. Teratology 1981; 23:
233.
Gatti JM, Kirsch AJ, Troyer WA, Perez-Brayfield MR, Smith
EA and Scherz HC: Increased incidence of hypospadias in
small-for-gestational age infants in a neonatal intensivecare unit. BJU Int 2001; 87: 548.
Fredell L, Kockum I, Hansson E, Holmner S, Lundquist L,
Lackgren G et al: Heredity of hypospadias and the significance of low birth weight. J Urol 2002; 167: 1423.
Urothelial Inhibition of Transforming Growth

Factor- in a Bladder Tissue Recombination Model
Siam Oottamasathien,* Karin Williams, Omar E. Franco, Marcia L. Wills, John C. Thomas,
Ali-Reza Sharif-Afshar, Romano T. DeMarco, John W. Brock, III, Neil A. Bhowmick,
Simon W. Hayward and John C. Pope, IV
From the Division of Pediatric Urology, Department of Urologic Surgery, Vanderbilt Childrens Hospital and Departments of Pathology
and Cancer Biology, Vanderbilt-Ingram Cancer Center, Vanderbilt University Medical Center, Nashville, Tennessee
Purpose: We examined the role of transforming growth factor- in urothelial and bladder development. Transforming
growth factor- signaling was attenuated in the urothelial compartment and the subsequent effects were examined in a tissue
recombination model.
Materials and Methods: Urothelium was cultured from adult rat bladders and transfected with control vector C7 or
mutant DNIIR (dominant negative transforming growth factor- receptor II). Grafts were created by recombining transfected
urothelium plus embryonic day 18 bladder mesenchyma and placed beneath the renal capsule of athymic mouse hosts. Grafts
were harvested at 21 and 42 days. Final tissues were evaluated with staining and immunohistochemistry using hematoxylin
and eosin, Gomoris trichrome strain, broad-spectrum uroplakin, smooth muscle actin-, phosphorylated SMAD2 and Ki67
antigen. Bladder structures were defined as having smooth muscle, suburothelial connective tissue and mature urothelium
expressing uroplakin. Urothelial compartment diameters were measured and subcategorized as small 0.10 to 0.40,
medium0.41 to 1.0 and large greater than 1.1 mm.
Results: At 21 days 14 C7 control and 15 DNIIR grafts were evaluated. No bladder tissue was seen in the C7 grafts vs 49
in DNIIR tissue, including 30 small, 9 medium and 10 large tissues. At 42 days 14 C7 and 12 DNIIR grafts were evaluated.
Six bladder structures (5 small and 1 medium) were seen in the C7 cohort vs 27 (14 small, 7 medium and 6 large) in the
DNIIR group. Immunohistochemical detection of phosphorylated-SMAD2 was significantly attenuated in DNIIR tissue. In
addition, Ki67 proliferative indexes were 4.0-fold higher in the DNIIR cohort compared to those in C7 tissues.
Conclusions: We successfully observed that primary urothelium cultures can be genetically manipulated and recombined
with undifferentiated mesenchyma to grow bladder tissue. By attenuating transforming growth factor- signaling in the
urothelium superior bladder tissue growth occurred, suggesting that transforming growth factor- is a growth inhibitor in
this organ system.
Key Words: urothelium, bladder, transforming growth factors, transfection, mesoderm
any disease states can lead to bladder dysfunction. Normal homeostatic cell-cell interactions
between the urothelium and stroma are critical
for normal bladder development and function.15 Although stromal-epithelial interactions are integral to normal bladder development, the autocrine, paracrine and
endocrine factors specifically involved remain poorly understood. TGF- isoforms are pleiotropic growth factors that
function physiologically to regulate the development, differentiation, cellular proliferation and immune responses in
organ systems. TGF- acts as a potent growth inhibitor of
most epithelial cell types and more recently it was recognized to be also growth inhibitory to stromal fibroblasts of
many tissues in vivo, including the bladder.6,7 To better
Study received approval from the Vanderbilt University Institutional Animal Care and Use Committee.
Supported by National Institutes of Health Grant R01-DK068593
(JCP IV).
* Correspondence: Department of Urologic Surgery, A1302 MCN,
Vanderbilt University Medical Center, Nashville, Tennessee 372322765 (telephone: 615-322-5823; FAX: 615-322-8990; e-mail: siam.
oottamasathien@vanderbilt.edu).
0022-5347/07/1784-1643/0
elucidate the role of TGF- in bladder development and

disease we impaired TGF- signaling in vitro in cultured
bladder urothelial cells and examined those effects in an in
vivo tissue recombination system.
Tissue recombination models are an efficient way in
which to evaluate stromal-epithelial interactions and bladder development in an in vivo environment. Previous experiments with tissue recombinants composed of embryonic rat
bladder mesenchyma with freshly isolated embryonic rat
bladder urothelium produced bladder tissue with mature
urothelium and stromal differentiation.2 Many investigators
have reported the successful generation of primary cultures
of rodent bladder urothelium.5,8 11 Our group recently reported the first successful use of primary cultured bladder
urothelium to recapitulate bladder tissue formation in a
tissue recombination model.5 Based on the premise that
mesenchyma and epithelium are crucial for appropriate organ differentiation the tissue recombination technique can
be modified to accommodate in vitro genetic manipulation of
the components before grafting. Thompson12 and Yang13
et al first pioneered these techniques to examine malignant
and benign disease progression in prostate models. With the
1643
Vol. 178, 1643-1649, October 2007

Printed in U.S.A.
DOI:10.1016/j.juro.2007.03.163
1644
TRANSFORMING GROWTH FACTOR- UROTHELIAL INHIBITION
ability to generate primary cultures of rodent bladder

urothelium coupled with the capability of producing bladder
tissue in a tissue recombination model we genetically manipulated bladder urothelium by over expressing mutant
DNIIR. We examined those effects in vivo after recombination with embryonic bladder mesenchyma. These studies
pave the way for an improved understanding of bladder
development and disease.
Cell Culture/Generation of Pure
Adult Rat Bladder Urothelial Cell Populations
All animal experiments were performed in accordance with
the Institutional Animal Care and Use Committee at
Vanderbilt University. Bladders were obtained from adult
female Sprague-Dawley rats (Harlan, Indianapolis, Indiana)
weighing 200 to 250 gm. Bladder harvest, microdissection,
urothelial separation and primary urothelial cell culture
were performed as previously described.5
liquid DAB (Dako). Negative controls were set up at the

same time. No immunoreactivity was observed in these negative control sections.
Real-Time Semiquantitative PCR Analysis
Total RNA was isolated from 5 106 transfected urothelial
cells using an RNAeasy Mini/Maxi kit according to the
manufacturer protocol. RNA quality was checked by running the product on 1.2% formaldehyde-agarose gel.
For cDNA synthesis 1 g total RNA was added to a reaction
mix using an iScript cDNA Synthesis Kit. For real-time
semiquantitative PCR 1 l cDNA template was added to
SYBR Green master mix. To this mix we added PAI-1 forward oligo 5=-GCCAACCCACTACGCCTTCACTCA 3= and
reverse oligo 5=-CCTTTGGCCTGGGTCATCCTTCAT 3=. As
an internal control, we used glyceraldehyde-3-phosphate dehydrogenase with forward oligo 5=-CAACTACATGGTCTACATGTTC 3= and reverse oligo 5=-CTCGCTCCTGGAAGATG 3=.
ICycler PCR conditions were performed according to manufacturer protocols. Relative expression is expressed as the
mean SEM.
In Vitro Genetic Manipulation

of Cultured Adult Rat Bladder Urothelium
DNA plasmids were created with a pLZRS-EGFP backbone (Nolan Laboratory, Stanford, California). The cytomegalovirus promoter from pIRES-EGFP (Clontech, Palo
Alto, California) was excised as a BglII/BamHI fragment.
The fragment was ligated into the BamHI site of the pLZRSEGFP backbone to generate the C7 construct. The C7
construct contains a multiple cloning site under the control of the cytomegalovirus promoter and IRES, which
permits EGFP expression. In addition, the C7 plasmid
contains an episomal property, allowing continued replication of DNA sequences in host daughter cells. The plasmid construct for mutant DNIIR was created by cloning
the mutant sequence into the multiple cloning site in the
C7 backbone.14 The 2 plasmids were used for in vitro
genetic manipulation studies, the C7 plasmid served as
a control and DNIIR served as the experimental condition.
Transfected cell lines were created by applying lipofection
technology with Lipofectamine 2000 (Sigma) according to
manufacturer protocols. Transfected cells were allowed to
grow to 95% confluence and sorted according to EGFP expression via FACS analysis. Transfected EGFP positive cells
were then grown to confluence, passaged and cryopreserved,
as previously described.5
Tissue Recombination Grafts of

Genetically Manipulated Bladder
Urothelium With Embryonic Bladder Mesenchyma
Pregnant Sprague-Dawley rats were sacrificed at embryonic day 18 (plug day considered day 0). The embryos were
isolated and the bladders were removed. The mesenchyma
and urothelium were separated manually under microscopic examination, leaving the mesenchyma behind as a
bladder shell. Embryonic bladder mesenchyma was cut
along 1 wall of the bladder shell using microscissors to
form a more planar structure. Transfected cultured bladder urothelial cells (C7 and DNIIR) were detached from
their flasks by gentle trypsinization and viable cells were
counted on a hemocytometer using trypan blue exclusion.
Urothelial cells (100,000) were resuspended in 40 l rat
collagen matrix and plated. One embryonic bladder mesenchymal shell was placed into each collagen plug before
incubation at 37C to promote plug solidification. Culture
medium was then applied to solidified grafts. Control
grafts consisting of embryonic bladder mesenchyma alone,
primary cultured bladder urothelium alone and collagen
matrix alone were also performed.
Immunocytochemistry
After transfected urothelial cell lines were created (C7 and
DNIIR) immunocytochemistry was performed to confirm
pure urothelial transfectant populations. Slides of cultured
transfected urothelial cells were blocked in Rockland 5%
normal swine serum and 0.1% bovine serum albumin in PBS
for 1 hour. Antibodies to green fluorescent peptide (Santa
Cruz Biotechnology, Santa Cruz, California) (mouse, 1:200)
were incubated overnight at 4C. After the appropriate
washes sections were incubated in biotinylated anti-mouse
antibody (Dako, Carpinteria, California) for 30 minutes at
room temperature. Subsequently sections were incubated in
ABC-horseradish peroxidase (Vector Laboratories, Burlingame, California) for 30 minutes and washed extensively.
Bound antibodies were then visualized by incubation with
Xenograft Harvest and Processing

Hosts were sacrificed 21 and 42 days after grafting, respectively. Mice were sacrificed by anesthetic overdose followed
by cervical dislocation. Grafts were harvested, processed
and paraffin embedded. Sections were cut at 5 m for staining and IHC, as previously described.5
Athymic Nude Mouse Host Xenografting

Male athymic nude mice (CD-1 nu/nu) (Charles River Laboratories, Wilmington, Massachusetts) at ages 7 to 8 weeks
were hosts for subrenal capsule grafts. Surgical grafting
technique was performed as previously described.5
Staining and IHC

Sections for Gomoris trichrome staining were done elsewhere using Gomoris One-Step Green Trichrome Stain Sys-

tem (DakoCytomation, Glostrup, Denmark) and an automated stainer platform.
For IHC all incubations were performed at room temperature unless otherwise stated. Slides were deparaffinized
and hydrated through xylene and graded alcohols. Endogenous peroxidase activity was blocked with peroxidase blocking reagent (Dako) for 15 minutes and sections were washed
in PBS for 5 minutes. IHC was performed using a Vectastain Elite ABC peroxidase kit according to the manufacturer protocol. Sections were incubated overnight at 4C in
a humidified chamber with primary antibodies, including
rabbit broad-spectrum uroplakin (diluted 1:1000), mouse
smooth muscle -actin (Sigma) (diluted 1:1,000), mouse
phosphorylated-SMAD2 (Cell Signaling, Beverly, Massachusetts) (diluted 1:1,000) and rabbit Ki67 antigen (Vector
Laboratories) (diluted 1:1,800). Following overnight incubation the slides were washed in PBS for 5 minutes. Sections
were then incubated for 30 minutes with biotinylated secondary antibody solution diluted in PBS, followed by Vectastain Elite ABC Reagent diluted in PBS for 30 minutes
Between incubations the sections were washed for 5 minutes
in PBS. Immunoreactivity was visualized by incubating sections in Liquid DAB Substrate-Chromogen System (Dako)
for 2 to 4 minutes or using a DAB peroxidase substrate kit
(Vector Laboratories) for 10 minutes. The sections were
washed in double distilled H2O, counterstained with hematoxylin, dehydrated and coverslipped. Negative controls
were set up at the same time as the primary antibody
incubations. They included incubation with PBS in place of
primary antibody. No immunoreactivity was observed in
these negative control sections.
1645
performed. Figure 1, A and B shows immunocytochemical

detection of EGFP in all cell populations (brown areas). As a
control, immunocytochemical staining for EGFP was performed on nontransfected cultured bladder urothelium,
which remained undetectable (data not shown).
Before creating in vivo tissue recombination grafts it was
important to biochemically validate the DNIIR construct.
PAI-1 was used as a reporter for this purpose.15 RNA isolates were obtained from nonserum starved C7 and DNIIR
urothelial cells, and mRNA expression levels for PAI-1 were
detected by real-time PCR. Figure 1, C shows relative expression levels for PAI-1 from C7 and DNIIR mRNA isolates. PAI-1 relative expression levels in the control C7
group were approximately 5-fold higher than those in the
DNIIR group (1.18 103 vs 2.35 104). These findings
confirmed that the DNIIR urothelial cells were biochemically behaving in a fashion consistent with attenuated
TGF- signaling.
Tissue Recombinant Xenografts
A total of 28 C7 control and 27 DNIIR grafts were created.
Each graft consisted of 100,000 transfected urothelial cells
plus 1 embryonic bladder mesenchymal shell. Grafts were
harvested after 2 incubation periods of 21 and 42 days,
respectively. The total number of grafts harvested during
each incubation period was 14 C7 and 15 DNIIR grafts at
21 days, and 14 C7 and 12 DNIIR grafts at 42 days. Grossly
the grafts were whitish-tan and 3 to 4 mm in diameter.
Many DNIIR grafts had a cystic appearance (fig. 2, A), which
was not observed in C7 grafts. Gomoris trichrome staining
of the recombinants revealed well differentiated smooth
Pathological and Statistical Analysis

All slides were reviewed in blinded fashion by a genitourinary
pathologist. The total number of bladder structures for all
grafts performed was tallied and the widest urothelial compartment diameters were measured. Three subcategories of
urothelial compartment diameter were defined as small
0.10 to 0.40, medium 0.41 to 1.0 and large greater than
1.1 mm. Measurements were made by a calibrated micrometer and normalized according to magnification levels. Tests
of statistical significance for each incubation period (21 and
42 days) comparing differences between the total number of
bladder structures for each experimental condition (C7 vs
DNIIR) were done using the Mann-Whitney U test with
p 0.05 considered statistically significant.
RESULTS
In Vitro Genetic Manipulation
of Cultured Adult Rat Bladder Urothelium
After the initial isolation, propagation and verification of
mesenchyma-free populations of cultured adult rat bladder
urothelium, as previously described,5 urothelial cells after
passage 20 were used for in vitro genetic manipulation studies. To select for successful transfectants FACS was applied
to select EGFP expressing cells. Each transfected cohort of
C7 and DNIIR cells was sorted, resulting in approximately
10,000 EGFP positive cells of the 5 106 sent for sorting.
No adverse growth affects were observed after FACS.
Transfected cells were grown to 95% confluence in a T-25
Primeria flask within 1 week. To further verify successful
transfectants immunocytochemical detection of EGFP was
FIG. 1. A, immunocytochemical detection of EGFP (brown areas) in

cultured adult bladder urothelial cells after transfection with C7
vector construct. All cells were successfully transfected with target
DNA. B, immunocytochemical detection of EGFP in cultured adult
bladder urothelial cells after transfection with mutant DNIIR construct. All cells were successfully transfected with target DNA.
A and B, scale bar indicates 50 m. C, real-time semiquantitative
PCR for PAI-1 from mRNA isolates obtained from urothelium transfected with C7 or DNIIR revealed 5-fold difference in PAI-1 relative expression between C7 and DNIIR transfectants, confirming
in vitro biochemical attenuation of TGF- signaling in DNIIR
transfectants.
1646
FIG. 2. A, gross appearance of harvested grafts (arrowheads) on

host kidney. B, small bladder structure with 0.10 to 0.40 mm widest
urothelial diameter. Asterisk indicates bladder structure central
lumen. Gomoris trichrome stain. C, medium bladder structure with
0.41 to 1.0 widest urothelial diameter. Gomoris trichrome stain.
B and C, scale bar indicates 100 m. D, large bladder structure with
greater than 1.1 mm widest urothelial diameter. Gomoris trichrome
stain. Scale bar indicates 400 m.
muscle fibrils surrounding organized suburothelial connective tissue layers (dense blue areas) (fig. 2, B to D). These
connective tissue layers were consistent with lamina propria, which is normally seen in native bladders. In addition,
the connective tissue layers further surrounded central
urothelial clusters. A central lumen with cellular debris
could be seen in some grafts (fig. 2, D). Various sized bladder
structures were observed in the grafted components.
Figure 2, B to D shows examples of small, medium and large
bladder structures.
Immunohistochemical detection of broad-spectrum uroplakin confirmed the presence of mature urothelial cells.
Uroplakin is a selective marker for urothelial cell differentiation and the only epithelial cells in the body that express
uroplakin are urothelium. Uroplakin staining (brown areas)
was present in urothelial clusters, indicating that the cells
were acting in a manner consistent with functional urothelium (fig. 3, B). There was no observable uroplakin staining
in graft stromal cells. Immunohistochemical detection of
smooth muscle -actin was observed as brown areas in the
smooth muscle fibrils in the grafts (fig. 3, C).
Immunohistochemical detection of EGFP confirmed the
presence of genetically manipulated urothelial cells in the
grafted components. This verified that the recombinant tissue growth was not from urothelial cell carryover in mesenchymal preparations. Figure 3, D shows that all cells in the
urothelial compartment stained positive (brown) for EGFP,
confirming that the urothelial cells originated from the genetically manipulated specimens. In addition, grafts of embryonic bladder mesenchyma alone without urothelial constituents survived as fibroblasts but were completely devoid
of complex bladder structures (data not shown), consistent
with previous reports in the literature.5 This further substantiated that the embryonic bladder mesenchyma preparations used were devoid of urothelial contaminants. Grafts
composed of transfected urothelium alone without embry-
onic bladder mesenchyma and collagen matrix alone without

any cellular constituents yielded no evidence of complex
differentiation or bladder tissue formation (data not shown).
The phenotypic characteristics of all bladder structures
were categorized according to overall size based on the widest urothelial diameter of each structure. Three subcategories based on size criteria were identified, including small
0.10 to 0.40, medium 0.41 to 1.0 and large greater than
1.1 mm. To be classified as a bladder structure mature
urothelium expressing uroplakin, smooth muscle fibrils and
suburothelial connective tissue had to be present and confirmed by IHC and staining with broad-spectrum uroplakin,
smooth muscle -actin and Gomoris trichrome. After 21
days of incubation a total of 14 C7 and 15 DNIIR grafts
were evaluated. No bladder structures were identified in
C7 grafts. However, 49 bladder structures were present in
DNIIR grafts. Of the 49 positive DNIIR bladder structures
subcategorization by size showed 30 small, 9 medium and 10
large structures. The Mann-Whitney U test comparing the
number of bladder structures observed at this time between
C7 and DNIIR samples revealed statistical significance
(p 0.001). After 42 days of incubation 14 C7 and 12
DNIIR grafts were evaluated. Of the 14 C7 grafts 6 bladder
structures were observed, including 5 small and 1 medium
structure by size subcategorization. The 12 DNIIR grafts
yielded 27 bladder structures, of which 14 were small, 7
were medium and 6 were large. The Mann-Whitney U test
comparing the number of bladder structures observed at this
time for C7 vs DNIIR revealed statistical significance
(p 0.005).
To biochemically explain the phenotypic growth differences observed between C7 and DNIIR bladder structures
IHC was performed to detect phosphorylated-SMAD2 and
Ki67 antigen. These stains were performed in 42-day bladder structures since this point yielded positive bladder tissue growth for C7 as well as DNIIR grafts. The downstream signaling cascade of the TGF- pathway includes
SMAD2 phosphorylation. IHC for phosphorylated-SMAD2
revealed whether the phenotypic growth differences ob-
FIG. 3. Medium bladder structure. A, H & E. B, IHC detection of

uroplakin (brown areas) confirming mature urothelium. C, IHC
detection of smooth muscle -actin (brown areas). D, IHC detection
of EGFP (brown areas). Scale bar indicates 100 m.

served may have been mediated in part through attenuated
TGF- signaling. Figure 4, A and B shows phosphorylatedSMAD2 staining in the recombinant urothelial compartment on C7 grafts. However, DNIIR bladder structures
had almost no phosphorylated-SMAD2 positive cells in the
recombinant urothelial compartment (fig. 4, C and D). These
findings show that DNIIR urothelial cells were biochemically acting in a fashion consistent with attenuated TGF-
signaling. To compare whether attenuated TGF- signaling
in the recombinant urothelial compartment translated to a
higher proliferative index IHC for Ki67 antigen was performed on the same bladder structures stained for phosphorylated-SMAD2. Figure 5, A and B demonstrates Ki67 staining results in C7 bladder structures. Overall no urothelial
cells were captured as proliferating (fig. 5, A), while only 1 or
2 urothelial cells were proliferating (brown areas) in the
tissue section (fig. 5, B). In contrast, figure 5, C and D shows
Ki67 staining results in DNIIR bladder structures, in which
multiple urothelial cells were proliferating (brown areas).
Ki67 proliferative indexes were calculated for the 6 C7
bladder structures and 6 random DNIIR bladder tissues.
Calculations of Ki67 proliferative indexes were based on the
total number of Ki67 positive urothelial nuclei divided by
the total number of urothelial cells in a bladder structure.
The Ki67 proliferative index for C7 bladder structures was
0.08 vs 0.32 for DNIIR tissues. Overall this 4-fold difference
correlated with the difference in the total number of bladder
structures in this 42-day incubation period (6 C7 vs 27
DNIIR bladder structures, representing a 4.5-fold difference). In addition, bladder tissues demonstrated mature
urothelial differentiation (figs. 4 and 5), as exemplified by
prominent uroplakin expression (fig. 6).
DISCUSSION
To gain a better understanding of bladder development and
disease investigations involving urothelial cell biology will
be of crucial importance. Urothelial growth and differentia-
FIG. 4. IHC detection of phosphorylated-SMAD2 protein (brown

nuclear stain) in same structures as in figures 5 and 6. A and B,
different C7 bladder structures with positive staining in almost
every urothelial nucleus. C and D, different mutant DNIIR bladder
structures with sparse staining in urothelium, showing TGF- signaling attenuation. Asterisk indicates bladder structure central
lumen. Scale bar indicates 50 m.
1647
FIG. 5. IHC detection of Ki67 antigen (brown nuclear stain) in same

structures as in figures 4 and 6. A and B, different C7 bladder
structures with sparse urothelial proliferation. C and D, different
mutant DNIIR bladder structures with significant urothelial proliferation. Asterisk indicates bladder structure central lumen. Scale
bar indicates 50 m.
tion are germane to a number of disease states that affect

the bladder, in part due to the complex cell-cell interactions
that occur between the urothelial compartment and the
stroma. A previous study by Baskin et al demonstrated the
necessity of reciprocal cell-cell signaling interactions between the stromal and epithelial compartments in normal
bladder development.2 Stromal signaling is required to influence epithelial cell differentiation. In turn, the epithelium
influences stromal development and maturation. When
either of these constituents is absent, bladder development
fails to occur. Although to our knowledge the inductive factors driving stromal-epithelial interactions in the bladder
are unknown, various growth factors have been implicated,
including TGF- isoforms.16
The TGF- signaling pathway was shown to affect the
normal growth and development of different organ systems,
and yet its exact role in the bladder remains unknown.
FIG. 6. IHC detection of broad-spectrum uroplakin (brown areas) in

same structures as in figures 4 and 5. A and B, different C7
bladder structures. C and D, different mutant DNIIR bladder structures. Asterisk indicates bladder structure central lumen. Scale bar
indicates 50 m.
1648
TGF- exerts its effects through a heteromeric receptor complex that consists of the membrane associated serine-threonine kinases, TGF- receptor type I and TGF- receptor type
II. Each receptor type is essential for TGF- ligand induced
effects on the cells.14 The TGF- activated signaling cascade
can occur via multiple pathways, including SMAD proteins,
phosphatidylinositol 3-kinase, protein phosphatase 2A/p70
S6 kinase, various mitogen-activated protein kinase pathways and Rho proteins.17 To our knowledge no studies exist
on the role of TGF- in the urothelium. Multiple studies
have elucidated the role of TGF- as a growth inhibitor in
other epithelial cell types, including the liver, intestine,
skin, breast, prostate and lung.18 20 Based on these reports
we examined the role of TGF- in urothelial development by
applying in vivo tissue recombination models.
Although Thompson et al pioneered genetic manipulation and tissue recombination techniques to study prostate development,12 to our knowledge no reports in the
literature exist with this application to study bladder
development. We reported the ability to generate longterm mesenchyma-free primary cultures of rodent bladder
urothelium that can recapitulate bladder tissue formation
in a tissue recombination model.5 These findings illustrate the ability of long-term (greater than passage 20)
primary cultures of bladder urothelium to retain signaling
properties that can induce complex stromal differentiation and bladder tissue formation. With this knowledge
and by adopting the concepts of Thompson et al12 we
genetically manipulated the urothelial compartment to
over express mutant DNIIR. Our goal was to generate
transfected primary urothelial cell lines and determine
whether the tissue recombination model could be used to
investigate the effects of attenuated TGF- signaling in
an in vivo environment.
Our results illustrate that the attenuation of urothelial
TGF- signaling led to bladder structures that grew features
that were more rapid, abundant and robust in size. This
suggests that ubiquitous TGF- signaling in the bladder
urothelium acts in a growth inhibitory fashion, paralleling
findings observed in epithelial compartments of other organ
systems.18 20
Mechanistically the attenuation of TGF- signaling in
the urothelium, thereby removing TGF- growth inhibitory
effects, could potentially allow the urothelium to proliferate
at an increased rate. This could directly translate to enhancement in the number of urothelial-stromal interactions
with overall improvement in the rate of bladder structure
growth and development. In the recombinants IHC detection for phosphorylated-SMAD2 revealed intact signaling in
the recombinant urothelium of C7 bladder structures but
minimal detection in the recombinant urothelium of DNIIR
bladder structures. This confirmed the biochemical attenuation of TGF- signaling in the recombinant urothelium of
the DNIIR bladder structures. The same bladder structures
were also subjected to IHC detection for Ki67 antigen, revealing that proliferative indexes were 4-fold higher in the
recombinant urothelium of the DNIIR bladder structures.
These data provide further evidence that TGF- signaling
attenuation led to improved proliferative capacity in the
recombinant urothelial compartment and argue that TGF-
may be a growth inhibitory factor in urothelial development.
Although a slightly greater number of DNIIR grafts were
available for analysis at 21 vs 42 days (15 and 12 grafts,
respectively), this could not entirely explain why a greater

number of bladder structures were observed in the earlier
21-day incubation period vs those at 42 days (49 vs 27). A
significant number of bladder structures (30), which at 21
days were subcategorized as small structures, if allowed to
incubate for a longer period could have potentially coalesced
and grown to form more medium and large bladder structures. This in turn would have decreased the overall number
of bladder structures, ie 5 small structures fusing to form 1
large structure, and it could explain the discrepancies seen
between the 2 incubation periods.
CONCLUSIONS
Primary cultures of rodent bladder urothelium can be generated and genetically manipulated. To our knowledge we
report the first application demonstrating that a chimeric
bladder was formed from genetically manipulated primary
cultured bladder urothelium recombined with embryonic
bladder mesenchyma. In addition, we provide evidence that
the role of TGF- in bladder urothelium may be as a growth
inhibitor. These findings could result in future therapeutic
targets along the TGF- pathway with hopes of improving
urothelial cell biology for the multitude of disease states that
affect the bladder.
ACKNOWLEDGMENTS
Dr. T. T. Sun, New York University provided broad-spectrum uroplakin. Gomoris trichrome staining was done at
the Vanderbilt mouse histology core facility. Mary Dietrich
performed the statistical data analysis, Katrina Saba provided technical assistance and Robert J. Matusik critically
reviewed the manuscript.

ABC avidin-biotin complex
DAB 3,3=-diaminobenzidine tetrahydrochloride
DNIIR dominant negative form of type II TGF-
receptor
EGFP enhanced green fluorescence peptide
FACS fluorescence activated cell sorting
IHC immunohistochemistry
PAI-1 plasminogen activator inhibitor type 1
PCR polymerase chain reaction
SMA smooth muscle -actin
TGF- transforming growth factor-
REFERENCES
1.
2.
3.
4.
Baskin LS, Hayward SW, Sutherland RA, DiSandro MJ,

Thomson AA, Goodman J et al: Mesenchymal-epithelial
interactions in the bladder. World J Urol 1996; 14: 301.
Baskin LS, Hayward SW, Young P and Cunha GR: Role of
mesenchymal-epithelial interactions in normal bladder development. J Urol 1996; 156: 1820.
DiSandro MJ, Li Y, Baskin LS, Hayward S and Cunha G:
Mesenchymal-epithelial interactions in bladder smooth
muscle development: epithelial specificity. J Urol 1998;
160: 1040.
Liu W, Li Y, Cunha S, Hayward G and Baskin L: Diffusable
growth factors induce bladder smooth muscle differentiation. In Vitro Cell Dev Biol Anim 2000; 36: 476.

5.
6.
7.
8.
9.
10.
11.
12.
13.
Oottamasathien S, Williams K, Franco OE, Thomas JC, Saba

K, Bhowmick NA et al: Bladder tissue formation from cultured bladder urothelium. Dev Dyn 2006; 235: 2795.
Bhowmick NA, Neilson EG and Moses HL: Stromal fibroblasts
in cancer initiation and progression. Nature 2004; 432: 332.
Sharif-Afshar AR, Donohoe JM, Pope JCT, Adams MC, Brock
JW 3rd and Bhowmick NA: Stromal hyperplasia in male
bladders upon loss of transforming growth factor-beta signaling in fibroblasts. J Urol 2005; 174: 1704.
Roszell JA, Douglas CJ and Irving CC: Polyamine-stimulated
growth of cultured rat urinary bladder epithelial cells. Cancer Res 1977; 37: 239.
Liebert M, Wedemeyer G, Chang JH, Stein JA, McKeever PE,
Carey TE et al: Comparison of antigen expression on normal urothelial cells in tissue section and tissue culture.
J Urol 1990; 144: 1288.
Southgate J, Hutton KA, Thomas DF and Trejdosiewicz LK:
Normal human urothelial cells in vitro: proliferation and
induction of stratification. Lab Invest 1994; 71: 583.
Kurzrock EA, Lieu DK, deGraffenried LA and Isseroff RR: Rat
urothelium: improved techniques for serial cultivation, expansion, freezing and reconstitution onto acellular matrix.
J Urol 2005; 173: 281.
Thompson TC, Southgate J, Kitchener G and Land H: Multistage carcinogenesis induced by ras and myc oncogenes in a
reconstituted organ. Cell 1989; 56: 917.
Yang G, Timme TL, Park SH, Wu X, Wyllie MG and Thompson
TC: Transforming growth factor beta 1 transduced mouse
prostate reconstitutions: II. Induction of apoptosis by doxazosin. Prostate 1997; 33: 157.
14.
1649
Lin HY, Wang XF, Ng-Eaton E, Weinberg RA and Lodish HF:

Expression cloning of the TGF-beta type II receptor, a functional transmembrane serine/threonine kinase. Cell 1992;
68: 775.
15. Hua X, Miller ZA, Wu G, Shi Y and Lodish HF: Specificity in
transforming growth factor beta-induced transcription of
the plasminogen activator inhibitor-1 gene: interactions of
promoter DNA, transcription factor muE3, and Smad proteins. Proc Natl Acad Sci U S A 1999; 96: 13130.
16. Baskin LS, Sutherland RS, Thomson AA, Hayward SW and
Cunha GR: Growth factors and receptors in bladder development and obstruction. Lab Invest 1996; 75: 157.
17. Bhowmick NA, Chytil A, Plieth D, Gorska AE, Dumont N,
Shappell S et al: TGF-beta signaling in fibroblasts modulates the oncogenic potential of adjacent epithelia. Science
2004; 303: 848.
18. Romero-Gallo J, Sozmen EG, Chytil A, Russell WE, Whitehead
R, Parks WT et al: Inactivation of TGF-beta signaling in
hepatocytes results in an increased proliferative response
after partial hepatectomy. Oncogene 2005; 24: 3028.
19. Barnard JA, Beauchamp RD, Coffey RJ and Moses HL: Regulation of intestinal epithelial cell growth by transforming
growth factor type beta. Proc Natl Acad Sci U S A 1989; 86:
1578.
20. Forrester E, Chytil A, Bierie B, Aakre M, Gorska AE, SharifAfshar AR et al: Effect of conditional knockout of the type II
TGF-beta receptor gene in mammary epithelia on mammary gland development and polyomavirus middle T antigen induced tumor formation and metastasis. Cancer Res
2005; 65: 2296.
Association of Transforming Growth Factor-1

Gene Polymorphism With Familial Vesicoureteral Reflux
Seika Kuroda, Valeria Solari and Prem Puri*
From the Childrens Research Centre, Our Ladys Childrens Hospital and National Childrens Hospital Tallaght, Dublin, Ireland
Purpose: Familial clustering of vesicoureteral reflux suggests that genetic factors have an important role in the pathogenesis
of vesicoureteral reflux. Transforming growth factor-1 is a multifunctional peptide that controls proliferation and differentiation in many cell types. Recently an association between the transforming growth factor-1 509 and 869 gene
polymorphism, and renal parenchymal scarring was reported. We investigated the genetic contribution of transforming
growth factor-1 in familial vesicoureteral reflux by examining the genotype frequencies of transforming growth factor-1
polymorphic variants.
Materials and Methods: The study included 141 families in which 1 or more siblings had primary vesicoureteral reflux.
Renal parenchymal scarring was assessed using dimercapto-succinic acid scans. Genotyping was performed in 280 patients
with vesicoureteral reflux, including 133 index patients and 147 siblings, and in 74 controls for the position 509 and the
coding region at position 10 (869) of the transforming growth factor-1 gene polymorphism by polymerase chain reaction,
gel analysis and appropriate restriction digest.
Results: The genotype frequency of 509CC was significantly increased in the familial vesicoureteral reflux group compared
to controls (58% vs 33%, p 0.01), whereas 509TT genotype frequency was significantly lower in the familial vesicoureteral
reflux group compared to controls (7.5% vs 28%, p 0.01). Similarly there was a significant increase in the 869TT genotype
(52% vs 32%, p 0.05), while the 869CC genotype was significantly lower in patients with familial vesicoureteral reflux
compared to controls (11% vs 24%, p 0.01). There were no significant differences in transforming growth factor-1 genotype
distribution between patients with vesicoureteral reflux with and without renal parenchymal scarring.
Conclusions: To our knowledge this study demonstrates for the first time the association of the cytokine transforming
growth factor-1 gene polymorphism in patients with familial vesicoureteral reflux. Individuals with the transforming
growth factor-1 509CC and 869TT genotype may have increased susceptibility to vesicoureteral reflux.
Key Words: bladder; ureter; polymorphism, genetic; vesico-ureteral reflux; transforming growth factor beta
rimary VUR is the most common urological anomaly

in children and it has been reported in 30% to 50% of
those who present with urinary tract infection. The
association of VUR, urinary tract infection and renal damage is well known.1,2 The hereditary and familial nature of
vesicoureteral reflux is now well recognized.3 Several studies have shown that siblings of children with VUR have a
much higher chance of reflux than the general pediatric
population with a reported incidence of 26% to 50%.35 The
transmission of VUR from parent to child has been reported
at a rate of up to 66%.3 Furthermore, 43% of infants born to
mothers with reflux nephropathy were found to have VUR.6
Familial clustering of VUR implies that genetic factors may
have an important role in the pathogenesis of reflux. A
recent study of 46 twin pairs provides the strongest evidence
for a genetic origin, in that there was 80% to 100% concordance for VUR in monozygotic twins compared with 35% to
50% concordance in dizygotic twins.4
Urinary tract development depends on reciprocal inductive signals between the ureteral bud and the metanephric
Study received hospital Ethics Committee approval.

* Correspondence: The Childrens Research Centre, Our Ladys
Childrens Hospital, Crumlin, Dublin 12, Ireland (telephone: 3531-4096420; FAX: 353-1-4550201; e-mail: prem.puri@ucd.ie).
0022-5347/07/1784-1650/0
mesenchyma. Upon induction by the ureteral bud, which is

an epithelial outgrowth of the wolffian duct, the metanephric mesenchyma undergoes a series of morphogenetic events
that convert the mesenchyma to epithelium and eventually
generates most of the mature kidney tubules. The collecting
duct system is elaborated by the branching ureter, of which
the growth depends on signals from the metanephric mesenchyma. As the ureter and kidney form, the distal ends of
the mesonephric ducts widen and become incorporated into
the posterior bladder wall, encompassing the 2 ureteral orifices and trigone. Abnormal position or development of the
ureteral end can lead to anomalies of the urethrovesical
junction, such as VUR.7,8
TGF- belongs to a family of multifunctional growth factors that have profound regulatory effects on many developmental and physiological processes.9 There is growing evidence that TGF-1 regulates the growth and branching of
the ureteral bud as it forms the collecting system and ureter.10,11 Several polymorphisms in the TGF-1 gene have
been identified. There is mounting evidence that the ability
of an individual to produce high or low levels of TGF-1 may
be genetically predetermined, and polymorphisms at 509
and 869 may be associated with higher or lower TGF-1
synthesis in vitro.12,13 Recently an association between
TGF-1 509 and 869 gene polymorphism, and VUR and
1650
Vol. 178, 1650-1653, October 2007

Printed in U.S.A.
DOI:10.1016/j.juro.2007.03.199
GENE POLYMORPHISM AND FAMILIAL VESICOURETERAL REFLUX
1651
RPS was reported.14 16 We investigated whether TGF-1

polymorphism is involved in familial VUR.
Subjects
Included in the study were 280 patients, including 162 girls
and 118 boys, in a total of 141 families in which 1 or more
siblings had primary VUR. They consisted of 133 index
patients and 147 siblings. The control group consisted of 74
Irish individuals without renal disease. Voiding cystourethrogram was performed to determine the presence and
grade of reflux. RPS was evaluated with 99technetiumdimercapto-succinic acid renal scan. All renal scans were
performed 3 to 4 months after a urinary tract infection. The
group was subdivided into 85 patients with and 195 without
RPS. The study was approved by the hospital Ethics Committee for obtaining DNA samples from patients for VUR
genetic studies.
DNA Extraction and TGF-1 Genotyping
Genomic DNA was extracted from peripheral blood leukocytes
using a salting out procedure and the concentration was determined by an ND-1000 spectrophotometer (NanoDrop, Wilmington, Delaware). All samples were diluted to 100 ng/l
with tris-acetate-ethylenediaminetetraacetic acid. Genotyping
was performed by PCR restriction fragment length polymorphism. For 509 polymorphism of the TGF-1 gene the
primer sets used were forward primer 5=-CAGTAAATGTATGGGGTCGCAG-3= and reverse primer 5=-GGTGTCAGTGGGAGGAGGG -3=. PCR was performed in a final volume of 25
l containing 100 ng genomic DNA, 10 pmol concentrations
FIG. 2. Restriction analysis of TGF-1 869T-C polymorphism. Amplified 199 bp fragments were digested with Not1 and separated on
2% agarose gel.
FIG. 1. Restriction analysis of TGF-1 509C-T polymorphism. Amplified 153 bp fragments were digested with Bsu36I and separated
on 2% agarose gel.
of each primer, a 200 m/l concentration of deoxynucleoside

triphosphate, 1.5 mmol/l MgCl2 and AmpliTaq Gold. PCR
reactions were performed on a Primus 96-plus thermal cycler (MWG Biotech, Ebersberg, Germany). PCR began with
denaturation at 95C for 10 minutes, followed by 40 cycles of
denaturation at 95C for 30 seconds and annealing at 60C for
1 minute. PCR products (2 l) were digested with BSU36I
restriction enzyme and separated on 2% agarose gel. After
ethidium bromide staining 118 and 35bp fragments appeared on the polymorphic 509C allele, whereas a 153 bp
fragment appeared on the common 509T allele.
For codon 10 polymorphism the primer sets used were
forward primer 5=- CTCCGGGCTGCGGCTGCGGC-3= and
reverse primer 5=-CGAGCCAGGGGGAGGTGCCG -3=. PCR
was performed in a final volume of 25 l containing 100 ng
genomic DNA, 10 pmol concentrations of each primer, a 200
m/l concentration of deoxynucleoside triphosphate, 1.5
mmol/l MgCl2 and Platinum Taq. PCR began with denaturation at 95C for 5 minutes, followed by 35 cycles of
denaturation at 94C for 30 seconds, annealing at 62C for 30
1652
seconds and extension at 72C for 1 minute. PCR products

(3 l) were digested with Not1 restriction enzyme and separated on 2% agarose gel. After ethidium bromide staining
182 and 17 bp fragments appeared on the polymorphic
509C allele, whereas a 199 bp fragment appeared on the
common 509T allele.
Statistics
Statistics were performed using Statcel2 (OMS Publishing,
Tokorozawa, Japan). Allele genotype frequencies between
the VUR and control groups were evaluated by the chisquare test.
TABLE 2. TGF-1 polymorphisms in patients with VUR with and

without RPS
No. Pts (%)
Polymorphism
Overall
509:
CC
CT
TT
C:T
Leu10Pro:
TT
TC
CC
C:T
RPS
85
No. RPS
133
53 (62)
25 (29)
7 (8.2)
0.77/0.23
110 (56)
71 (36)
14 (7.2)
0.75/0.25
38 (45)
38 (45)
9 (11)
0.67/0.33
91 (47)
83 (43)
21 (11)
0.68/0.32
RESULTS
TGF-1 509T-C and 869C-T is a biallelic restriction fragment length polymorphism site that was readily detected by
Bsu36I and Not1 restriction analysis (figs. 1 and 2).
Tables 1 and 2 list genotype and allele frequencies. The
genotype frequency of 509CC was significantly increased
in the familial VUR group compared to controls (58% vs
33%, p 0.01), whereas 509TT genotype frequency was
significantly lower in the familial VUR group than in controls (7.5% vs 29%, p 0.01). Similarly there was a significant increase in the 869TT genotype (46% vs 31%,
p 0.05), while the 869CC genotype was significantly
lower in patients with familial VUR compared to controls
(11% vs 24%, p 0.01).
There was no significant difference in gender distribution
between patients with (60.2% female and 39.8% male) and
without (51.8% female and 8.2% male) RPS (p 0.19) Reflux
grade was significantly higher in the group of patients with
RPS vs without RPS (p 0.01). There were no significant
differences in TGF-1 genotype distribution between patients who had VUR with RPS and patients without RPS.
DISCUSSION
We investigated the contribution of the TGF-1 polymorphism in familial VUR. The 2 polymorphisms investigated
in this study (509T-C and 869C-T) in the TGF-1gene
were reported to be associated with transcriptional activity
of the gene or the serum level of the gene product.16,17 Our
results show that the 509T-C polymorphism of the TGF-1
gene is associated with familial VUR and the CC genotype is
more prevalent in index patients and siblings with VUR
than in controls. Similarly the genotype frequency of 869
TABLE 1. TGF-1 polymorphisms in patients with VUR

and controls
No. Pts (%)
Polymorphism
Overall
509:
CC
CT
TT
C:T
Leu10Pro:
TT
TC
CC
C:T
VUR
280
Index VUR
133
Controls
74
163 (58.2)
96 (34.3)
21 (7.5)
0.76/0.24
77 (58)
45 (34)
11 (8)
25 (33.8)
29 (39.1)
20 (27.0)
0.53/0.47
129 (46)
121 (43.2)
30 (10.7)
0.71/0.29
60 (45)
62 (47)
11 (8)
24 (32.4)
34 (45.9)
16 (21.6)
0.55/0.45
TT is more increased in patients with familial VUR than in

controls. These results indicate that individuals with
TGF-1 509CC and 869TT gene polymorphism may be at
increased risk for VUR.
There is increasing evidence to suggest that TGF- affects critical processes during urinary tract development,
including cell proliferation, differentiation, cell fate determination apoptosis and morphogenesis, by activating discrete
signaling pathways.17 In particular TGF-1 was shown to
inhibit the ureteral bud in vitro.18 Culture of embryonic
kidney explants in the presence of TGF-1 affected whole
organ growth and the ability of the ureteral bud to undergo
branching morphogenesis.11,19 In the current study polymorphism at the 509 and 869 position of TGF-1 was
studied. Our study findings suggest that polymorphisms at
TGF -1-509CC and 869TT may result in abnormal development of the urethrovesical junction, causing VUR.
Significant variability exists in the evolution of acute
pyelonephritis following urinary tract infection in patients
with VUR since some children have RPS and others do not.
However, the exact pathogenesis of renal scarring leading
to RN is unknown. Substantial evidence in the literature
supports the evidence that TGF-1 is a key mediator of
fibrotic disease. Recently the polymorphism of TGF-1 at
position 869, Lue10-Pro (T-C), codon 10 was found to be
associated with diabetic nephropathy.20 The polymorphic
variant of the TGF-1 gene at position 509 was reported by
Cotton et al in children with renal scarring following urinary
tract infections.14 More recently a significant association
between codon region 10 and the polymorphic variant 509
of the TGF-1 gene in renal scarring in children with VUR
was reported.15,16 In our study there was no significant
association between any of the 2 TGF-1 polymorphisms
analyzed and the risk of RPS. The reason for these differences between our study and others remains unclear. An
explanation may be that we had few patients with severe
renal damage in this series.
The Achilles heel of association studies is stratification,
that is the finding of an apparent association that is actually
due to a difference between cases and controls that is unrelated to the disease being studied. It is not possible to say
that the current result is not the result of such an artifact,
especially since the control cohort was smaller than the case
cohort. This could have been avoided by a different study
design, incorporating patient genotyping and then using the
transmission disequilibrium test and/or linkage analysis.
Genotyping the parents would also have permitted the de-

termination of haplotypes, which potentially extract more
information from the sample than simple genotypes. More
studies using larger control population as well as genotyping
of parents of patients with VUR are required to confirm our
findings.
CONCLUSIONS
Familial clustering of VUR implies that genetic factors may
have an important role in the pathogenesis of reflux. Our
study of the genetic contribution of TGF-1 to familial VUR
demonstrated associations between the TGF-1 509CC
and 869TT gene polymorphisms, and VUR.

PCR
RPS
TGF
VUR
polymerase chain reaction

renal parenchymal scarring
transforming growth factor
vesicoureteral reflux
REFERENCES
Smellie JM, Edwards D, Hunter N, Normand ICS and Prescod
N: Vesico-ureteric reflux and renal scarring. Kidney Int,
suppl., 1975; 4: S62.
2. Gleeson FV and Gordon I: Imaging in urinary tract infection.
Arch Dis Child 1991; 66: 1282.
3. Noe HN, Wyatt RJ, Peeden JN Jr and Rivas ML: The transmission of vesicoureteral reflux from parent to child. J Urol
1992; 148: 1869.
4. Kaefer M, Curran M, Treves ST, Bauer S, Hendren WH, Peters
CA et al: Sibling vesicoureteral reflux in multiple gestation
births. Pediatrics 2000; 105: 800.
5. Chertin B and Puri P: Familial vesicoureteral reflux. J Urol
2003; 169: 1804.
6. North RA, Taylor RS and Gunn TR: Pregnancy outcome in
women with reflux nephropathy and the inheritance of
vesico-ureteric reflux. Aust N Z J Obstet Gynaecol 2000; 40:
280.
7. Thomas JC, DeMarco RT and Pope JC 4th: Molecular biology
of ureteral bud and trigonal development. Curr Urol Rep
2005; 6: 146.
8.
9.
10.
11.
12.
13.
14.
15.
1.
16.
17.
18.
19.
20.
1653
Bush KT, Sakurai H, Steer DL, Leonard MO, Sampogna RV,

Meyer TN et al: TGF-beta superfamily members modulate
growth, branching, shaping, and patterning of the ureteric
bud. Dev Biol 2004; 266: 285.
Lawrence DA: Transforming growth factor-beta: an overview.
Kidney Int, suppl., 1995; 49: S19.
Karihaloo A, Nickel C and Cantley LG: Signals which build a
tubule. Nephron Exp Nephrol 2005; 100: e40. Epub February 22, 2005.
Michael L and Davies JA: Pattern and regulation of cell proliferation during murine ureteric bud development. J Anat
2004; 204: 241.
Grainger DJ, Heathcote K, Chiano M, Snieder H, Kemp PR,
Metcalfe JC et al: Genetic control of the circulating concentration of transforming growth factor type beta1. Hum Mol
Genet 1999; 8: 93.
Yamada Y: Association of a Leu(10)Pro polymorphism of the
transforming growth factor-beta1 with genetic susceptibility to osteoporosis and spinal osteoarthritis. Mech Ageing
Dev 2000; 116: 113.
Cotton SA, Gbadegesin RA, Williams S, Brenchley PE and
Webb NJ: Role of TGF-beta1 in renal parenchymal scarring
following childhood urinary tract infection. Kidney Int
2002; 61: 61.
Lee-Chen GJ, Liu KP, Lai YC, Juang HS, Huang SY and Lin
CY: Significance of the tissue kallikrein promoter and
transforming growth factor-beta1 polymorphisms with renal progression in children with vesicoureteral reflux. Kidney Int 2004; 65: 1467.
Solari V, Owen D and Puri P: Association of transforming
growth factor-beta1 gene polymorphism with reflux nephropathy. J Urol 2005; 174: 1609.
Vrljicak P, Myburgh D, Ryan AK, van Rooijen MA, Mummery
CL and Gupta IR: Smad expression during kidney development. Am J Physiol Renal Physiol 2004; 286: 625.
Clark AT, Young RJ and Bertram JF: In vitro studies on the
roles of transforming growth factor-beta 1 in rat metanephric development. Kidney Int 2001; 59: 1641.
Rogers SA, Ryan G, Purchio AF and Hammerman MR: Metanephric transforming growth factor-beta 1 regulates
nephrogenesis in vitro. Am J Physiol 1993; 264: F996.
Wong TY, Poon P, Chow KM, Szeto CC, Cheung MK and Li PK:
Association of transforming growth factor beta (TGF-)
T869C(Leu10Pro) gene polymorphism with type 2 diabetic
nephropathy in Chinese. Kidney Int 2003; 63: 1831.
Pediatric Urology Medal

A. Barry Belman
t is a tremendous honor to recognize the recipient of the
2006 Pediatric Urology Medal, the highest accolade offered in our subspecialty worldwide. I had a similar
honor 6 years ago when I presented this medal to my former
mentor, Dr. John Woodard. I now have the honor of recognizing my mentor, my associate of 20 years and one of my
closest friends, Dr. A. Barry Belman.
I would like to share some of Barrys life and personality,
and highlight his career achievements. To do so with someone as unique as Doctor Belman requires a combination of
humor and deep heartfelt respect.
After growing up in Tucson, Arizona, Barry attended
college at the University of Arizona where he joined a fraternity that had several premed students. This would eventually influence his career choice but initially he started out
in agriculture school studying hydrology, focusing on irrigation methods in the desert. Perhaps this too had some
influence on his ultimate choice of urology as a specialty. He
then developed an interest in veterinary medicine but his
mom told him why waste your time with animals when you
could be helping people.
In fact, Barry was close to his mom who was an influential figure in his life. Not only did she give him love, nurturing and guidance, but she, like Barry, had a great sense of
humor. Once during a visit with Barry and Paula, his mom
at the age of 75 walked into their house, turned to look at
herself in a mirror in the entrance hall and declared Theyre
just not making mirrors the way they use to!
After college Barry was admitted to medical school at
Northwestern in Chicago. He would spend the next 12 years
of his life in Chicago as a medical student, resident and
junior attending before coming to Childrens Hospital in
Washington D. C. For many reasons these would be the most
pivotal years of his personal and professional life.
The first life changing event occurred during the last day
of an externship when he by chance happened to see an
attractive young lady named Paula Yonover on the first day
of her research project in the hospital cafeteria. Barry was
having lunch with a guy on whom Paula had a huge crush
but Barry asserted himself (can you imagine that??), managed to get her phone number and a year later married
Paula on the day after he graduated from medical school.
They started their family with his son Peter, followed by 3
wonderful daughters, Lisa, Trina and Jessica. He is a devoted family man, having now been married to Paula for 42
years in a marriage that Paula describes as wonderful,
respectful and with lots of humor. As a father, Barry was a
strict disciplinarian but loving and involved with his kids.
The family continues to grow, and Barry and Paula are now
expecting grandchildren 7 and 8.
The second pivotal event for Barry in Chicago was when
he was offered a position in the urology residency program at
Northwestern. For the next 5 years he was exposed to the
0022-5347/07/1784-1654/0
most important influential person of his professional and

academic career, Dr. John Grayhack, chairman of the Northwestern Department of Urology. John Grayhack was Doctor
Belmans mentor and role model. To quote Barry, he taught
absolute honesty and to always think in a straight line.
Doctor Grayhack introduced Doctor Belman and all of his
residents to the science of medicine and would not accept
anything else but objectivity in thinking and patient management. If the residents could not justify a procedure, they
would get blasted. Under Doctor Grayhacks mentorship, the
residents at Northwestern were taught to practice evidence-based medicine decades before the term was ever
created.
During his chief residency Doctor Belman spent 6 months
with Dr. George Kaplan who was the acting chief of Pediatric Urology. Barry credits George for stimulating him to go
into pediatric urology. However, it was Dr. Lowell King, who
returned from Presbyterian-St. Lukes Hospital to become
the permanent chief of pediatric urology at Northwestern,
who gave Doctor Belman his first big opportunity. With
Doctor Grayhacks support, Lowell hired Barry without even
an interview or discussion of salary. As Barry has said
himself, Lowell did not know what he was getting . . . Im
sure he regretted it.
Doctor Belman also spent a year at Northwestern with
Doctor Douglas Stephens, who Barry describes as the
worlds most keen medical observer. Whereas Doctor Grayhack had introduced Barry to the science of medicine,
Douglas Stephens taught and reinforced it. For Douglas,
every patient was a laboratory. Today at the age of 95,
Douglas Stephens continues to be one of pediatric urologys
most phenomenal thinkers.
During his early attending years at Northwestern Doctor
Belman also came to know and respect Dr. Jean Smellie, a
British pediatrician. According to Barry, Jean Smellie represented the ideal that we should all be striving for objective thinking, compulsive attention to detail and conscientious follow-through. Jean Smellie was a major early
proponent for nonsurgical management of vesicoureteral reflux, an approach also endorsed early in this country by
Lowell King. Until then reflux was considered to be primarily a surgical condition but the strong commitment from a
handful of individuals, including Jean Smellie, Philip Ransley, Lowell King and Barry Belman, would change the management of this condition for the next 3 decades.
Another big opportunity while at Northwestern came
when Panayotis Kelalis and Lowell King invited Doctor Belman to contribute to the 1st edition of Clinical Pediatric
Urology. Initially he was supposed to only write the chapter
on hypospadias, but when Lowell King returned from a
6-month sabbatical with only a few chapters written, Doctor
Belman volunteered his services and quickly wrote 6 chapters of the book. This resulted in Panayotis and Lowell
1654
Vol. 178, 1654-1655, October 2007

Printed in U.S.A.
DOI:10.1016/j.juro.2007.07.069
PEDIATRIC UROLOGY MEDAL

naming him as an Associate Editor of the 1st edition, which
was a major step in his academic career.
Doctor Belman came to Childrens Hospital 30 years ago
in August 1976. During the next 30 years he would accumulate an extensive list of academic and professional accomplishments. Doctor Belman has authored more than 130
published articles with numerous contributions in the areas
of vesicoureteral reflux, urinary tract infection, hypospadias, testis tumors, varicocele and ureteropelvic junction
obstruction. He has authored more than 30 book chapters on
a wide variety of topics, served on the Editorial Board of The
Journal of Urology from 1991 to 1999, and served as associate editor of the 1st edition of Clinical Pediatric Urology,
co-editor of the 2nd and 3rd editions, and chief editor of the
4th edition of this text which is often referred to as the bible
of pediatric urology.
Even more impressive are the number of novel ideas that
can be credited to Doctor Belman. These ideas reflect a stop,
look and think approach which he has consistently applied
to the management of urological disorders in children. In his
first article, which was published as the lead article in the
New England Journal of Medicine, Doctor Belman was the
first in the world literature to report renin-mediated hypertension associated with unilateral hydronephrosis. The first
of his many independent ideas was published in a 1970
issue of JAMA where he described the nonoperative management of unilateral renal vein thrombosis, a disease
previously treated exclusively with surgical nephrectomy. In
1972 he categorized the association of different genitourinary anomalies with the level of imperforate anus. In 1982
he was the first to recognize and report an association between racial factors and the incidence of reflux and urinary
tract infection, noting a much lower rate in children of
African descent. He was also the first to observe the association between ipsilateral testicular hypotrophy and varicoceles in adolescents, as well as the phenomenon of catch-up
growth after varicocelectomy.
Doctor Belman described and promoted the use of
Durham Smiths deepithelialized flap for coverage in primary hypospadias repairs, greatly reducing the rate of fistula formation. He performed the first testis-sparing enucleation of a prepubertal testis teratoma while in Chicago and
this was later published in a larger series from Childrens in
1990, triggering the application of this approach to a variety
of other benign testicular tumors. In 1998 he published a
novel modification of the onlay hypospadias repair, the split
prepuce in situ onlay, and his 17-year experience with this
technique was reported at this meeting. More recently, Doctor Belman first observed and reported that intrauterine
torsion leading to a vanished testis is almost always a late
scrotal event, even in cases of abdominal blind-ending vas
deferens and vessels.
1655
Doctor Belman also served in an impressive number of

key leadership roles in our national organizations. In 1978
Doctor King appointed him Program Chairman for the
American Academy of Pediatrics (AAP), Section on Urology
meeting. From 1979 to 1983 he served on the American
Board of Urology exam committee. In 1982 he was Program
Chairman for the Society for Pediatric Urology meeting. He
served on the Executive Committee for the AAP Section on
Urology from 1982 to 1984, as Secretary from 1984 to 1987
and as Chairman in 1988 to 1989. He was elected President
of the Society for Pediatric Urology in 1995 to 1996.
Doctor Belman has also contributed extensively to the
training of medical students, residents and urologists. From
1983 to 1990 he served as an examiner for the American
Board of Urology oral examination. He was the American
Medical Association representative to the Residency Review
Committee from 1995 to 2000. He gave the AUA postgraduate course on pediatric urology from 1988 to 1996 and
served as the first AAP Section on Urology Education Committee chairman from 1990 to 1993. When serving in these
roles, Doctor Belman always tried to introduce new young
speakers to give them a chance for exposure. In addition, he
instructed countless medical students, mentored more than
300 residents and most recently has helped to establish and
fund the A. Barry Belman Fellowship in Pediatric Urology.
In fact, aside from close relationships and treating children with urological problems, teaching has been the most
gratifying aspect of Doctor Belmans career. He truly has a
passion and gift for bringing pediatric urology down to a
level that makes it understandable for any level of trainee. It
is the teaching of pediatric urology at all levels that Doctor
Belman believes has been his biggest impact. For this, Doctor Belman has received numerous teaching awards.
Those of us who have worked and trained with Doctor
Belman will attest to his love and ability to educate and
train, and we are indebted to him for the many opportunities
that he created for us. As evidence of his gift for teaching and
inspiration, 14 former residents who trained with and were
influenced by Doctor Belman have gone on to pursue a
career in pediatric urology. In fact, when I interviewed with
him in 1985 it was my own observation of his relationship
with the residents and his fellow Steve Skoog which helped
convince me to accept his offer to join him in practice. It was
the best decision of my professional career. Doctor Belman
has truly had a great life and career, and as those of us who
know him best will attest, he has never completely abandoned the little boy inside. I am honored to recognize the
2006 recipient of the Pediatric Urology Medal, Dr. A. Barry
Belman.
H. Gil Rushton, Jr.
Division of Urology
Childrens Hospital
Washington, D. C.
Hypospadias/Urethra
Traumatic Obliterative Urethral Strictures in Pediatric Patients:
Failure of the Cut to Light Technique at Long-Term Followup
Matthew K. Tollefson, Richard A. Ashley, Jonathan C. Routh and Douglas A. Husmann*,
From the Department of Urology, Mayo Medical School and Mayo Clinic, Rochester, Minnesota
Purpose: The cut to light procedure or antegrade and retrograde urethroscopy with endoscopic incision of obliterative
urethral strictures is an accepted method of treating posterior urethral distraction injuries. We report long-term followup on
22 pediatric patients treated with this technique.
Materials and Methods: The records of all pediatric patients with a posterior urethral distraction injury treated from 1986
through 2006 were reviewed. The cut to light procedure was performed for obliterative strictures less than 1 cm via a cold knife
or laser in 11 patients each. A urethral catheter was left in situ for 3 weeks. Patients were then followed a minimum of 2 years.
Results: A total of 22 patients at a median age of 11 years (range 3 to 16) were treated. Followup revealed stricture
recurrence in all patients. Median time to recurrence was 3 months after catheter removal (range 5 days to 1 year). Although
intermittent catheterization to maintain patency was attempted in all patients, the development of impassable urethral
strictures in 16 (73%) and complaints of chronic pain with catheterization in 4 (18%) led to end-to-end urethroplasty.
Successful reconstruction with open surgery was achieved in 19 of 20 patients (95%). Two patients (9%) maintained urethral
patency with intermittent catheterization.
Conclusions: In children with obliterative posterior urethral distraction injuries a cut to light procedure had a long-term
success rate of 0%. All patients required daily intermittent catheterization (9%) or formal urethral reconstruction (91%) to
maintain patency. We would strongly recommend against treating pediatric obliterative urethral strictures with a cut to light
procedure.
Key Words: urethral stricture, pelvis, wounds and injuries, endoscopy, reconstructive surgical procedures
istraction of the posterior urethra is a devastating

complication of pelvic trauma. It typically occurs following blunt trauma to the pelvis and it is estimated
to occur in approximately 10% of traumatic pelvic fractures.1
Initial treatment of these patients remains controversial
regarding primary realignment or suprapubic tube urinary
diversion with delayed urethral repair.2,3 Primary realignment can be much more difficult in pediatric patients since
pelvic bone mobility and small prostate size can create significant distraction injuries that are not amenable to initial
endoscopic management.3,4 Initial management with suprapubic tube diversion is a rapid and safe method to divert
urine in patients, who frequently have multiple coexisting
injuries that require expedient surgical management in an
attempt to save the life of the patient. The resultant urethral
stricture may be repaired in delayed fashion.
The management of obliterative posttraumatic posterior
urethral strictures in children is challenging. Often the location and length of the stricture, pelvic anatomy of the
pediatric patient and concomitant bladder neck injuries
makes primary urethral repair difficult or impossible.5 Due
to these anatomical differences even delayed urethral repair
has a higher failure rate in children than in adults.6 There-
fore, many of these patients are treated with a combined

transpubic perineal approach to achieve satisfactory end-toend urethroplasty. This combined approach has a high success rate in terms of urethral patency but it remains a
difficult procedure that can result in significant blood loss
and prolonged hospital courses.
In the late 1980s the cut to light procedure emerged as a
minimally invasive endourological method to decrease the
morbidity associated with complex urethral reconstructions.7 Briefly, this method involves simultaneous urethroscopy through a suprapubic tube tract as well as through the
urethra. Light is visualized through the obliterative stricture and the scar tissue is incised with a cold knife urethrotome or ablated with a laser. When urethral continuity is
restored, a urethral catheter is left in situ to allow re-epithelialization of the urethral mucosa. Initial experience with this
procedure was promising.8,9 However, long-term followup of
the procedure is lacking with minimal literature on pediatric
patients. Therefore, we analyzed our long-term management
with the cut to light procedure to treat obliterative posterior
urethral strictures less than 1 cm in pediatric patients.

* Correspondence and requests for reprints: Department of Urology, Mayo Clinic, 200 First St. Southwest, Rochester, Minnesota
55905(telephone:507-284-3711;FAX:507-284-4951;e-mail:husmann.
douglas@mayo.edu).
Financial interest and/or other relationship with Qmed.
0022-5347/07/1784-1656/0
Patient Selection
Upon approval from the institutional review board we identified all pediatric patients with a posterior urethral distraction injury from 1984 to 2004 who were treated with an endoscopic cut to light procedure. Patients selected for the procedure
1656
Vol. 178, 1656-1658, October 2007

Printed in U.S.A.
DOI:10.1016/j.juro.2007.03.182
TRAUMATIC OBLITERATIVE URETHRAL STRICTURES IN PEDIATRIC PATIENTS

had a posttraumatic posterior urethral stricture that obliterated
the urethral lumen. All strictures were less than 1 cm in total
length. Stricture length was estimated by simultaneous cystoscopy through a suprapubic tube tract and urethroscopy with radiographic confirmation. Additionally, as the stricture was incised, the length of scarred tissue was noted.
Procedure
Patients were taken to the operating room and placed in the
dorsolithotomy position. General anesthesia was induced.
Flexible urethroscopy was performed through the suprapubic tube site to the location of the urethral stricture. Simultaneous urethroscopy was performed through the urethra.
The stricture was then incised using a cold knife urethrotome or a KTP laser. A urethral catheter was placed intraoperatively and left in situ for 3 weeks. Patients were
treated on an outpatient basis and they were discharged
from the hospital on the day of the procedure.
Followup Protocol
All patients were followed at 3-month intervals or at patient
request due to symptom development. Followup consisted of
patient subjective assessment, uroflowmetry and ultrasound
of post-void residual urine for a minimum of 2 years following stricture ablation. Subsequent retrograde urethrography and cystourethroscopy were done when significant flow
rate abnormalities were found or if the patient had diurnal
symptoms of obstructive voiding.
RESULTS
A total of 22 male pediatric patients were treated endoscopically with the cut to light procedure for a less than 1 cm
posttraumatic obliterative posterior urethral stricture. The
median age of these patients was 11 years (range 3 to 16).
The cut to light procedure was performed using a cold knife
urethrotome in 11 patients, while 11 underwent stricture
elimination with a KTP laser. There were no intraoperative
complications, including rectal injury or secondary hemorrhage requiring hospital admission. The procedure was performed 3 to 4 months following the traumatic injury after
immediate urinary diversion with a suprapubic tube.
Followup revealed stricture recurrence in all 22 patients
treated. Median time to stricture recurrence was 3 months
following urethral catheter removal (range 5 days to 1 year).
Upon the development of obstructive voiding symptoms all
22 patients were initiated on daily intermittent catheterization. This was discontinued in 20 of the 22 patients (91%).
The reasons for intermittent catheterization discontinuation
were the development of impassable urethral strictures in
16 patients (73%) and complaints of chronic pain associated
with catheter passage in 4 (18%). Two patients (9%) maintained urethral patency with daily intermittent catheterization at a followup of 3 and 6 years, respectively.
The 20 patients in whom the cut to light procedure and
intermittent catheterization failed subsequently underwent
formal open end-to-end urethroplasty. This procedure was
successful for maintaining urethral patency in 19 of the 20
patients (95%). One patient required repeat end-to-end urethroplasty to maintain urethral continuity. No patient
treated with formal end-to-end urethroplasty required intermittent catheterization to prevent stricture recurrence.
1657
DISCUSSION
The management of posttraumatic posterior urethral disruptions remains a challenging and controversial issue in
the field of urological reconstruction. Posterior urethral disruptions are typically the result of severe pelvic fractures
sustained in motor vehicle accidents. Initial management of
these difficult cases is primary re-alignment with placement
of a urethral catheter or suprapubic tube diversion with
delayed repair of the subsequent obliterative urethral stricture. Primary re-alignment may be more challenging in
pediatric patients, leaving many patients presenting for delayed repair. In our experience the cut to light procedure was
universally associated with stricture recurrence and we recommend that these patients should undergo definitive endto-end urethroplasty.
Pediatric patients are unique since the prostate and
puboprostatic ligaments are not fully developed and the
bladder resides in a more intraperitoneal position compared
to that in an adult. Due to the mentioned findings the
location of traumatic posterior urethral disruption is less
predictable in pediatric patients with supraprostatic,
transprostatic and prostatomembranous urethral injuries
described.10 Additionally, the size of the pelvis, the mobility
of the pelvic bones and the increased risk of concomitant
traumatic bladder neck injuries in children can make primary realignment of the pediatric urethra challenging.6
The primary goal of the repair of obliterative urethral
strictures after complete posterior urethral disruption is
maintaining urethral continuity without the need for daily
intermittent catheterization. The open surgical approach to
these injuries can be difficult because the large hematoma
and urinary extravasation induced by the traumatic injury
may result in a significant inflammatory response. Historically the transpubic approach has been fraught with severe
complications, including a blood loss of up to 5,500 ml,9
lengthy hospital stay11 and recurrent strictures in up to 30%
of cases.12 However, advances in surgical exposure, anesthesia and experience with the procedure have significantly
decreased these complications.13 Important factors to consider when performing these procedures are complete excision of periurethral scar tissue, delicate tissue handling and
tension-free mucosa-to-mucosa reapproximation. Currently
patients undergoing transpubic or perineal urethroplasty
may expect a 3 to 5-day hospital course, a low rate of blood
transfusion and long-term success exceeding 90%.2,10,13,14
For these reasons open end-to-end urethroplasty may be
considered the gold standard reconstruction method.
The first direct vision internal urethrotomy was described in 1974 by Sachse.15 As experience and technique
with endourological management of stricture disease expanded, the surgical armamentarium available to clinicians
included the cold knife, electrocautery and Nd:YAG or KTP
lasers. In 1986 the cut to light procedure was first described7
and it emerged as a promising technique to manage these
cases endourologically.16 The procedure is typically performed on an outpatient basis and it has decreased the early
morbidity associated with patient treatment. Attempts are
made to replicate the excellent results associated with open
urethroplasty. Unfortunately early reports of the cut to light
technique were based on few patients with limited followup.
Subsequent studies demonstrated a high recurrence rate
1658
TRAUMATIC OBLITERATIVE URETHRAL STRICTURES IN PEDIATRIC PATIENTS
associated with the procedure and information on long-term

followup of this technique in children is lacking.
When treating young patients, the importance of performing a durable procedure is highlighted. These patients will
require urethral patency for several decades and long-term
followup of any new technique is necessary. Although the cut to
light procedure is attractive for decreasing morbidity, blood
loss and recovery time, we found that the long-term urethral
patency rates of this technique are disappointing. In our experience the cut to light procedure was unsuccessful for preventing stricture recurrence in all patients with formal end-to-end
urethroplasty (20 of 22 or 91%) or daily intermittent catheterization (2 of 22 or 9%) necessary to maintain urethral patency.
These results are consistent with reports in the adult population indicating a high stricture recurrence rate3 and the need
for intermittent catheterization in the majority of patients. Our
outcomes strongly suggest that an aggressive surgical approach to this problem provides the most durable and complication-free result.
CONCLUSIONS
The cut to light procedure had a long-term success rate of 0%
in our hands with all patients requiring formal end-to-end
urethroplasty (20 of 22 or 91%) or daily intermittent catheterization (2 of 22 or 9%) to maintain urethral continuity.
Because the success rate of open urethroplasty exceeds 90%,
we strongly recommend formal end-to-end urethroplasty in
pediatric patients with obliterative urethral strictures as
first line treatment to render them stricture-free.
REFERENCES
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
Koraitim MM: Pelvic fracture urethral injuries: the unresolved

controversy. J Urol 1999; 161: 1433.
Zhang J, Xu YM, Qiao Y, Jin SB, Wu DL and Gu BJ: An
evaluation of surgical approaches for posterior urethral distraction defects in boys. J Urol 2006; 176: 292.
Levine J and Wessells H: Comparison of open and endoscopic
treatment of posttraumatic posterior urethral strictures.
World J Surg 2001; 25: 1597.
Gundogdu H, Tanyel FC, Buyukpamukcu N and Hicsonmez A:
Primary realignment of posterior urethral ruptures in children. Br J Urol 1990; 65: 650.
Podesta ML: Use of the perineal and perineal-abdominal
(transpubic) approach for delayed management of pelvic
fracture urethral obliterative strictures in children: longterm outcome. J Urol 1998; 160: 160.
Flynn BJ, Delvecchio FC and Webster GD: Perineal repair of
pelvic fracture urethral distraction defects: experience in 120
patients during the last 10 years. J Urol 2003; 170: 1877.
Gupta NP and Gill IS: Core-through optical internal urethrotomy in management of impassable traumatic posterior
urethral strictures. J Urol 1986; 136: 1018.
Dogra PN and Nabi G: Nd-YAG laser core-through urethrotomy in obliterative posttraumatic urethral strictures in
children. Pediatr Surg Int 2003; 19: 652.
Lieberman SF and Barry JM: Retreat from transpubic urethroplasty for obliterated membranous urethral strictures.
J Urol 1982; 128: 379.
Onen A, Ozturk H, Kaya M and Otcu S: Long-term outcome of
posterior urethral rupture in boys: a comparison of different surgical modalities. Urology 2005; 65: 1202.
Webster GD and Sihelnik S: The management of strictures of
the membranous urethra. J Urol 1985; 134: 469.
12.
Netto NR Jr: The surgical repair of posterior urethral strictures by the transpubic urethroplasty or pull-through technique. J Urol 1985; 133: 411.
13. Tunc HM, Tefekli AH, Kaplancan T and Esen T: Delayed
repair of post-traumatic posterior urethral distraction injuries: long-term results. Urology 2000; 55: 837.
14. Pratap A, Agrawal CS, Tiwari A, Bhattarai BK, Pandit RK and
Anchal N: Complex posterior urethral disruptions: management by combined abdominal transpubic perineal urethroplasty. J Urol 2006; 175: 1751.
15. Sachse H: Treatment of urethral stricture: transurethral slit in
view using sharp section. Fortschr Med 1974; 92: 12.
16. el-Abd SA: Endoscopic treatment of posttraumatic urethral obliteration: experience in 396 patients. J Urol 1995; 153: 67.
EDITORIAL COMMENT
This article is a stake to the heart of the practice of attempting to cure posterior urethral distraction injury by so-called
cut to the light procedures, at least in children. Reconstructive urologists have always been suspicious of the cut the to
the light procedure because we have all seen it fail previously in our referred patients and in the rare times that we
use it ourselves it never seems to work. While the stunning
failure of cut to the light to cure a single stricture in these 22
patients is strong evidence that this procedure is a poor one,
doubters might suggest that cut to the light only failed
because the patients were children. Flynn et al reported that
adolescents at least have higher failure rates after open
surgical treatment of posterior urethral distraction injury
(reference 6 in article). However, I believe that the published
data, the results in my clinical practice and the clinical
results of the full-time reconstructive urologists with whom
I communicate indicate clearly that cut to the light seldom
works in adults or children. These patients should most
often be treated with open anastomotic urethroplasty, if
necessary after referral to a urologist with experience with
the open surgical repair of these difficult problems. Period.
Although it is a less important point, this article also casts
doubt on the practice of self-catheterization to try to keep
strictures open after urethrotomy. Of these patients 73% had
recurrence on a self-catheterization regimen and another 18%
had so much pain with catheter passage that they had to
abandon it. Again, few reconstructive urologists attempt this
maneuver since it always seems to fail, not to mention that it
unnecessarily condemns the patient to a lifetime of painful
self-catheterization, which would not be necessary after surgical cure by open urethroplasty. An excellent study by Greenwell et al suggests that self-catheterization has no value, at
least for anterior strictures,1 and the current authors add
doubt about its usefulness for posterior urethral distraction
injuries. This suboptimal management scheme remains wildly
popular, judging by the referral population seen at our clinic,
despite the real doubts as to its efficacy. It is another specter in
need of a stake to the heart, in my opinion.
Richard A. Santucci
Wayne State University School of Medicine
Detroit, Michigan
1.
Greenwell TJ, Castle C, Andrich DE, MacDonald JT, Nicol DL

and Mundy AR: Repeat urethrotomy and dilation for the
treatment of urethral stricture are neither clinically effective
nor cost-effective. J Urol 2004; 172: 275.
Long-Term Patient Reported Outcome of Urinary Symptoms

After Hypospadias Surgery: Norm Related Study in Adolescents
Kimihiko Moriya,* Hidehiro Kakizaki, Hiroshi Tanaka, Takahiko Mitsui, Tsuyoshi Furuno,
Takeya Kitta, Hiroshi Higashiyama, Hiroshi Sano and Katsuya Nonomura
From the Departments of Urology, Hokkaido University Graduate School of Medicine, Sapporo and Asahikawa Medical College (HK),
Asahikawa, Japan
Purpose: Although there are some reports of the functional outcome after hypospadias surgery, long-term patient reported
data on urinary symptoms are sparse. We evaluated the long-term outcome of urinary control in patients with hypospadias
who were operated in childhood.
Materials and Methods: A detailed questionnaire was mailed to 33 patients with hypospadias who were between 18 and
26 years old and had undergone the first operation at age 6 years or younger, and to 50 age matched controls.
Results: A total of 22 patients (66.7%) and 38 controls (76.0%) returned the questionnaire. Mean age was 20.6 years in the
hypospadias group and 21.0 years in the control group. Of the 22 patients 8 had mild and 11 had severe hypospadias, while
severity was unknown in 3. Straight direction of the urinary stream was noted at a similar rate in the hypospadias and
control groups (77% and 71%, respectively). Regarding the urinary stream, no subjects in either group reported it as bad or
very bad. Of the hypospadias group 95% of subjects and 89% of controls reported mild or no urinary frequency bother. The
incidence of individuals who always or mainly voided while standing was similar in the hypospadias and control groups (91%
and 84%, respectively). However, 32% of subjects in the hypospadias group reported moderate or severe terminal dribbling,
in contrast to only 3% of controls. When subjects with mild and severe hypospadias were compared, 4 of 11 (36%) with severe
hypospadias needed to squeeze the urethra after voiding to eliminate residual urine in the urethra, whereas none with mild
hypospadias needed to squeeze the urethra.
Conclusions: Although patients with hypospadias mostly have good urinary control, terminal dribbling is not uncommon,
especially in those with severe hypospadias.
Key Words: abnormalities, urethra, hypospadias, urination, questionnaire
ypospadias is one of the most common congenital

anomalies in male children with a reported incidence of 0.8/1,000 to 8.2/1,000 live male births.1 The
goal of HS surgery is to reconstruct the urethra to the tip of
the glans and straighten the penis. Successful HS surgery
ensures a cosmetic penile appearance, voiding while standing and unhampered sexual function in adulthood. Despite a
great surgical interest in short-term results and the incidence of problems such as fistula and stricture, there are
only a few studies describing the long-term sexual and cosmetic outcome of HS surgery compared with control subjects.25
Although there are some reports of uroflowmetry or urodynamic data after HS surgery,6 8 long-term patient reported data on urinary symptoms are sparse.9 14 According
to the previous literature spraying or sprinkling of the urinary stream and terminal dribbling were not uncommon in
those who underwent HS surgery.7,9 14
We evaluated the long-term outcome of HS surgery with
respect to urinary control. For this purpose patient reported
* Correspondence: Department of Urology, Hokkaido University

School of Medicine, Noth-15, West-7, Kita-Ku, Sapporo, 060-0824,
Japan (telephone: 81-11-716-1161, extension 5949; FAX: 81-11-7067853; e-mail: k-moriya@med.hokudai.ac.jp).
0022-5347/07/1784-1659/0
urinary symptoms in adolescents were compared with those

in age matched normal subjects.
Patients and Control Subjects
Of the 135 patients who were treated for HS at age 6 years
or younger at Hokkaido University Hospital between 1983
and 1991, 42 were 18 years or older and could be contacted
for this survey. These patients were contacted by telephone
and the purpose of this survey was explained. A questionnaire was then mailed to 33 patients who agreed to participate in this survey. Timing of the survey after the actual
completion of surgery was 11.3 to 20.9 years (mean 16.2).
Controls were 50 age matched, normal male volunteers
without a history of HS.
Instruments
As a part of a comprehensive survey investigating long-term
cosmetic, sexual and urinary outcome of HS surgery, the
questionnaire that was used consisted of 2 domains, including 1 about the cosmetic and sexual outcome, and the other
about urinary symptoms. The results of cosmetic and sexual
outcome have already been reported.5 The domain about
urinary symptoms included 5 items, such as urinary stream
direction, urinary stream, urinary frequency, voiding pos-
1659
Vol. 178, 1659-1662, October 2007

Printed in U.S.A.
DOI:10.1016/j.juro.2007.03.176
1660
PATIENT REPORTED OUTCOME OF URINARY SYMPTOMS AFTER HYPOSPADIAS SURGERY
ture and terminal dribbling. Additionally, their subjective

dissatisfaction with urination was also asked.
Data were analyzed using the Student t test, the chi-square
test with the Yates correction or the Fisher exact probability, as appropriate, p 0.05 considered statistically significant.
FIG. 2. Urinary stream
RESULTS
Response Rate to Questionnaire
Of the 33 patients 22 (66.7%) in the HS group and 38 of 50
controls (76.0%) returned the questionnaire. Mean SD age
when they returned the questionnaire was 20.6 2.1 years
in the patient group and 21.0 1.9 years in controls.
Of the 22 patients who returned the questionnaire the
severity of HS was classified as mild by 8, severe by 11 and
unknown by 3. As operative procedures, 1-stage urethroplasty with a parameatal foreskin flap15 was done in 16
cases, the Thiersch-Dupley technique was used in 1, the
Mathieu technique was used in 1, orthoplasty alone was
done in 1 and the procedure was unknown in 3. The mean
number of operations per patient was 1.6 (range 1 to 3). Age
at the first and final operations was 31 to 75 (mean 46) and
35 to 81 months (mean 53), respectively. Clinical characteristics in participants with HS did not differ significantly
from those in nonresponders with regard to age at first and
final operations, the number of operations and age at this
survey.
Outcome of Urinary Control
Straight direction of the urinary stream was noted at a
similar rate in the HS and control groups (77% and 71%,
respectively), and spraying of the stream was uncommon in
each group (fig. 1). Regarding the urinary stream, which was
rated as very good, good, fair, bad or very bad, no respondents in either group reported it as bad or very bad (fig. 2).
The degree of urinary frequency bother was rated as severe,
moderate, mild or none. Mild or no bother was reported by
95% of respondents in the HS group (mild in 50% and none
in 45%) and by 89% of controls (mild in 50% and none in
39%) (fig. 3). The incidence of individuals who always or
mainly voided while standing was similar in the 2 groups,
that is 90% in the HS group (always in 72% and mainly in
18%) and 84% in controls (always in 55% and mainly in 29%)
(fig. 4). However, 32% of respondents in the HS group reported moderate (9%) or severe (23%) terminal dribbling, in
contrast to only 3% and 0% of controls, respectively (fig. 5).
Thus, the incidence of individuals who reported no or mild
FIG. 1. Direction of urinary stream
terminal dribbling was significantly lower in the HS group

than in controls (overall 68%, including none in 32% and
mild in 36% vs overall 97%, including none in 52% and mild
in 45%, p 0.003, fig. 5). When compared between mild and
severe HS, 4 of 11 respondents (36%) with severe HS
squeezed the urethra after voiding to eliminate residual
urine in the urethra, whereas none with mild HS needed to
squeeze the urethra (fig. 6).
Subjective dissatisfaction with urination was reported by
3 patients in the HS group (13.6%), including 3 with terminal dribbling and 1 with urinary frequency, while 1 complained of the 2 symptoms. Two controls (5.3%) reported
subjective dissatisfaction with urination, including 1 with
terminal dribbling and 1 with difficult voiding. The rate of
dissatisfaction with urination was not significantly different
between the 2 groups and none of the participants in either
group wished to visit the hospital due to any urinary symptoms.
DISCUSSION
Long-term cosmetic, sexual and urinary outcome of HS surgery is critical for evaluating surgical techniques. Other
relevant issues in HS surgery are patient psychological maturation and infertility.16 Although several reports of longterm cosmetic and sexual outcomes have been published to
date,25 there are few reports of the long-term outcome of
urinary control comparing patients with HS with age
matched controls.3
Functional evaluation using uroflowmetry is universally
accepted to estimate the outcome of HS surgery because
uroflowmetry is noninvasive, available even for small children and accepted as a useful tool for detecting asymptomatic strictures.6-8 The results of uroflowmetry cannot be
biased by operator interpretation. However, it is well known
that the functional outcome does not always correlate with
urinary symptoms and patient satisfaction.17 Therefore,
subjective assessment judged by the patients themselves
should be included to estimate the long-term outcome of
FIG. 3. Urinary frequency bother
FIG. 4. Voiding posture
surgery, in addition to objective assessment using urodynamic study (uroflowmetry, etc) and surgeon appraisal of
the cosmetic results (meatal position, scar, chordee, etc).18
While various surgical techniques with excellent functional
outcomes have been reported, there is sparse literature on
patient urinary symptoms long after surgery. The reason for
this scarcity of long-term outcome analysis of urinary control would be the difficulty in contacting patients long after
surgery. In the current study we evaluated the long-term
outcome of urinary control after HS surgery in adolescents.
According to the previous literature urinary stream
spraying and terminal dribbling are not uncommon in patients with HS. About 33% to 66% of patients complained of
urinary stream spraying after surgery using various techniques.6,7,10,11 In the current study a straight direction of
the urinary stream was reported by 77% of patients with
HS, which was similar to the 71% in controls, and urinary
stream spraying was not common. The recent literature
shows that spraying is not a common complication compared
to reports in the 1980s and early 1990s.7,14 This presumptive decrease in the frequency of urinary stream spraying in
the recent era might be attributable to the improvement in
fine surgical instruments and refinement of surgical techniques, such as terminalization of the urethral meatus.10
Such advances seem to contribute to a better outcome of HS
surgery in its functional and cosmetic aspects.
Regarding the urinary stream, the degree of bother in
urinary frequency and voiding posture, there were no obvious differences between the 2 groups. However, the urinary
stream should be evaluated objectively using uroflowmetry
or other urodynamic studies to evaluate the validity and
efficacy of HS repair, although patient reported subjective
urinary symptoms were similar to those of controls. Unfortunately since patients in the current study were not followed regularly at the outpatient clinic at the time of this
survey, we had no data on uroflowmetry or another urodynamic study to compare symptoms and objective findings.
1661
FIG. 6. Terminal dribbling depending on HS severity
In contrast, an obvious difference was observed in the

incidence of individuals who reported moderate or severe
terminal dribbling between the patient group and controls.
Terminal dribbling is a relatively common problem after HS
surgery because there is no effective procedure currently
available to improve this problem, especially in cases with a
long, reconstructed urethra without adequate supporting
tissue. Since the HS abnormality is characterized by abnormal or deficient development of the corpus spongiosum distal to the meatus, the reconstructed urethra is not supported
by the corpus spongiosum and, therefore, the reconstructed
urethra does not contract at the end of micturition. While
basic principles of urethroplasty, such as not making too
wide a urethra and making a meatus large enough to avoid
reconstructed urethra dilatation, are important for decreasing terminal dribbling, some urinary stasis within these
relatively long urethral substitutions would be inevitable.12,14
Although a significant difference was observed in the rate
of terminal dribbling in the 2 groups, the rate of dissatisfaction with urination did not differ significantly and none of
the participants wished to visit the hospital due to any
urinary symptoms. From these results it seems that their
urinary symptoms can be coped with and do not bother them
severely.
CONCLUSIONS
Patients with HS mostly had good urinary control. While
urinary stream spraying seems to be improved by the advance in surgical techniques, terminal dribbling was still not
uncommon in patients with HS, especially in those with
severe HS.

HS hypospadias
REFERENCES
FIG. 5. Incidence of individuals who reported mild or no terminal

dribbling was significantly lower in HS group than in controls (68%
vs 97%, p 0.003).
1. Levitt SB and Reda EF. Hypospadias Pediatr Ann 1997; 17: 48.
2. Mureau MA, Slijper FM, Nijman RJ, van der Meulen JC,
Verhulst FC and Slob AK: Psychosexual adjustment of children and adolescents after different types of hypospadias
surgery: a norm-related study. J Urol 1995; 154: 1902.
3. Aho MO, Tammela OK, Somppi EM and Tammela TL: Sexual
and social life of men operated in childhood for hypospadias
and phimosis. A comparative study. Eur Urol 2000; 37: 95.
4. Bubanj TB, Perovic SV, Milicevic RM, Jovcic SB, Marjanovic
ZO and Djordjevic MM: Sexual behavior and sexual func-
1662
5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
15.
16.
17.
18.
tion of adults after hypospadias surgery: a comparative

study. J Urol 2004; 171: 1876.
Moriya K, Kakizaki H, Tanaka H, Furuno H, Higashiyama H,
Sano H et al: Long-term cosmetic and sexual outcome of
hypospadias surgery: norm-related study in adolescent.
J Urol 2006; 176: 1889.
Garibay JT, Reid C and Gonzalez R: Functional evaluation of
the results of hypospadias surgery with uroflowmetry.
J Urol 1995; 154: 835.
van der Werff JF, Boeve E, Brusse CA and van der Meulen JC:
Urodynamic evaluation of hypospadias repair. J Urol 1997;
157: 1344.
Hammouda HM, El-Ghoneimi A, Bagli DJ, McLorie GA and
Khoury AE: Tubularized incised plate repair: functional
outcome after intermediate followup. J Urol 2003; 169: 331.
Svensson J and Berg R: Micturition studies and sexual function in operated hypospadiacs. Br J Urol 1983; 55: 422.
Bracka A: A long-term view of hypospadias. Br J Plast Surg
1989; 42: 251.
Aho MO, Tammela OK and Tammela TL: Aspects of adult
satisfaction with the result of surgery for hypospadias performed in childhood. Eur Urol 1997; 32: 218.
van der Werff JF and Ultee J: Long-term follow-up of hypospadias repair. Br J Plast Surg 2000; 53: 588.
Nelson CP, Bloom DA, Kinast R, Wei JT and Park JM: Longterm patient reported outcome and satisfaction after oral
mucosa graft urethroplasty for hypospadias. J Urol 2005;
174: 1075.
Lam PN, Greenfield SP and Williot P: 2-stage repair in infancy
for severe hypospadias with chordee: long-term results after puberty. J Urol 2005; 174: 1567.
Koyanagi T, Nonomura K, Yamashita T, Kanagawa K and
Kakizaki H: One-stage repair of hypospadias: is there no
simple method universally applicable to all types of hypospadias? J Urol 1994; 152: 1232.
Snodgrass WT: Assessing outcomes of hypospadias surgery.
J Urol 2005; 174: 816.
Abrams P: In support of pressure-flow studies for evaluating
men with lower urinary tract symptoms. Urology 1994; 44:
153.
Mureau MA, Slijper FM, Slob AK, Verhulst FC and Nijman
RJ: Satisfaction with penile appearance after hypospadias
surgery: the patient and surgeon view. J Urol 1996; 155:
703.
EDITORIAL COMMENT
These authors evaluated long-term outcomes of urinary control in patients who underwent HS repair as a child and who
are now adults. Cosmetic appearance and sexual function
has been thoroughly studied in the literature, although
many studies have not evaluated urinary control. Thus, the
topic is relevant.
Urinary control was evaluated by a questionnaire that
was mailed to 33 patients who were treated for HS at age 6
years or younger and 50 age matched controls who had not
undergone penile surgery. To participate the subjects had to
be 18 years old or older. The questionnaire consisted of 5
items, including the urinary stream direction, urinary
stream, urinary frequency, voiding pressure and terminal
dribbling. The responses of men who had undergone HS
repair were compared to those of age matched controls. This
is a well designed study.
Only 22 of 33 patients and 38 of 50 controls answered the
questionnaire. Given the prevalence of HS, the sample size
is small. Although the authors included the type of repair
that the patients had undergone and the severity of HS, the
patients in each group were too few to substantiate further

analysis.
The most important result in this study is that subjectively men with a history of HS repair are satisfied with
urinary control compared with that of age matched controls.
Terminal dribbling was the only measure that was significantly worse in the HS group than in controls. However, this
did not translate into a clinically significant difference in
terms of overall bothersome symptoms.
A significant drawback was the uroflowmetry was not
measured. Uroflowmetry is an inexpensive test that can be
easily done at a routine office visit and can assess patients
for urethral strictures. In addition, a quantitative symptom
score, like the American Urological Association symptom
score, may have been a more useful measure of bothersome
symptoms. For these reasons it is unlikely that the results of
this study can be generalized to the population of patients
with a history of HS repair. However, it is encouraging that
patients are satisfied with urinary control several years
after initial repair. Future research evaluating long-term
urinary control and quality of life of patients with a history
of HS repair is important.
Terry W. Hensle
Childrens Hospital of New York
New York, New York
REPLY BY AUTHORS
Hypospadias is one of the most common congenital genital
anomalies in male children. However, long-term outcome
studies of hypospadias surgery in a large series are often
difficult because many patients relocate during the 2 decades and some refuse to participate in a study that addresses a sensitive issue. Another reason for not participating in our study was that the parents did not want their sons
to be reminded of the hypospadis, as the surgery had been
completed more than 10 years ago and its memory had
already been forgotten. Although our small study revealed
that most patients with hypospadias have good urinary control in the long term, we agree that further study of surgical
techniques, optimal age for surgery, impact on the degree of
hypospadias, etc is necessary in a larger series.
We agree that uroflowmetry is a useful tool to assess the
silent urethral stricture after hypospadias surgery. Although patients in this study became adults and were not
followed regularly at the time of the survey, uroflowmetry
was routinely performed at regular followup visits for several years after surgery. However, the data on uroflowmetry
were not sufficient and, thus, were not included in the
present study. On the other hand, while AUA symptom score
and International Prostate Symptom Score are validated
and widely used for the evaluation of urinary symptoms,
these questionnaires would not be enough to evaluate the
outcome of hypospadias surgery because of the lack of specific postoperative symptoms, such as spraying of the urinary stream or milking the urethra after micturition. Therefore, our original questionnaire was used in this study. Since
surgeon and patient satisfaction after hypospadias surgery
has been reported differently, a validated specific questionnaire for hypospadias surgery is warranted to evaluate outcomes from the patient point of view.
Split Prepuce In Situ Onlay

Hypospadias Repair: 17 Years of Experience
S. Sedberry-Ross, B. C. Stisser, C. G. Henderson, H. G. Rushton* and A. B. Belman
From the Childrens National Medical Center and George Washington University, Washington, D. C.
Purpose: We report our 17-year experience using split prepuce in situ onlay hypospadias repair, including long-term
followup of the first 100 patients initially reported on in 1998.
Materials and Methods: We identified 421 patients who underwent in situ onlay repair. Charts were retrospectively
reviewed to determine preoperative management, intraoperative details and complications.
Results: In situ onlay repair was used to repair glanular hypospadias in 22 cases (5.2%), coronal hypospadias in 184 (43.7%),
distal shaft hypospadias in 152 (36.1%), mid shaft hypospadias in 51 (12.1%), proximal shaft hypospadias in 7 (1.6%) and
hypospadias in the penoscrotal region in 5 (1.2%). Repair was successful with 1 procedure in 376 patients (89.4%), which
increased to 99.8% after a second procedure. Complications were defined as any problem that gave the surgeon or family
reason for concern. Functional complications requiring reoperation occurred in 45 patients (10.6%). Minor complications
requiring simple procedures or early postoperative evaluation occurred in 17 patients (4%). Concerns not requiring intervention occurred in 27 patients (6.4%). There were no urethral strictures. Three patients (0.7%) were lost to followup. Repair
is pending in 1 patient.
Conclusions: In situ onlay repair preserves the urethral plate and allows the formation of a well vascularized flap with
adequate tissue to completely cover the neourethra, resulting in a low rate of major complications. With longer followup,
inclusion of more mid shaft repairs and expansion to more proximal degrees of hypospadias our complication rates are higher
than previously reported but there have been no urethral strictures in 17 years of experience. Since complications present at
a median of 158 days (mean 570) after the procedure, long-term followup is indicated.
Key Words: hypospadias, urethra, abnormalities, surgical flaps, complications
ore than 200 techniques for hypospadias repair

have been described. In 1987 Elder et al reported a
modification of the pedicle flap for hypospadias repair, adapting the transverse preputial island flap as the
onlay island flap.1 This modification preserved the urethral
plate and avoided circular anastomosis, which is the most
common cause of fistula and stenosis. In 1998 Rushton and
Belman described a modification of the onlay island flap and
their initial experience with split prepuce in situ onlay hypospadias repair.2 In situ onlay repair was reportedly applicable in the mid to the distal shaft and in some cases of
proximal shaft hypospadias. In contrast to the transverse
preputial island flap, in situ onlay repair eliminated the
mobilization of a separate vascular pedicle to the inner preputial layer, thereby optimizing the blood supply to the
inner foreskin flap and simplifying repair. In the report of
the first 100 boys there were 5 complications, including
hematoma in 1 and urethrocutaneous fistula in 4. A good
cosmetic result was obtained in all cases. This report expands the experience to 17 years, including long-term followup of the initial 100 cases.
* Correspondence: Division of Pediatric Urology, Childrens National

Medical Center, 111 Michigan Ave., Northwest, Washington, D. C.
(telephone: 202-884-5042; FAX: 202-884-4739; e-mail: hrushton@
cnmc.org).
0022-5347/07/1784-1663/0
PATIENTS AND METHODS

Billing codes, operative logbooks and scheduling records
were reviewed. A total of 421 boys with hypospadias who
underwent in situ onlay repair were identified. Charts were
retrospectively reviewed in detail to examine preoperative
management, intraoperative details and postoperative care,
including complications, complaints or other information
concerning in situ onlay repair. Details of this technique
were previously described.2
For this report a complication was defined as any problem
that gave the surgeon or family reason for concern. Functional complications were defined as problems requiring surgical intervention to achieve normal urethral function or
improve the cosmetic appearance after in situ onlay repair.
Minor complications were defined as problems requiring
simple procedures or early postoperative evaluation. Concerns were those reported by the family that did not interfere with urethral function, produce an unacceptable cosmetic result or require intervention.
RESULTS
A total of 421 patients underwent in situ onlay repair between 1988 and 2005. Average patient age was 1.2 years
(median 7 months). Average followup was 229 days (median
32, range 2 to 4,779). The native meatus was located in the
glans in 22 patients (5.2%), at the corona in 184 (43.7%), in
the distal shaft in 152 (36.1%), in the mid shaft in 51
1663
Vol. 178, 1663-1667, October 2007

Printed in U.S.A.
DOI:10.1016/j.juro.2007.03.198
1664
SPLIT PREPUCE IN SITU ONLAY HYPOSPADIAS REPAIR
(12.1%), in the proximal shaft in 7 (1.6%) and in the penoscrotal region in 5 (1.2%). Urethroplasty measured length
was available in 136 patients. In these boys the mean measured length was 14 mm (median 13, range 9 to 30). Of the
boys 82 (19.4%) had mild to moderate chordee, which was
successfully corrected with dorsal plication or subcutaneous
tissue release. None of these boys required further surgical
intervention for chordee correction. A total of 36 patients
(8.5%) received parenteral testosterone preoperatively.
Hypospadias was successfully repaired with 1 procedure
in 376 patients (89.4%). This increased to 99.8% after a
second procedure. Functional complications requiring reoperation included UCF in 37 patients (8.7%), meatal stenosis
in 4 (0.9%), acquired megalourethra in 2 (0.4%), urethral
diverticulum in 3 (0.7%) and urethral disruption in 1 (0.2%).
The UCF rate was 4.5% after glanular repair, 8.1% after
coronal repair, 6.6% after distal shaft repair, 17.6% after
mid shaft repair, 28.5% after proximal shaft repair and 40%
after penoscrotal repair (see table). One patient with a distal
UCF was treated in the office with excision of a thin meatal
web, which was categorized as a minor complication. Another patient diagnosed with a UCF had resolution of that
fistula. One patient with a UCF awaits repair and 1 with a
UCF was lost to followup. These 2 patients were included in
the functional complication category. A total of 35 fistulas
(8.3%) were repaired successfully at 1 operation using a
de-epithelialized or full-thickness skin advancement flap in
all except 1.3 One patient (0.2%) required a second procedure
for fistula repair.
After repair megalourethra developed in 2 patients
(0.4%) and acquired urethral diverticula developed in 3
(0.7%). In 1 of these cases an unrecognized megalourethra
was identified during fistula repair. Complete urethroplasty
was performed to correct the 2 defects. The second acquired
megalourethra presented as post-void urethral dilatation.
Urethroplasty was performed without further complications. In 2 patients urethral diverticula were identified during fistula repair. Diverticulectomy was performed in each
case. The other patient presented on 2 occasions with bac-
teriuria. Voiding cystourethrogram revealed a urethral diverticulum, which was excised.

Urethral disruption occurred in 5 patients (1.2%), although 2 with minimal partial glanular urethral disruption
did not require further surgery. Complete disruption of repair occurred in 1 boy. He underwent successful onlay urethroplasty 3 years after initial repair. Another boy had urethral disruption due to catheter trauma. The family was
instructed to present for followup after healing for reevaluation. However, the patient moved out of the country and
was lost to followup. Similarly 1 boy with mild urethral
disruption failed to return for followup.
Meatal stenosis, defined subjectively as a narrowed meatus and a fine caliber, narrow, forceful stream with a characteristic upward deviation, was diagnosed in 11 patients.
Four patients (0.8%) had significant meatal stenosis that did
not respond to meatal dilations. In 1 case meatotomy was
performed in conjunction with UCF repair. Meatotomy was
performed in the other 3 patients without further complications. Other patients in whom the meatus appeared small
postoperatively underwent calibration with 8Fr and 10Fr
sounds. If they passed easily, meatal stenosis was excluded.
Minor complications included mild meatal stenosis in 7
patients (1.7%), who were treated with meatal dilation and
subsequently had resolution of symptoms. Seven inclusion
cysts (1.7%) were excised and 2 hematomas required drainage. One patient had obstructive voiding symptoms suspicious for urethral stricture. This patient underwent cystoscopy, which revealed a normal caliber urethra without
evidence of stricture. Another patient was rehospitalized for
pneumomediastinum, which was treated conservatively and
resolved.
Concerns included 3 patients (0.7%) with small inclusion
cysts that resolved. Three patients (0.7%) had mild penile
torque. Suture tracts were reported in 5 patients (1.1%).
Four patients (0.8%) with small hematomas were evaluated.
Each case resolved without intervention. Skin bridges/tags
were present in 2 patients (0.4%) and minor bleeding was
reported in 3 (0.7%). Two patients (0.4%) were evaluated for
All complications by native meatal location

No. Shaft (%)
Complication
UCF
Meatal stenosis
Megalourethra
Urethra:
Diverticula
Stricture
Disruption
Inclusion cyst
Penile torque
Suture track
Bleeding
Hematoma
Skin bridge/tag
Meatal web/flap
Penile adherence
Infection
Other
Totals
No. Glans (%)
No. Coronal (%)
Distal
Mid
Proximal
No. Penoscrotal (%)
No. Total (%)
22 (5.2)
1 (4.5)
184 (43.7)
15 (8.1)
5 (2.7)
152 (36.1)
10 (6.6)
3 (2.0)
51 (12.1)
9 (17.6)
2 (3.9)
1 (1.9)
7 (1.6)
2 (28.5)
1 (14)
5 (1.2)
2 (40)
421
39 (9.2)
11 (2.6)
2 (0.4)
1 (4.5)
2
(9)
* No evidence of stricture on cystoscopy.

Pneumomediastinum.
3
3
3
2
1
4
1
1
1
1
1
*
(1.6)
(1.6)
(1.6)
(1.1)
(0.5)
(2.1)
(0.5)
(0.5)
(0.5)
(0.5)
(0.5)
41 (22.2)
1 (0.6)
1 (1.9)
6 (3.9)
2 (3.9)
1 (1.9)
3 (2.0)
1 (0.7)
2 (1.3)
1 (20)
1
(14)
3
0
5
10
3
5
3
6
2
2
1
2
1
1 (1.9)
1 (1.9)
1 (0.7)
27 (17.8)
18
(35)
(57)
3 (60)
(0.7)
(1.2)
(2.4)
(0.7)
(1.1)
(0.7)
(1.4)
(0.4)
(0.4)
(0.2)
(0.2)
(0.2)
95 (22.5)

meatal web/flaps. One child had penile adherence and 2
(0.4%) were treated for a minor infection.
Using our broad definition of complication the all inclusive complication rate for in situ onlay repair was 22.5% (see
table). Complications occurred in 9% of boys with glanular
hypospadias, in 22.2% with coronal hypospadias, in 17.8%
with mid shaft hypospadias, in 35% with distal shaft hypospadias, in 57% with proximal hypospadias and in 60% with
penoscrotal hypospadias. However, the reoperative rate was
28 of 358 cases (7.8%) for glanular, coronal and distal shaft
repairs, 12 of 5 (23%) for mid shaft repairs and 4 of 12 (33%)
for proximal/penoscrotal repairs. Of note, 2 of the 12 boys
with severe hypospadias had more than 1 abnormality, increasing the overall functional complication rate in this
group. In 1 boy meatal stenosis was initially treated with
dilation and a UCF subsequently developed. During repair a
urethral diverticulum was discovered and repaired. Similarly a boy with penoscrotal hypospadias had a UCF and
intraoperatively a megalourethra was found and repaired.
The other 2 complications in boys with a proximally located
meatus were isolated UCF, which were repaired successfully.
When reviewing the first 100 cases, average followup was
492 days (median 44, range 3 to 4,364). There were 23
complications (23%). The most common complication was
urethrocutaneous fistula (10 or 10%). The 2 acquired megalourethras reported in this series occurred in the first 100
patients. Meatal stenosis occurred in 2 cases (2%), of which
each required surgical meatotomy. The overall reoperative
rate for functional complications was 12%. The average time
from initial operation to complication presentation was 570
days (median 158).
DISCUSSION
The goal of hypospadias surgery is to provide a satisfactory
cosmetic result and a functional urethra without long-term
complications. In situ onlay repair adheres to the general
principles that result in successful hypospadias repair. By
preserving the entire vascular pedicle of the foreskin the
blood supply of the inner preputial flap used for the neourethra is optimized.2 Additionally, in situ onlay repair not only
offers tissue to form the urethra, but also provides adequate
redundant pedicle that can be used to cover the neourethra
and prevent overlapping suture lines, which theoretically
lowers the incidence of UCF formation. Finally, the urethral
plate remains intact. According to Erol et al, the urethral
plate is a well vascularized structure with a rich nerve
supply, and an extensive muscular and connective tissue
backing.4 Due to these features preservation of the urethral
plate was suggested to decrease complication rates.4 These
principles coupled with meticulous technique and delicate
tissue handling offer the opportunity for excellent functional
and cosmetic outcomes.
Cook et al reported that most pediatric urologists use the
TIP (Snodgrass) procedure for distal and mid shaft hypospadias repairs, while applying the TIP, transverse island
flap onlay or the tubularized island flap for repairing more
proximal degrees of hypospadias.5 The choice is dictated by
parameters such as urethral plate width or depth, the
amount of skin available and the presence or absence of
chordee. For example, Holland et al concluded that patients
with a shallow urethral plate are more likely to have a
1665
neourethra caliber of 6Fr or less and, thus, a higher incidence of meatal stenosis if the TIP technique is used for
hypospadias repair.6 They also reported that patients with a
urethral plate less than 8 mm wide were significantly more
likely to have a urethral fistula. However, other groups
found no difference in outcome when comparing urethral
plate characteristics or urethral plate width in boys undergoing TIP repair.7
Except severe chordee requiring transection of the urethral plate, in situ onlay repair is applicable in virtually all
cases of mid shaft and more distal hypospadias, including
boys with mild to moderate chordee that may be corrected
with the release of skin and subcutaneous tissue, and/or
dorsal plication, regardless of urethral plate width or depth.
An increase in the number of mid shaft hypospadias repairs
coupled with the application of this technique to more severe
forms of hypospadias, longer followup and our liberal, all
inclusive definition of complication resulted in an overall
complication rate of 22.5%. However, functional complication rate and reoperative rates were low.
Our overall fistula rate of 9.2% was higher than initially
reported. This was attributable not only to the inclusion of
more severe forms of hypospadias, but also to the late presentation of small fistulas that became clinically apparent
only after toilet training. The incidence of fistula in this
series is similar to that of other repairs. Fistula rates for TIP
repair vary from 0% to 16%.8 10 Fistula rates for transverse
island onlay repair are also highly variable. Some investigators reported rates as high as 17% when used to repair
proximal hypospadias, while others reported no UCF.11,12 In
our series the highest incidence of UCF occurred in patients
with proximal or penoscrotal hypospadias, suggesting that
this technique is best reserved for mid shaft or more distal
hypospadias. However, except in 1 patient, all fistulas were
repaired successfully at 1 operation. There were no additional complications in the patients after UCF repair.
Acquired megalourethra and urethral diverticula are
rare complications of hypospadias surgery and they occurred
in less than 1% of our patients. Some series show no development of acquired urethral diverticula or megalourethra
with transverse onlay repair.11,12 Amukele et al reported
urethral diverticula in 2 patients (0.7%) undergoing
Thiersch-Duplay repair.13 We suspect that these complications may be underreported because in 3 of our 5 patients
these complications were diagnosed incidentally during surgery to correct a UCF.
In the literature the incidence of meatal stenosis following hypospadias repair varies. Cheng reported a 1% incidence of meatal stenosis in a multi-institutional study of TIP
repair.9 Wiener et al reported a 2% incidence of meatal
stenosis with transverse island onlay repair.11 In our series
the incidence of meatal stenosis was 2.6%. However, only 4
patients (0.8%) had significant meatal stenosis that required
surgical meatotomy for definitive treatment. All other patients were treated successfully with outpatient meatal dilation. We suspect that the incidence of meatal stenosis may
have been higher in other reported series if they had used
our liberal definition of meatal stenosis.
There were 5 cases (1.2%) of urethral disruption, of which
2 were mild and did not need repair. Three patients had
more significant disruption. One child had a large fat pad,
which resulted in a retracted penis that interfered with
normal healing. This child underwent successful repair at a
1666
later date. The second child sustained catheter trauma,

which resulted in bleeding and subsequent retraction of the
meatus to a glanulocoronal location. Repair was offered but
the parent refused and the child was subsequently lost to
followup. The other patient was also lost to followup. Interestingly in many series there are no reported cases of failed
repair or of dehiscence. However, Snodgrass and Lorenzo
reported 1 boy who underwent repair dehiscence in a series
of 33 boys (3.0%) undergoing proximal hypospadias repair
with the TIP procedure.14 Amukele et al reported a 0.7%
incidence of repair dehiscence using the Thiersch-Duplay
technique.13
Overall reoperative rates compared favorably to those of
other series. Reoperative rates for distal repairs were 7.26%
for glanular, coronal and distal shaft repair, 23% for mid
shaft repair and 33% for proximal/penoscrotal repair. The
complication rate for the TIP technique for mid shaft and/or
proximal hypospadias is 17.5% to 33%.14,15 More recently
Snodgrass and Yucel reported an initial complication rate of
53% for proximal repair, which improved to 25% after modifications of their technique.16
Perhaps the most important finding in our series of 421
patients was a 0% incidence of urethral strictures with in
situ onlay repair. Jayanthi8 and Patel et al12 also reported
no evidence of stricture with the modified Snodgrass repair
and with island onlay repair, respectively, in cohorts similar
to those in the current series. Snodgrass and Lorenzo reported 1 urethral stricture in a series of 33 boys undergoing
TIP repair for proximal hypospadias.14 Several reports have
documented abnormal flow curves with obstructive patterns
after TIP repair even in the absence of an actual stricture. It
was proposed that the obstructed flow pattern may be secondary to a stiff urethra with low compliance.1719 Urethral
strictures were reported with the Thiersch-Duplay technique (1.1%) and with onlay island flap repair (2.63%).13,20
Of the potential functional complications urethral stricture
is the most complex and difficult to repair. Therefore, repairs
with a 0% stricture rate are appealing.
The main advantage to split prepuce in situ onlay hypospadias repair is the avoidance of mobilizing a separate
vascular pedicle to the inner preputial layer. Eliminating
this step maximizes blood supply to the flap used to form the
neourethra and provides additional well vascularized, redundant, de-epithelialized tissue. This technique maintains
virtually the entire blood supply of the half of the foreskin
used to create the inner foreskin flap, thereby significantly
decreasing the risk of devascularization. Furthermore, coverage by the redundant pedicle tissue provides further vascular supply and protects against fistula formation and
metal stenosis.2 The technique is simple and it results in
excellent cosmetic outcomes.
Most reviews tend to provide short-term results with
little reference to long-term followup. That was true in the
first report of in situ onlay repair. However, when re-reviewing the initial 100 patients, which allowed the longest period
for potential complications to present, we found that the
average time between operation and complication presentation was 570 days (median 158). This is attributable to the
late presentation of small fistulas that became clinically
apparent only after toilet training and it indicates that patients need long-term followup after hypospadias surgery.
The limitations of this review of in situ onlay repair
include the retrospective nature of the study. Additionally,
as with most series, long-term followup is not available for

all patients. Therefore, the complication rates may actually
be higher than reported. There also is the possibility that
patients sought care elsewhere if a complication occurred,
thus, creating a second avenue for potential underreporting.
We attempted to objectively and critically review in situ
onlay repair, and so we chose to report all problems observed. However, the incidence of functional complications
and the overall reoperative rate are low and similar to those
in other series. This procedure is most applicable to mid
shaft and more distal hypospadias, and it can be considered
for the more proximally located native meatus. In situ onlay
repair may be used in cases of mild to moderate chordee
regardless of urethral plate width and depth, and it is successful in 89.4% after 1 operation. Success of the procedure
increases to 99.8% after a second operation. Most importantly there are no reports of urethral strictures in any
patient in 17 years of performing this repair and cosmetic
results have been excellent.
Duckett wrote, There is no one good method for all hypospadias repairs . . . the urologist ought to be proficient in
performing a number of procedures to be prepared for all
possible eventualities.21 In situ onlay repair is a simple
technique that should be considered as one of the repairs
that are available in the armamentarium of the hypospadiologist.

TIP tubularized incised urethral plate
UCF urethrocutaneous fistula
REFERENCES
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
Elder JS, Duckett JW and Synder HM: Onlay island flap in the
repair of mid and distal penile hypospadias without chordee. J Urol 1987; 138: 377.
Rushton HG and Belman AB: The split prepuce in situ onlay
hypospadias repair. J Urol 1998; 160: 1134.
Santangelo K, Rushton HG and Belman AB: Outcome analysis
of simple and complex urethrocutaneous fistula closure using a de-epithelialized or full-thickness skin advancement
flap for coverage. J Urol 2003; 170: 1589.
Erol A, Baskin LS, Li YW and Liu WH: Anatomical studies of
the urethral plate: why preservation of the urethral plate is
important in hypospadias repair. BJU Int 2000; 85: 728.
Cook A, Khoury AE, Neville C, Boyli DJ, Farhat WA and Pippi
Salle JJ: A multicenter evaluation of technical preferences
for primary hypospadias repair. J Urol 2005; 174: 2354.
Holland AJA and Smith GHH: Effect of the depth and width of
the urethral plate on tubularized incised plate urethroplasty. J Urol 2000; 164: 489.
Nguyen M and Snodgrass W: Effect of urethral plate characteristics on tubularized incised plate urethroplasty. J Urol
2004; 171: 1260.
Jayanthi V: The modified Snodgrass hypospadias repair: reducing the risk of fistula and meatal stenosis. J Urol 2003;
170: 1603.
Cheng EY: Snodgrass hypospadias repair with vascularized
dartos flap: the perfect repair for virgin cases of hypospadias? J Urol 2002; 168: 1723.
Guralnick ML, al-Shammari A, Williot PE and Leonard MP:
Outcome of hypospadias repair using the tubularized incised plate urethroplasty. Can J Urol 2000; 7: 986.

11.
12.
13.
14.
15.
16.
Weiner J, Sutherland R and Roth D: Comparison of onlay and

tubularized island flaps of inner preputial skin for the
repair of proximal hypospadias. J Urol 1997; 158: 1172.
Patel R, Aseem R and Synder HM: The island tube and island
onlay hypospadias repairs offer excellent long-term outcomes: 14 year followup. J Urol 2004; 172: 1717.
Amukele S, Weiser A and Jeffrey A: Results of 265 consecutive
proximal hypospadias repairs using the Thiersch-Duplay
principal. J Urol 2004; 172: 2382.
Snodgrass WT and Lorenzo A: Tubularized incised-plate urethroplasty for proximal hypospadias. BJU Int 2002; 89: 90.
Chen SC, Yang SS, Hsieh CH and Chen YT: Tubularized
incised plate urethroplasty for proximal hypospadias. BJU
Int 2000; 86: 1050.
Snodgrass W and Yucel S: Tubularized incised plate for midshaft
and proximal hypospadias repair. J Urol 2007; 177: 698.
17.
1667
Marte A, Di Iorio G, De Pasquale M and Cotrufo AM: Functional evaluation of tubularized-incised plate repair of midshaft-proximal hypospadias using uroflowmetry. BJU Int
2001; 87: 540.
18. Holmdahl G, Karstrom L and Abrahamsson K: Hypospadias
repair with tubularized incised plate. Is uroflowmetry necessary postoperatively? J Pediatr Urol 2006; 2: 304.
19. Wolffenbuttel KP, Wondergem N and Hoefnagels JJ: Abnormal urine flow in boys with distal hypospadias before and
after correction. J Urol 2006; 176: 1733.
20. Castanon M, Munoz E and Carrasco R: Treatment of proximal
hypospadias with a tubularized island flap urethroplasty
and the onlay technique: a comparative study. J Pediatr
Surg 2000; 35: 1453.
21. Duckett JW: The island: flap technique for hypospadias repair.
J Urol 2002; 167: 2148.
Trauma/Imaging
The Posterior Sagittal Approach
for Recurrent Genitourinary Pathology
Shumyle Alam, Marc A. Levitt, Curtis A. Sheldon and Alberto Pea
From the Departments of Pediatric Surgery (Division of Pediatric Urology), Colorectal Center for Children (MAL, AP), Cincinnati
Childrens Medical Center, Cincinnati, Ohio
Purpose: The posterior sagittal transanorectal approach offers the surgeon excellent exposure to the lower genitourinary system.
The approach has been successfully used for revisional surgery involving the urinary tract and rectum. We have used this approach
to treat mllerian remnants, acquired or congenital fistulas and acquired urethral strictures. In the setting of revisional surgery
abdominal or perineal approaches may not provide the same level of exposure to the lower urinary tract.
Materials and Methods: Of 911 males who underwent a posterior sagittal approach 12 were identified in whom this
approach was used to manage lower urinary tract problems without associated anorectal malformation (group 1). An
additional 58 patients with associated anorectal malformations were included in analysis (group 2).
Results: All cases were approached with a posterior sagittal incision to facilitate exposure after failed prior surgical attempts
to treat traumatic strictures of the proximal urethra in 4 and mllerian remnants in 8 (group 1). In group 2 there were 19
urethral diverticula, 8 strictures and 31 persistent rectourethral fistulas. Followup was 6 months to 15 years. There were no
recurrences identified in group 1. In group 2 there were 5 postoperative urethrocutaneous fistulas. Data on potency were
available on 4 patients in group 1 and 14 in group 2. All 14 patients in group 2 reported intact potency after surgery. One
patient in group 1 was impotent before and after surgery.
Conclusions: The posterior sagittal approach is helpful for addressing lower urinary tract pathology when exposure is
inadequate from a perineal or abdominal approach. We propose that better exposure with this approach may limit inadvertent injury to nerves and result in more complete treatment of the presenting pathological condition. This approach provides
an alternative that warrants careful consideration when dealing with the lower genitourinary tract in pediatric patients,
especially in the setting of revisional surgery.
Key Words: abnormalities, reconstructive surgical procedures, urinary tract
he posterior sagittal approach provides excellent exposure to the structures of the perineum and anorectal
canal, and it was originally described for treating anorectal and cloacal malformations.1,2 It allows visualization of
the anatomy of the perineum, especially for pathological conditions located in the space between the concavity of the sacrum and the posterior pubis.3 For the genitourinary surgeon
this would provide an alternate way to access the posterior
urethra, bladder neck and seminal vesicles, especially in settings in which prior surgical treatment had failed.
Access to the structures of the pelvis has been performed
successfully with a transabdominal, retropubic, retroperitoneal or perineal approach. However, exposure can be lacking,
especially in the setting of revisional surgery. Scar can obliterate tissue planes and even with meticulous dissection the delicate structures of the perineum can be damaged in this setting. Laparoscopy also has an important role in addressing this
pathological conditions but the technique becomes limited in
the setting of revisional surgery.
In the adult literature, the York-Mason approach has been
used to address acquired rectourethral fistulas with good success.4,5 The posterior sagittal approach with splitting of the
rectum (transanorectal) has also been applied to children with
urogenital sinus anomalies in the setting of a normal anus.6 It
0022-5347/07/1784-1668/0
seems logical that a posterior approach could also benefit pediatric patients by addressing symptomatic mllerian remnants, fistulas, seminal vesicle cysts or even proximal urethral
strictures that are located too high to allow comfortable access
via a perineal approach and too low to reach from the abdomen.
The anterior sagittal transrectal approach has also been
used by urologists in the past to address lower urinary tract
pathology. This approach is a modification of the posterior
sagittal approach.7 The difference between the 2 approaches is
patient positioning and the avoidance of splitting the posterior
rectal wall. However, in the setting of revisional surgery, ie
after perineal surgery the posterior sagittal approach allows
one to access the perineum without going through prior scar.
We have successfully used the posterior sagittal approach
for acquired genitourinary pathology in the setting of associated anorectal malformations.8 This includes acquired urethral atresia, posterior urethral diverticula and persistent fistulas (fig. 1). Success in managing urethral pathology in these
patients prompted us to use the approach in patients without
associated anorectal malformations in the setting of revisional
surgery.

A total of 911 males underwent a posterior sagittal approach
from 1981 to 2006. Of these patients 12 were retrospectively
1668
Vol. 178, 1668-1671, October 2007

Printed in U.S.A.
DOI:10.1016/j.juro.2007.03.189
POSTERIOR SAGITTAL APPROACH FOR RECURRENT GENITOURINARY PATHOLOGY
FIG. 1. Child with anorectal malformation and urethral stricture.

Child is prone and rectum was mobilized so that it is out of field.
Proximal urethra is easily visualized.
identified as having undergone posterior sagittal transanorectal approaches for genitourinary pathology without associated anorectal malformations (group 1). These 12 patients
underwent surgery performed by one of us (AP). An additional 58 patients were identified for comparison with urinary tract pathology as well as associated anorectal malformations (group 2) All patients underwent revisional surgery
using a posterior sagittal approach. The patients were assigned to different groups to compare outcomes after performing a posterior sagittal approach in the setting of revisional surgery in those with and without an associated
ARM. Institutional review board approved retrospective
analysis was used.
All patients were male and 2 to 43 years old at surgery.
Followup was 6 months to 15 years. Females were excluded from analysis. We have used the transanorectal
approach in men with acquired fistulas after prostate
cancer treatment but they were not included in analysis.
Patients referred without a prior attempt at repair regardless of associated genitourinary pathology were also
excluded from analysis.
pose the posterior urethra. These patients had also undergone a prior attempt at repair. In the patients without
associated anorectal malformations (group 1) there were 4
traumatic strictures of the proximal urethra and 8 mllerian remnants (table 1). To date there have been no
recurrences in this group. Two patients in the series were
lost to followup. All patients were continent at a minimum
of 6 months of followup. Potency data were known on 4
patients with 3 (75%) reporting intact erections and ejaculation. One patient who had a stricture following a crush
injury was impotent before and after surgery (table 2).
There was 1 documented case of paternity in a patient
with a large seminal vesicle cyst.
There were 3 patients in whom urethral strictures were
managed with cystostomy tubes and diverting colostomies
before referral. One patient in this group was treated with a
suprapubic tube alone and underwent repair without fecal
diversion. Only 1 of the patients with posterior urethral
pathology underwent diversion with colostomy. No patient
in this series was in urinary retention postoperatively.
The 4 patients with urethral strictures were unique because 3
(75%) underwent diversion with colostomy before referral. These
patients were victims of blunt trauma with pelvic fractures and
they received diversion because of associated injuries. All 4 patients underwent a prior attempt at stricture repair, in addition to
surgery related to the trauma.
The patients with mllerian remnants were also unique.
They were referred not for simple excision of the pathological conditions, but rather because prior attempts at excision
had failed and the patients were symptomatic with recurrent urinary tract infections and orchiepididymitis. Only 1
patient in this group underwent diversion before surgery.
Diversion was performed because in this patient several
prior attempts at repair had failed and he was considerably
older at operation (40 years).
Surgical success in group 1 was defined as treatment of
the primary pathological conditions without evidence of recurrence. To our knowledge no patients have required urinary diversion or further procedures. Followup studies, such
as formal urodynamics or noninvasive uroflowmetry, were
unfortunately not readily available at the time of this chart
review. Institutional review board approval is currently being sought to contact patients to determine long-term outcomes after repair.
In the remaining 58 patients with ARM (group 2) indications for surgery were urethral diverticula in 19, acquired
strictures in 8 and persistent rectourethral fistulas in 31
(table 1). One patient with a persistent rectourethral stricture had what was thought to be a low anorectal malformation and he was treated with perineal anoplasty. He also had
a delayed diagnosis of posterior urethral valves. It was not
until after transurethral incision of the valve leaflets that an
unrecognized rectourethral fistula became apparent.
TABLE 1
RESULTS
All patients in group 1 underwent a midline posterior
sagittal transanorectal approach after failed abdominal or
perineal surgery. Those in group 2 underwent a posterior
sagittal approach with mobilization of the rectum to ex-
1669
Stricture
Posterior urethral diverticulum
Persistent fistula
Totals
No.
Group 1
No.
Group 2
4
8
0
8
19
31
12
58
1670

TABLE 2
Group
Continent urine
Continent stool
Potent with/without
ejaculation
Urethrocutaneous
fistula
Lost to followup
No. Group 1 (%)
No. Group 2 (%)
10 (100)
10 (100)
3/4 (75)
38
(74.5)
41
(80)
14/14 (100)
A total of 38 patients (74.5%) were continent of urine in

group 2 after surgery. Residual urethral pathology did not
influence the preservation of continence. Rather, the number of failed prior attempts at repair or associated neurological anomalies were more predictive in this group, as described. No patient in this group was incontinent as a result
of surgery.
There was postoperative urinary retention in 1 patient with
recurrent urethral stricture, which was managed by a suprapubic tube. Overall there were 5 failures (8%) in group 2.
In group 2, 14 of 14 patients (100%) reported erections
sufficient for penetration, of whom 8 reported successful
ejaculation. These patients represent those in whom these
data were known. There were no reports of impotence in
group 2. There were 2 documented cases of paternity among
these patients. Eight patients were lost to followup (table 2).
No patients reported loss of potency after repair.
DISCUSSION
Our intent is to expose the pediatric genitourinary surgeon
to another option for addressing posterior urethral and perineal pathology, especially after the more traditional approaches have failed. We described the posterior sagittal
approach for addressing urethral pathology in the past as a
primary means of surgery. The novelty of this series of
patients is to suggest that the posterior sagittal approach
can be used safely even after multiple failed prior attempts
at repair. It was demonstrated in the past that the posterior
sagittal transanorectal approach does not lead to an increased rate of fecal incontinence, stenosis or infection in
children with or without anorectal malformation.1,8,9 This
aspect makes the approach for patients without ARM a
viable alternative for consideration. The success that we
have had using the transanorectal approach in females with
a urogenital sinus anomaly and normal anus serves to further this argument.10
The concept of splitting the rectum is not the initial
choice of most urologists. We believe that the exposure afforded by the posterior sagittal transanorectal approach and
the minimal morbidity noted by splitting the rectum justifies
this approach.10 Exposure afforded by splitting the rectum is
unparalleled via any other approach (fig. 2). The rectum in
ARM reoperations is usually moved out of the way to care for
an associated mislocated anus, stricture or prolapse, which
allows excellent exposure of the posterior urethra. In patients with a normal anus the anus and rectum are split,
preserving its function after reconstruction.
We used a protective colostomy in the majority of revisional surgeries for anorectal malformation because of the
need for rectal tapering, extensive mobilization or opposing
suture lines. Four patients in group 1 received a protective
colostomy. There was no increased complication rate in

cases that were not diverted because the rectum was only
split and then precisely reconstructed to help with exposure.
Colostomy can be safely avoided in patients with a normal rectum when the bowel is adequately prepared. We use
standard 1 day bowel preparation before surgery. A peripherally inserted central catheter is placed preoperatively in
anticipation of total parental nutrition. Postoperatively the
patient receives nothing by mouth for 7 to 10 days. We have
not observed wound infections or breakdown of repair without diversion and we do not believe that diverting colostomy
is necessary when these guidelines are followed. In the case
of a mislocated rectum with associated ARM colostomy can
also be avoided, provided that there is no need for tapering
or extensive dissection.
Although longer followup and a larger sample population
are needed to make definitive statements with regard to the
preservation of potency, the posterior sagittal approach allows the surgeon to remain midline and away from the
neurovascular bundles. In addition, there is little to no blind
dissection with this approach. The preservation of potency is
always a concern when addressing pelvic pathology in the
setting of revisional surgery and this direct anatomical approach would likely facilitate that goal.
In our series we have also observed that continence of
stool or urine is not affected by this approach. Of course, this
is a concern when addressing residual pathological conditions in children with associated ARM but in a normal
individual this does not apply. Although the preservation or
creation of urinary continence cannot be demonstrated in
every patient with ARM, one must remember the high incidence of neurological issues in children with an ARM.11,12 In
children without an ARM the confounding variable may be
the preexisting condition, especially in the setting of
trauma. An important measure for assessing the value of
this approach is the recurrence of pathological conditions,
which we did not note in our series.
CONCLUSIONS
The posterior sagittal transanorectal approach is a safe and
reasonable alternative for accessing difficult perineal
anatomy, especially in the setting of revisional surgery. The
few patients who have demonstrated that they are potent
with intact ejaculation help further the argument in favor of
this approach.
FIG. 2. Child without ARM. Mllerian remnant is easily addressed

by dividing rectum.

The benefits of the posterior sagittal approach are the
ability to stay midline, excellent exposure and visualization,
and in this series the avoidance of an incision through prior
scar (fig. 2). In our small series we observed no increased
rate of damage to the urinary tract or deleterious effects on
potency after surgery. The posterior sagittal transanorectal
approach should be included in the decision process, especially in settings in which prior surgical attempts have
failed.

ARM anorectal malformation
REFERENCES
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
DeVries PA and Pea A: Posterior sagittal anorectoplasty.

Pea A: The surgical management of persistent cloacas. Results in 54 patients treated with the posterior sagittal approach. J Pediatr Surg 1989; 24: 590.
Pea A and Hong A: The posterior sagittal trans-sphincteric
and trans-rectal approaches. Tech Coloproctol 2003; 7: 35.
Kilpatrick FR and York Mason A: Post-operative rectoprostatic
fistula. Br J Urol 1969; 41: 649.
Stephenson R and Middleton R: Repair of rectourinary fistulas
using a posterior sagittal transanal transrectal (modified
York-Mason) approach: an update. J Urol 1996; 155: 1989.
Belman AB, King LR and Kramer SA: Imperforate anus, urogenital sinus, and cloaca. In: Clinical Pediatric Urology, 4th
ed. London: Martin Dunitz 2000; pp 811 859.
Rossi F, De Castro R and Ceccarelli P: Anterior sagittal
transanorectal approach to the posterior urethra in the
pediatric age group. J Urol 1998; 160: 1173.
Hong AR, Rosen N, Acua MF, Pea A, Chaves L and Rodriguez G: Urological injuries associated with the repair of
anorectal malformations in male patients. J Pediatr Surg
200; 37: 339.
Siegal J, Brock W and Pea A: Pediatric urology: transrectal
posterior sagittal approach to prostatic utricle (mllerian
duct cyst). J Urol 1995; 153: 785.
Pea A: The transanorectal approach for the treatment of
urogenital sinus. Dial Pediatr Urol 1998; 21: 4.
Pea A and Levitt MA: Anorectal malformations. Unpublished
data.
Levitt MA and Pea A: Complications after the treatment of
ARM and redo operations. In: Anorectal Malformations in
Children. Edited by AM Holschneider and J Hutson. Heidelberg: Springer 2006; chapt 24, pp 319 328.
EDITORIAL COMMENT
To me the interest of the current report centers around the
12 patients without anorectal malformation. These 12 patients include 4 with previously operated posterior urethral
strictures and 8 with mllerian diverticula. It is reassuring
to know that bowel and urinary control are not affected by
this approach. Unfortunately these authors provide little
information to understand the justification of such an approach in cases of urethral stricture. The nature of the
previous operations and the length of the stricture are not
reported, and data on postoperative urinary flow rates are
not presented. Good results after failed initial repair can be
achieved in most cases by repeat perineal anastomosis with
or without partial pubectomy.1,2
1671
Regarding the excision of mllerian diverticula, it has

been my experience that this operation is rarely indicated.
The authors failed to mention whether these patients underwent previous hypospadias repairs or how the previously
attempted excision was performed. In cases of hypospadias a
symptomatic mllerian diverticulum is usually secondary to
a distal urethral stricture, for which the treatment usually
solves the problem. In rare cases in which excision is required the laparoscopic approach has given good results.
Finally, in difficult cases of vaginal reconstruction or
rectourethral fistulas I have found the anterior sagittal
transrectal approach (reference 7 in article) without diverting colostomy and with prompt resumption or enteral nutrition to be satisfactory.3
The authors again bring to our attention this approach,
which expands the available options to resolve these difficult
cases.
Ricardo Gonzlez
Thomas Jefferson University
Alfred I. DuPont Hospital for Children
Wilmington, Delaware
1.
Shenfeld O, Gofrit O, Gdor Y, Landau E and Pode D. Anastomotic urethroplasty for failed previously treated membranous urethral rupture. Urology 2004; 63: 837.
2. Hafez A, El-Asmy A, Sarhan O, El-Hefnawy A and Ghoneim M:
Perneal anastomotic urethroplasty for the managing posttraumatic urethral strictures in children: long term outcome.
BJU Int 2005; 95: 403.
3. Mauermann J, Gonzlez R, Franc-Guimond J and Filipas D:
The anterior sagittal transrectal approach (ASTRA) for traumatic urethrovaginal fistula closure. J Urol 2004; 171: 1650.
REPLY BY AUTHORS
As stated in the article the purpose of this presentation is
not to revolutionize the way in which these relatively rare
pathological conditions are addressed, but rather to add to
the options available to the surgeon.
With respect to the specific question concerning children
with mllerian remnants, all had undergone prior abdominal or perineal attempts at repair. Of these cases 7 were
associated with hypospadias, and 1 was associated with a
seminal vesicle cyst and ipsilateral renal anomaly. In this
group transabdominal attempts at resection failed in 2 patients. One patient had a failed endoscopic attempt at unroofing of the diverticulum. The remainder of the group had
undergone multiple attempts as hypospadias repair, about
half of whom were hypospadias cripples. All patients had
problems with repeat infections, and 1 had an orchiectomy
before presentation while 2 underwent vasectomy secondary
to recurrent epididymitis. There were no documented cases
of stricture after repair but several fistulas were noted in the
charts.
All 4 patients in the stricture group had multiple failed
attempts at repair. We agree that the perineal approach is
an option, but this was already attempted in 2 of these
patients without success. We acknowledge the limitation in
documenting continence without postoperative uroflow studies and this is a weakness of our study. Attempts are being
made to contact the patients for long-term followup.
Dirt Bikes and All Terrain Vehicles:

The Real Threat to Pediatric Kidneys
Hsi-Yang Wu* and Barbara A. Gaines
From the Department of Pediatric Urology, Childrens Hospital of Pittsburgh, Pittsburgh, Pennsylvania
Purpose: Recent reviews show that bicycles are the major cause of significant renal injury with few injuries occurring during
contact sports. All-terrain vehicles are also responsible for significant pediatric renal trauma. We determined whether dirt
bikes and all-terrain vehicles cause more significant renal injuries than contact sports.
Materials and Methods: A retrospective review of our pediatric trauma database revealed 115 consecutive patients treated
for renal trauma from 2000 to 2005. A total of 20 bicycle injuries occurred, including 6 on dirt bikes. A total of 13 all-terrain
vehicle injuries occurred, including 4 involving rollovers. A total of 12 contact sport injuries occurred, including 2 during
pick-up games. The mean grade of renal injury was compared among the mechanisms, with grades IIIV considered high
grade.
Results: In descending order of renal injury the mechanisms were dirt bike (2.8), all-terrain vehicle rollover (2.8), bicycle
(2.3), all-terrain vehicle (2.1), contact sports (1.8) and organized contact sports (1.4). Dirt bikes and all-terrain vehicle
rollovers caused significantly greater renal trauma than organized contact sports (2.8 vs 1.4, p 0.007 and 0.02, respectively),
whereas overall bicycle and all-terrain vehicle accidents resulted in similar renal trauma grades compared to those of all
contact sports. The 2 high grade renal injuries during contact sports occurred during pick-up football games without
protective gear.
Conclusions: Physician advice regarding children with a solitary kidney should include avoiding dirt bikes and all-terrain
vehicles. Efforts to limit all-terrain vehicle use in children younger than 16 years would decrease the risk of significant renal
injury in this population more effectively than limiting contact sports participation.
Key Words: wounds and injuries, kidney; off-road motor vehicles
atients with a solitary kidney who wish to participate

in contact sports require individual assessment according to AAP recommendations.1 Recent reviews
show that bicycles and ATVs are the major recreational
cause of significant renal injury with few injuries occurring
during contact sports.2 6 We determined whether dirt bikes
and ATVs cause more significant renal injuries than contact
sports such as football and hockey since this would also
suggest a change in physician counseling of patients with a
solitary kidney.

After obtaining approval with a waiver of consent from the
University of Pittsburgh institutional review board the
records of all children admitted to our institution with renal
injury between January 1, 2000 and December 31, 2005
were reviewed. Data on subjects were obtained from the
trauma registry using Collector (Digital Innovation, Forest
Hill, Maryland). Demographic information, injury mechanism, protective equipment use, overall injury severity, renal injury grade and outcome were evaluated. Injury grade
was based on the American Association for the Surgery of
Study received approval with a waiver of consent from the University of Pittsburgh institutional review board.
* Correspondence: G205 DeSoto Wing, Childrens Hospital of
Pittsburgh, 3705 5th Ave., Pittsburgh, Pennsylvania 15213 (telephone: 412-692-7932; FAX: 412-692-7939; e-mail: wuhy@chp.edu).
0022-5347/07/1784-1672/0
Trauma scoring system for renal injury.7 The ISS was determined from International Classification of Diseases, 9th
revision codes using the Collector database Tri-Code software addition. The mean grade of renal injury was compared
among mechanisms with grades IIIV considered high
grade. For study purposes and according to AAP criteria
basketball, football, hockey and soccer were considered contact sports. Bicycling, baseball, gymnastics, horseback
riding, kickball, running, skating, skiing, sledding, snowboarding and softball were categorized as limited or noncontact sports.1 Statistical analysis was performed using 1-way
ANOVA.
RESULTS
A total of 116 blunt injuries were found in 115 consecutive
patients, including 92 boys (80%) and 23 girls (20%). A total
of 20 bicycle injuries occurred, including 6 on dirt bikes, and
13 ATV injuries occurred, including 4 involving rollovers. All
injuries except 1 ATV injury involved a child younger than
16 years. A total of 12 contact sport injuries occurred, including 2 during pick-up games. Low grade renal injuries
predominated. Injury was grades I to V in 45, 31, 23, 15 and
2 cases, respectively. Two injuries occurred in patients with
a solitary kidney, including 1 grade I injury during baseball
and 1 grade III injury during soccer. Three patients with
abnormally positioned kidneys were identified. Two injuries
1672
Vol. 178, 1672-1674, October 2007

Printed in U.S.A.
DOI:10.1016/j.juro.2007.03.160
RENAL INJURIES CAUSED BY DIRT BIKES AND ALL TERRAIN VEHICLES

occurred in horseshoe kidneys, including a grade I bicycle
injury and a grade IV motor vehicle collision injury. One
grade II injury from a motor vehicle collision occurred to a
pelvic kidney. One patient with a grade IV injury from a fall
underwent nephrectomy 3 weeks after the injury. No other
patients required nephrectomy. The table lists the mechanism, number of patients, renal injury grade and ISS.
Dirt bikes and ATV rollovers caused significantly
greater renal trauma than organized contact sports (2.8 vs
1.4, p 0.007 and 0.02), whereas overall bicycle and ATV
accidents resulted in similar grades of renal trauma compared to all contact sports (p 0.24 and 0.55, respectively). When dirt bike injuries and ATV rollovers were
combined, the difference in renal trauma grade compared
with that of organized contact sports injuries remained
significant (2.8 vs 1.4, p 0.004). The 2 high grade renal
injuries (grades III and IV, respectively) during contact
sports occurred during pick-up football games without
protective gear. Limited contact sports had a slightly
higher grade of renal injury than organized contact
sports, which was not statistically significant (2.1 vs 1.4,
p 0.08). When dirt bikes (2.8) and ATV rollovers (2.8)
were compared with all contact sports (1.8, p 0.06 and
0.14), limited contact sports (2.1, p 0.18 and 0.31) and
falls (2.4, p 0.41 and 0.6, respectively), the difference in
renal trauma grade became progressively smaller. ISS
was significantly greater for all bicycle and ATV injuries
than for organized contact sports (12.9 and 18.5 vs 5.6,
p 0.007 and 0.006, respectively). The increase in ISS
was mainly due to splenic and liver lacerations, concussions and extremity fractures in the ATV and bicycle
groups.
DISCUSSION
Patients with a solitary kidney who wish to participate in
contact, collision or limited contact sports may do so after
individual assessment by a physician, according to the
AAP Committee on Sports Medicine and Fitness.1 Recent
reviews of pediatric patients with blunt renal injuries
pointed out that bicycling is the most common recreational cause of renal injury, whereas team contact sports
are an uncommon cause, and so a change in AAP recommendations about team contact sports was advocated.2 6
Mechanism of blunt renal injury by grade

Mechanism
No. Pts
Mean SD
Renal Injury Grade
Mean SD
Injury Severity Score
Dirt bike
ATV rollover
All bicycle
All ATV
All contact sports
Organized
contact sports
All limited
contact sports
Ski/sled
Horseback
Snowboard
Fall
Motor vehicle
collision
6
4
20
13
12
10
2.8 1.2
2.8 1.3
2.3 1.3
2.1 1.0
1.8 1.1
1.4 0.7
11.0 6.1
21.0 12.8
12.9 6.8
18.5 13.1
6.3 3.7
5.6 2.4
20
2.1 1.1
12.4 6.0
5
4
4
17
33
2.0 1.4
2.0 0.8
1.8 0.5
2.4 1.2
2.1 1.0
12.0 5.6
10.0 5.8
15.8 6.8
11.1 7.2
22.0 12.0
1673
While most ATV related injuries are orthopedic, head or

spinal cord injuries,8 ATVs caused as many renal injuries as
contact sports in this series. When they rolled over, ATVs
caused significantly higher grades of renal trauma than
organized contact sports. Our finding that ATVs and dirt
bikes cause significantly greater renal injuries than contact
sports confirms impressions from previous studies that the
rate of renal loss from contact sports is much lower than that
from limited or noncontact sports.3,6
The high rate of ATV injuries in this series may be due
to a greater popularity of ATVs in western Pennsylvania
compared to that at other locations or it may reflect an
increase in overall ATV use since the voluntary decree on
limiting ATV use expired in 1998. ATVs are commonly
used by children younger than 16 years despite laws requiring a helmet and training certificate.9 Compared to
bicycles the higher center of gravity, the additional mass
of the ATV, its ability to attain higher speeds and the
inability of a younger child to control it adequately contribute to the significantly higher ISS caused by ATVs.10
Since current laws restricting ATV use do not appear to
have an impact on patient morbidity,11 physician advice
to patients and parents may be the only chance to prevent
injuries from ATVs. Bicycle associated renal injuries are
usually due to handlebar trauma.12 Thus, it is not surprising that dirt bikes, which are more often used off road,
for jumps and for other stunts, would cause higher grades
of injury than other bicycles. The common factor that
differentiates ATVs and dirt bikes from other forms of
cycling would appear to be the higher risk of deceleration
injury.6
We also confirmed the findings of a previous study that
limited contact sports such as skiing, snowboarding and
horseback riding can cause high grades of renal injury,
which may result in nephrectomy.6 Interestingly the high
grade injuries from football occurred during pick-up
games, when no protective gear was worn. Although pediatric urologists,13 nephrologists14 and sports medicine
physicians15 are reluctant to allow patients with a solitary kidney to play contact sports, it would appear that a
grouping of sports that could result in sudden deceleration
and abdominal injury would be more useful for counseling
patients with a solitary kidney.
A reasonable recommendation by a pediatric urologist for
a patient with a solitary kidney who wishes to participate in
sports might include certain facts. 1) When normal protective gear for football or hockey is worn, the risk of renal
injury from organized contact sports is low. 2) ATVs and dirt
bikes are associated with a high risk of significant renal
injury. 3) Skiing, snowboarding, sledding and horseback
riding are as risky as organized contact sports.
CONCLUSIONS
Physician advice regarding children with a solitary kidney
should include the avoidance of dirt bikes and ATVs. The
risk of significant renal injury in organized football and
hockey with protective gear is low. Efforts to limit ATV use
in children younger than 16 years would decrease the risk of
significant renal injury in this population more effectively
than limiting participation in contact sports.
1674
RENAL INJURIES CAUSED BY DIRT BIKES AND ALL TERRAIN VEHICLES

7.

AAP American Academy of Pediatrics
ATV all-terrain vehicle
ISS Injury Severity Score
REFERENCES
1.
2.
3.
4.
5.
6.
Medical conditions affecting sports participation. American

Academy of Pediatrics, Committee on Sports Medicine and
Fitness. Pediatrics 2001; 107: 1205.
McAleer IM, Kaplan GW and LoSasso BE: Renal and testis
injuries in team sports. J Urol 2002; 168: 1805.
Gerstenbluth RE, Spirnak JP and Elder JS: Sports participation and high grade renal injuries in children. J Urol 2002;
168: 2575.
Wan J, Corvino TF, Greenfield SP and DiScala C: The incidence of recreational genitourinary and abdominal injuries
in the western New York pediatric population. J Urol 2003;
170: 1525.
Wan J, Corvino TF, Greenfield SP and DiScala C: Kidney and
testicle injuries in team and individual sports: data from
the national pediatric trauma registry. J Urol 2003; 170:
1528.
Johnson B, Christensen C, DiRusso S, Choudhury M and
Franco I: A need for reevaluation of sports participation
recommendations for children with a solitary kidney. J Urol
2005; 174: 686.
8.
9.
10.
11.
12.
13.
14.
15.
Moore EE, Shackford SR, Pachter HL, McAninch JW, Bronner

BD, Champion HR et al: Organ injury scaling: spleen, liver,
and kidney. J Trauma 1989; 29: 1664.
Cvijanovich NZ, Cook LJ, Mann NC and Dean JM: A population-based assessment of pediatric all-terrain vehicle injuries. Pediatrics 2001; 108: 631.
Keenan HT and Bratton SL: All-terrain vehicle legislation for
children: a comparison of a state with and a state without a
helmet law. Pediatrics 2004; 113: e330.
Brown RL, Koepplinger ME, Mehlman CT, Gittelman M and
Garcia VF: All-terrain vehicle and bicycle crashes in children: epidemiology and comparison of injury severity. J Pediatr Surg 2002; 37: 375.
Upperman JS, Schults B, Gaines BA, Hackam D, Cassidy LD,
Ford HR et al: All-terrain vehicle rules and regulations:
impact on pediatric mortality. J Pediatr Surg 2003; 38:
1284.
Nadler EP, Potoka DA, Shultz B, Morrison KE, Ford HR and
Gaines BA: The high morbidity associated with handlebar
injuries in children. J Trauma 2005; 58: 1171.
Sharp DS, Ross JH and Kay R: Attitude of pediatric urologists
regarding sports participation by children with a solitary
kidney. J Urol 2002; 168: 1811.
Grinsell MW, Showalter S, Gordon KA and Norwood VF: Single kidney and sports participation: perception versus reality. Pediatrics 2006; 118: 1019.
Anderson CR: Solitary kidney and sports participation. Arch
Fam Med 1995; 4: 885.
3-Dimensional Magnetic Resonance

Genitography: A Different Look at Cloacal Malformations
Steven M. Baughman, Randy R. Richardson,* Daniel J. Podberesky, Neal C. Dalrymple and
Elizabeth B. Yerkes,
From the Departments of Urology and Radiology (RRR, DJP, NCD), Wilford Hall United States Air Force Medical Center, Lackland Air
Force Base, Texas
Purpose: Standard contrast genitography and endoscopy have been the cornerstone of preoperative investigation of
anatomy in complex cloacal malformations. In many cases standard genitography is ambiguous and lengthy endoscopy
may be required to define the anatomy. Nonenhanced magnetic resonance imaging lacks definition of the nondistended
common channel and confluence. To our knowledge we introduce 3-dimensional magnetic resonance genitography for
delineating the unique anatomical features of these complex anomalies.
Materials and Methods: Contrast genitography, endoscopy and 3-dimensional magnetic resonance genitography were
performed preoperatively in 4 female infants with cloacal malformations. Three-dimensional magnetic resonance genitography was performed using a 3-dimensional spoiled gradient technique after the instillation of gadolinium-diethylenetetramine pentaacetic acid solution via the common channel and cutaneous vesicostomy, when present. Augmented pressure
distal colostogram with gadolinium via the mucous fistula was included. Three-dimensional images were constructed using
a commercially available workstation.
Results: Three-dimensional magnetic resonance genitography yielded anatomical information far superior to that of
standard genitography in all cases and it provided information on female genital tract anomalies and hindgut confluence
that were not fully appreciated at initial endoscopy. The confluence was most accurately represented by augmented
pressure colostogram. The length of the common channel was consistent with that on endoscopy. Rotating 3-dimensional
images illustrated the unique anatomical relationships. No magnetic resonance imaging specific complications were
observed.
Conclusions: Three-dimensional magnetic resonance genitography provides superior anatomical detail for the preoperative assessment of cloacal malformations. It augments the information obtained by standard magnetic resonance
imaging of the pelvic organs, levators and lumbosacral spine. Three-dimensional magnetic resonance genitography has
great potential as a standard for preoperative anatomical evaluation and it will likely simplify preoperative endoscopy
rather than supplant it.
Key Words: abnormalities; cloaca; anatomy; magnetic resonance imaging; imaging, three-dimensional
loacal malformations are among the most interesting

and challenging anomalies facing pediatric urological
and general surgeons. While the initial diagnosis is
straightforward, the importance of thorough preoperative
assessment of the anatomy cannot be overstated.
Retrograde contrast genitography and endoscopy are the
standard techniques to determine the length of the common
channel and anticipate the technical challenges and prognosis associated with the level of confluence. While each provides useful information, each also has shortcomings. MRI of
the lumbosacral spine and pelvis can provide additional
complementary information.13 To our knowledge we intro-
The views of this manuscript represent the views of the authors

and do not necessarily reflect those of the Department of Defense.
* Current address: Department of Radiology, St. Josephs Hospital, Phoenix, Arizona.
Current address: Department of Radiology, University of Texas
Health Science Center at San Antonio, San Antonio, Texas.
Correspondence: Department of Urology, 2300 Childrens Plaza,
Box 24, Childrens Memorial Hospital, Chicago, Illinois 60613.
Current address: Division of Urology, Childrens Memorial Hospital, Chicago, Illinois.
0022-5347/07/1784-1675/0
duce the technique of 3D-MRG and describe its usefulness

for delineating the complex anatomical relationships in
these challenging patients.

Standard contrast genitography, endoscopy and 3D-MRG
were performed preoperatively in 4 female infants with cloacal malformations. Genitography was performed at initial
diagnosis. Each child underwent creation of colostomy and
mucous fistula with planned delayed definitive reconstruction. Endoscopy and 3D-MRG were performed with the patient under anesthesia on separate occasions.
MR images were acquired at 1.5 Tesla using a Signa
Excite HD. General anesthesia was provided and monitored by a pediatric anesthesiology. Patients were imaged in
a head or pelvis coil using a combination of multiplanar
T1-weighted, repetition time 400 milliseconds, echo time 8
milliseconds, 2 excitations and fat saturated, T2-weighted,
repetition time 3,500 milliseconds, echo time 100 millisec-
1675
Vol. 178, 1675-1679, October 2007

Printed in U.S.A.
DOI:10.1016/j.juro.2007.03.196
1676
3-DIMENSIONAL MAGNETIC RESONANCE GENITOGRAPHY
onds, 3 excitations spin-echo sequences through the pelvis

and lumbosacral spine. T1 and T2-weighted sequences were
obtained using a 3 mm section thickness, a 3 to 4 mm
interslice gap and a 256 192 matrix. A thick slab heavily
T2-weighted MR urogram sequence was performed to evaluate the upper urinary tract. Field of view varied between
18 and 22 cm depending on patient size.
A 3D-SPGR sequence was acquired in the coronal or
sagittal plane (repetition time 6.6 milliseconds, echo time
1.5 milliseconds, 2 excitations) with a flip angle of 15 degrees, 3 mm section thickness, 3 mm interslice gap and a
256 192 matrix. Field of view varied between 18 and 22 cm
depending on patient size. For the SPGR sequence imaging
was performed during the hand instillation of a dilute solution of gadodiamide (Omniscan) via the common channel
and the mucous fistula as well as the cutaneous vesicostomy
when present. Gadolinium was diluted with normal saline to
a ratio of between 1:750 and 1:1,000.
To maintain adequate confluence hydrodistention the cloaca was prefilled with gadolinium solution and then actively
instilled during the SPGR sequence. Balloon occlusion was
used at each orifice with the balloon pulled snugly against
the internal rim of the vesicostomy and taped on tension in
the mucous fistula to prevent gadolinium backflow. The
balloon was taped against the perineal orifice with a crisscross dressing. Extension tubing was used to overcome access issues during the scan.
The initial patients also underwent dynamic gadolinium
enhanced MR angiogram (0.1 mmol/kg) to evaluate the renal
vasculature and ureters following contrast medium excretion. However, this made evaluation for vesicoureteral reflux impossible following retrograde instillation of contrast
material into the cloaca. Therefore, it was decided to eliminate this sequence in later patients.
Average total actual scanning time was approximately 30
minutes. Patient preparation, positioning and scan planning
added another 30 to 60 minutes to total examination time.
Post-processing, including multiplanar maximum intensity
projection images and rotating 3D images, were constructed
from the 3D-SPGR sequence on an Advantage Workstation
by an experienced radiologist. Average post-processing time
was approximately 30 minutes.
addition of an augmented pressure distal colostogram. Rotating 3D images offer a spatial perspective of the unique
anatomy that is missed by standard genitography and endoscopy.
No MRI specific complications were observed. The Appendix and figures 1 to 4 show the usefulness of 3D-MRG and
compare this technique to standard modalities.
DISCUSSION
Management of cloacal malformations challenges even the
most experienced specialists. Knowledge of the anatomy
not only prepares the surgeon, but also provides important prognostic information for the family regarding future urinary and fecal continence.4 The confluence of the
urinary, genital and gastrointestinal tracts in these children follows grossly predictable patterns and yet each
child is unique.
Standard contrast genitography delineates the length of
the common channel and confirms the suspicion of cloacal
malformation. However, as the dilated bladder, vagina
and/or rectum fills, the double density in this 2-dimensional
view can obscure important details of the confluence. In
addition, the fluoroscopy and x-rays used for this technique
expose the patient to ionizing radiation.
Endoscopy can be quite involved in patients of this
nature and details such as genital tract duplication can be
overlooked. The true spatial relationships between the
bladder, vaginas, rectum and perineum are left somewhat
to the imagination even after selective catheterization of
the internal orifices for anatomical confirmation with contrast instillation. MRI has an established role in these
patients for assessing the lumbosacral cord and defining
RESULTS
Three-dimensional MRG yielded anatomical information
superior to that of standard genitography in all cases. It
provided information on female genital tract anomalies
(duplication and partial agenesis) and the degree of rectal
atresia that was not fully appreciated at initial endoscopy.
These anomalies were also not defined by standard genitography.
The length of the common channel was well represented
by 3D-MRG in the 2 patients in whom a proper common
channel was present. It correlated with the length seen
during genitography and endoscopy. However, the common
channel was more challenging to image with 3D-MRG due to
the need for active injection against the perineum with the
patient in the MR scanner. The confluence and true level of
the rectal segment were most accurately represented by the
FIG. 1. Female infant with cloacal atresia and no perineal orifice.

Cystogram via vesicostomy did not define genital tract or rectal
fistula. Two cervixes and rectal fistula near bladder dome were
missed at initial endoscopy and were only identified later with
concomitant irrigation via mucous fistula, while 3D-MRG showed
2 small vaginal cuffs (arrows) and efflux through 1 fallopian tube
(arrowhead). Rectal fistula was not revealed by retrograde filling.
Intravenous gadolinium was given but patient also had bilateral
vesicoureteral reflux. Distal colostogram was not yet part of
protocol but it would have confirmed high rectal atresia. Intraoperative findings were consistent with 3D-MRG.
FIG. 2. Female infant with imperforate anus and genital tract

duplication. Child had normal labia and single perineal opening
in vaginal position. Endoscopically urethra was normal length,
and rectal and vaginal openings were just proximal to perineal
orifice. Confluence was missed by standard genitography. Gadolinium was given intravenously and via perineal catheter. Each
vagina opacified (arrows). Augmented pressure distal colostogram defined distal rectal segment (arrowhead). B, bladder. R,
rectum.
the visceral and muscular pelvic anatomy.1,2 Decompression of the distended systems after initial medical and
surgical intervention limits assessment of the rectal fistula, genital tract anomalies and common channel. Fast
spin-echo T2-weighted images have revealed the fistula in
newborns.3
FIG. 3. Female infant with persistent cloaca. Genitogram provided useful information on common channel length and reflux.
Lengthy endoscopy augmented by contrast instillation into internal orifices revealed almost 4 cm common channel with urethra
entering anterior wall of dilated vagina (V) and rectum entering
medial vaginal wall. Smaller duplicate vagina with stenotic opening was missed during initial endoscopy, standard genitogram
and 3D-MRG. This vagina entered common channel on right,
near orifice of larger vagina. Unenhanced MRI demonstrated
pelvic anatomy and truncated sacrum. Gadolinium was administered intravenously and via vesicostomy. Urethra was clearly
shown entering vagina (arrow). Gadolinium was actively instilled
via perineal orifice to opacify common channel (arrowhead). Distal colostogram defined rectal segment (R). B, bladder.
1677
FIG. 4. Female infant with persistent cloaca. Standard genitography filled only vagina (V) due to urethral and bladder (B) neck
compression by vaginal distention. Rectal segment was opacified
by colostogram only. Access to bladder endoscopically was difficult due to progressive vaginal distention. Vaginal orifice was
adjacent to urethral orifice at apex of common channel (CC).
Rectal fistula was identified on vaginal floor distally. Common
channel was less than 3 cm long. Gadolinium instillation via
perineal orifice and mucous fistula enhanced urethra (arrow) and
common channel (not shown). Compression of distal urethra by
distended vagina gave false impression that urethra entered
vagina directly, rather than entering common channel. Due to
long atretic distal rectal segment cloaca could not be opacified via
mucous fistula but high level of rectal atresia was well demonstrated (arrowhead). Intraoperative findings of high rectal atresia and high vaginal confluence were consistent with 3D-MRG. R,
rectum.
We propose 3D-MRG as an alternative to standard genitography and an enhancement of standard pelvic MRI. The
concept was inspired by the definition of the upper urinary
tract with MR urography.5 In addition, we used the important concept of identifying the level of rectal atresia with
augmented pressure distal colostogram, as previously emphasized by Gross et al.6 Indeed, the primary advantage of
3D-MRG over standard contrast genitography or standard
MRI is the information offered by 3D reconstruction of the
hydrodistended confluence and clear imaging of the rectal
segment. Augmented pressure distal colostogram is an important element of this study. The position of the rectal
fistula on endoscopy may not reflect the true level of the
segment suitable for pullthrough. The most accurate information is obtained by prefilling all elements of the cloaca
with gadolinium and then actively instilling solution during
the scan to maintain hydrodistention of the critical regions.
It is possible to opacify the common channel accurately
with 3D-MRG but this is the most challenging and least
satisfying aspect of the technique. Active instillation is required while the patient is deep in the scanner and a tight
seal between the balloon and perineal orifice can be difficult
to maintain. Definition of the distance from the rectal atresia and the vagina(s) to their respective targets on the perineum is important for surgical planning and it could be
further enhanced by placing a perineal marker, such as a
vitamin E capsule.
The enhanced information offered by 3D-MRG is most
useful in suspected high confluence cases. We do not anticipate that this technique would eliminate endoscopic evalu-
1678
ation in these complex patients. Rather, the techniques are

complementary. Preparations for abdominal exploration, in
addition to posterior sagittal dissection or a projected need
for vaginal replacement, are the main advantages of this
preoperative assessment.
A recently published case report described 3D reconstructed computerized tomography for evaluating complex
anatomy in an infant with a cloacal malformation.7 The
anatomy was beautifully demonstrated. The examination
time of the 3D computerized tomography is shorter but
3D-MRG offers several key advantages, including the lack of
ionizing radiation exposure and concomitant standard imaging of cord anomalies and pelvic floor anatomy.
Due to patient age general anesthesia is recommended
to maximize patient safety and minimize motion artifact.
The length of the study is further increased by imaging
the lumbosacral spine to assess sacral anomalies and
tethered cord but all images can be obtained using the
same anesthesia. The average actual scanning time for a
complete examination was 30 minutes. As the protocol is
refined, it is possible that certain sequences could be
eliminated, as we have already done with the dynamic IV
gadolinium MR angiogram. For example, the MR urogram
could be eliminated if the upper tracts were previously
evaluated adequately by ultrasound. The multiplanar pregadolinium T1 and T2 sequences could be narrowed down
to a single axial T2 sequence, and coronal and axial T1
sequences. These changes would shorten actual examination time to less than 20 minutes. In addition, as experience is gained with this technique, examination time and
post-processing time would be expected to decrease significantly. Three-dimensional SPGR can be performed with
the patient in the sagittal or coronal plane since images
are subsequently constructed into rotating 3D images
that can be reformatted later into any plane desired. If
software is not available for 3D reconstruction or reconstruction is not desired, the highest yield sequence would
be sagittal SPGR with instillation of diluted gadolinium
solution.
We continue to apply this technique to patients with
cloacal malformations. We have also found it useful in select,
previously operated patients who are considered candidates
for further reconstruction.
APPENDIX
3D-MRG vs Standard Genitography and Endoscopy
for Preoperative Assessment of Cloacal
Malformations
Factors in Favor of 3D-MRG
Elimination of contrast double densities
Lack of exposure to ionizing radiation
3D spatial illustration of anatomical relationships
Definition of true level of rectal atresia (caliber and distance from
perineum)
Definition of level of vaginal confluence (distance from perineum)
Identification of genital tract duplication
Identification of ureteral ectopia or vesicoureteral reflux
Assessment of lumbosacral spinal cord using same anesthesia/sedation
Simplification of endoscopic assessment
Factors Less in Favor of 3D-MRG
Less consistent demonstration of common channel (advantage: endoscopy)
Need for anesthesia/sedation (advantage: genitography)
Cost (advantage: genitography)

3D
3D-MRG
MR
MRI
SPGR
3-dimensional
3D MR genitography
magnetic resonance
MR imaging
spoiled gradient
REFERENCES
1.
2.
3.
4.
5.
6.
CONCLUSIONS
7.
For preoperative assessment of complex genitourinary

anomalies 3D-MRG provides excellent anatomical detail.
The technique augments the information previously obtained by standard MRI of the pelvic organs, levators and
lumbosacral spine. In addition, there is no exposure to ionizing radiation with this technique.
Also, 3D-MRG provides information complementary to
that of endoscopy and it will likely simplify rather than
replace endoscopic evaluation in these complex cases. It
provides a unique illustration of the spatial relationships,
particularly in the most complex cases. This protocol has
great potential as a tool for preoperative anatomical evaluation of the most challenging genitourinary malformations.
Sato Y, Pringle KC, Bergman RA, Yuh WT, Smith WL, Soper RT
et al: Congenital anorectal anomalies: MR imaging. Radiology 1988; 168: 157.
Jaramillo D, Lebowitz RL and Hendren WH: The cloacal malformation: radiologic findings and imaging recommendations. Radiology 1990; 177: 441.
Nievelstein RAJ, Vos A, Valk J and Vermeij-Keers C: Magnetic
resonance imaging in children with anorectal malformations: embryologic implications. J Pediatr Surg 2002; 37:
1138.
Levitt MA and Pea A: Outcomes from the correction of anorectal malformations. Curr Opin Pediatr 2005; 17: 394.
McDaniel BB, Jones RA, Scherz H, Kirsch AJ, Little SB and
Grattan-Smith JD: Dynamic contrast-enhanced MR urography in the evaluation of pediatric hydronephrosis: part
2, anatomical and functional assessment of ureteropelvic
junction obstruction. AJR Am J Roentgenol 2005; 185:
1608.
Gross GW, Wolfson PJ and Pea A: Augmented-pressure colostogram in imperforate anus with fistula. Pediatr Radiol
1991; 21: 560.
Adams ME, Hiorns MP and Wilcox DT: Combining MDCT,
micturating cystography, and excretory urography for 3D
imaging of cloacal malformation. AJR Am J Roentgenol
2006; 187: 1034.
EDITORIAL COMMENT
The cloaca remains one of the supreme challenges of perineal reconstruction because of the inherent wide variability in the anatomy. This can influence the length and size
of the cloacal channel, and the vaginal configuration,
which can be duplicated with varying sizes of the vagina
as well as entry points into the cloaca. The position of the
connection of the imperforate anus to the cloaca again is
highly variable, sometimes with a duplicated vagina, actually entering anterior to enlarged vaginal segments.

The result of this great variability is that planning a
reconstructive procedure for this anomaly has always
been one of the major challenges faced by reconstructive
pediatric urologists. I have in my practice routinely depended not only on distal colonic x-rays, but also on retrograde studies and I have always planned independent
cystoscopic assessment of the anatomy before planning
the definitive operation.
The advent of this 3D reconstruction by an MR technique
may obviate most of these steps by providing 1 elegant 3D
reconstruction of all of the pelvic anatomy, thus, permitting
the accurate planning of a reconstructive procedure. This is
1679
a great contribution to our field and this report should be

carefully reviewed by anyone performing this surgery. While
we are still fairly early in the evolution of MR availability, I
would hope that the day is on the near horizon when we will
all be able to avail ourselves of this type of elegant diagnostic
study.
Howard M. Snyder, III
Division of Urology
Vesicoureteral Reflux I
Intravesical Electrotherapy for
Neurogenic Bladder Dysfunction: A 22-Year Experience
Jennifer A. Hagerty,* Ingrid Richards and William E. Kaplan
From the Department of Pediatric Urology, Childrens Memorial Hospital, Chicago, Illinois
Purpose: Intravesical electrotherapy was previously shown to be effective for improving bladder capacity and compliance.
We evaluated our 22-year experience with this therapy in patients with neurogenic bladder.
Materials and Methods: The charts of 405 patients who received intravesical electrotherapy were reviewed. Cystometrograms were performed at the start of each treatment series. Bladder capacity and pressure were determined for each patient
before and after therapy. Patients were also questioned regarding the sensation of bladder filling.
Results: From 1985 to 2006, 372 patients with an average age of 5.5 years (range 0 to 43) had followup information available and
were included for evaluation. Patients received a median of 29 treatment sessions (range 2 to 197). Mean patient followup was 6.6
years (range 0 to 19.7). Of the 372 patients 286 (76.9%) had a 20% or greater increase in bladder capacity after treatment. In this
subset of patients bladder storage pressure at capacity was normal (less than 40 cm water) in 74.4% (213 of 286). Of the 17.2% of
patients (64 of 372) who had no change in bladder capacity 81.21% (52 of 64) had normal bladder storage pressures after treatment.
Bladder sensation was developed and sustained in 61.6% of patients (229 of 372), including 33.6% in the first series.
Conclusions: Bladder stimulation is effective for increasing bladder capacity without increasing storage pressure in a
majority of patients. This technique is safe and effective for improving bladder compliance. Some patients also have improved
sensation of bladder filling and they should be able to catheterize when feeling full rather than by the clock.
Key Words: bladder; urinary bladder, neurogenic; electric stimulation; urinary incontinence; meningomyelocele
eurogenic bladder dysfunction carries a risk to the

health and life of the patient. The primary goal of
treatment is preserving renal function and social
continence. This goal depends on achieving bladder compliance in these patients. Many patients are satisfactorily
treated with anticholinergics and CIC.13 However, there
are a subset of patients who are refractory to this conventional treatment and often require bladder augmentation to
ensure bladder compliance and often continence.2,4 The
risks of bladder augmentation are well documented.59
At our center for the last 22 years we have been performing intravesical bladder electrotherapy in these refractory
cases of neurogenic bladder dysfunction. We previously reported successful increases in compliance and modest improvement in sensation.10 13 These results were further
demonstrated in a multi-institutional trial.14
Our initial goals of therapy were volitional voiding and
improved incontinence. However, after treating several patients it was determined that these goals were not being met
and therapy was redirected toward treating bladder compliance. In this long-term evaluation we describe our results of
what is to our knowledge the largest published cohort of patients who have been treated with intravesical electrotherapy.
METHODS
Patients with a low capacity, poorly compliant bladder and/or
urinary incontinence secondary to neurogenic bladder were
* Correspondence: Pediatric Urology, Childrens Memorial Hospital, 700 East Fullerton Ave., Chicago, Illinois 60611 (e-mail:
jhagerty@childrensmemorial.org).
0022-5347/07/1784-1680/0
referred from multiple centers for intravesical electrotherapy.

The charts of 405 patients were reviewed. Only patients with
followup data available were included in the evaluation. Cystometrograms were performed at each treatment series. Bladder capacity and pressure were determined for each patient
before and after therapy. Patients were also questioned regarding the sensation of bladder filling. Patients received treatment
at our institution, as previously described.10 Briefly, a cystometrogram was performed. The bladder was emptied and the
volume obtained was considered the recorded capacity. Detrusor pressure at the end of the cystometrogram, when bladder
capacity was attained, was defined as bladder capacity pressure. The bladder was then filled to half capacity and observed
for any activity to guide bladder stimulation parameters.
These parameters were adjusted as needed throughout therapy to achieve the maximum benefit.
Bladder electrotherapy was performed with a sterile electrocatheter inserted via the urethra into the bladder. A
rectal probe was inserted to measure abdominal straining.
Detrusor pressure was measured. Bladder sensation was
detected throughout the study by careful continuous monitoring. In infants this was often determined by the response
of the child to bladder filling. Leak point pressure and compliance were determined. All data that were generated from
the stimulation process was documented by the observing
staff, including contraction strength, length and number,
the initial residual urine with which the child presented, the
size of the catheter used, the volume of fluid with which the
bladder was filled and end residual volume.
Each daily session lasted 90 minutes. The initial series
consisted of approximately 15 to 20 of these outpatient ses-
1680
Vol. 178, 1680-1683, October 2007

Printed in U.S.A.
DOI:10.1016/j.juro.2007.03.188
INTRAVESICAL ELECTROTHERAPY FOR NEUROGENIC BLADDER DYSFUNCTION

TABLE 1. Demographics
Characteristic
No. sex (%):
M
F
Age (yrs):
Mean
Median
Range
No. diagnosis (%):
Myelomeningocele
Spinal tumor/bleed/infarction
Transverse myelitis
Tethered cord
Sacral agenesis
Neurogenic bladder, other
Neurogenic bladder, unknown etiology
195 (52.4)
177 (47.6)
5.5
4.0
043
329 (88.4)
9 (2.4)
8 (2.2)
6 (1.6)
4 (1.1)
8 (2.2)
8 (2.2)
sions. A 3 to 6-month rest period without stimulation was

then begun. To gain further improvement or after a neurologically induced setback, ie cord tethering or shunt malfunction, repeat series were performed. The second series
commenced in exactly the same fashion, starting with slow
fill cystometrogram, daily therapy sessions and slow fill
cystometrogram at the completion of the sessions. The following series consisted of 5 to 10 daily outpatient sessions.
These 1 to 2-week series continued every 3 to 6 months until
success was achieved.
Medications that may affect bladder dynamics were routinely discontinued a few days before urodynamic studies
and during bladder stimulation therapy. These medications
were sometimes reinstituted between therapy sessions depending on patient condition. They were always reinstituted
between each series. In a few select patients with high
storage pressure medications were not discontinued during
therapy due to possible upper tract deterioration. Urodynamic studies were performed with these patients on the
same doses of medications, so that fair comparisons of studies could be made.
RESULTS
From 1985 to 2006, 405 patients received intravesical electrotherapy and 372 had followup information available and
were included for evaluation. Table 1 lists demographics.
Patients received a median of 29 treatment sessions (range
2 to 197). Mean patient followup was 6.6 years (range 0 to
19.7) following the last treatment series.
An improvement in bladder capacity, defined as a 20% or
greater increase from pretreatment capacity, was seen in
76.9% of patients (286 of 372). In this subset of patients
bladder storage pressure at capacity was normal (less than
40 cm water) in 74.4% (213 of 286). Of the 17.2% of patients
(64 of 372) who had no change in bladder capacity 81.21%
1681
(52 of 64) had normal bladder storage pressures after treatment. An additional 5.6% of patients (21 of 372) had a
decrease in bladder capacity and 23.8% (5 of 21) had increased bladder pressure.
Bladder sensation was developed and sustained in 61.6%
of patients (229 of 372). Sensation developed in 33.6% of
patients in the first series, in 14.0% in the second series and
in approximately 10% in each subsequent series. Age was
similar in patients who did and did not have sensation (6.25
and 5.25 years, respectively).
The lesion spinal level was known in 279 of 372 patients.
Table 2 lists the responses seen at each spinal level.
Patients were maintained on CIC as deemed necessary
for continence and to assist in maintaining low bladder
pressure. At followup 59.1% of patients (220 of 372) were on
CIC. In many patients CIC regimens increased because patients were now catheterizing by sensation instead of by the
clock.
On pretreatment cystometrogram 23.1% of patients (86 of
372) demonstrated a high risk bladder with increased detrusor pressure (more than 40 cm H2O) at capacity and they
were candidates for bladder augmentation. Ten of 372 patients underwent bladder augmentation, including 4 due to
poor compliance. The figure shows a representative patient
cystometrogram.
DISCUSSION
In 1956 by Katona and Berenji initiated intravesical stimulation in Hungary.15 In 1982 their series was substantiated
by Madesbacher et al.16 This was the first attempt to rehabilitate the bladder by the intravesical route instead of by
leaving permanently implantable devices in the body. The
major goals of these investigators was to establish detrusor
contractions in children and adults with neurogenic bladder
disease.15,16
Initially our main goal was to establish effective detrusor
contractions that could then help us teach a child to gain
urinary control through biofeedback techniques. It was established in our prior studies that this goal could only be
attained by 18% to 20% of patients. However, of greater
significance was the goal of increasing bladder capacity and
decreasing bladder pressure. Clinically this increase in bladder capacity allows greater bladder compliance, improved
urinary continence, an increased interval between bladder
catheterizations, less need for anticholinergic medication
and the possible avoidance of bladder augmentation in patients with a poorly compliant bladder. We previously noted
success in these areas, specifically in the myelomeningocele
population.10 14
The results of this long-term study demonstrate that
intravesical bladder stimulation therapy causes a signifi-
TABLE 2. Bladder capacity improvement at each spinal level

Spinal Lesion Level
No. Pts
Cervical
Thoracic
Lumbar
Lumbosacral
Sacral
Unknown
3
17
147
60
52
93
No. Greater Than 20%

Improved Capacity (%)
No. Detrusor Pressure Less Than

40 cm H2O Increased Capacity (%)
No. Sensation (%)
2 (66.7)
14 (82.3)
122 (83.0)
46 (76.7)
34 (65.4)
1 (50)
11 (78.6)
114 (93.4)
32 (70.0)
21 (61.8)
2 (66.7)
6 (35.3)
88 (59.9)
37 (61.7)
38 (73.1)
1682
Male patient treated with intravesical electrotherapy. After 39 sessions capacity improved from 190 to 310 cc at 10 cm H2O detrusor
pressure with resolution of uninhibited bladder contractions. Results were maintained at 3 years of followup.
cant increase in bladder capacity in more than half of the

patients treated with this modality. This increase in capacity occurs without a concomitant increase in storage pressures in the majority of patients. Also, in a modest number
of patients an improvement in bladder sensation was observed. This allowed these patients to sense when they
needed to empty the bladder. This is advantageous for potentially decreasing episodes of incontinence between catheterizations and decreasing the need for timed voiding.
Many of these patients graduated from the program and
maintained these results for many years. Most importantly
augmentation was avoided in the majority of patients.
Experience with bladder stimulation elsewhere has been
variable. In a study by Jannek a sensation of urgency developed in 100% of patients.17 In a randomized, controlled study
by Boone et al in 1992 there was no statistical difference in
bladder capacity between a treated and a sham treated
group.18 In 1993 Lyne and Bellinger reported that 5 of
their 12 patients had sustained increases in bladder capacity following therapy.19 In 1994 Decter et al also reported an increase in age adjusted bladder capacity in a
similar percent of patients.20 The reasons for the variabil-
ity in success at different centers may be multifactorial. It

must be emphasized that the administration of bladder
stimulation therapy is operator dependent. Adjustments
must continuously be made during the course of therapy,
which may substantially affect the overall result and clinical outcome. Given these reasons, there is a definite
learning curve associated with this treatment modality.
In a multi-institutional trial we noted that with training
by our staff our promising results could be duplicated
elsewhere.14
Variability in success may also be related to the timing
and number of urodynamic studies. One urodynamic study
alone may misrepresent overall patient health. Improvement is often seen during the course of treatment. Occasionally success is not seen for a couple of years after therapy.
Also, patients may have a setback that responds to another
series of bladder stimulation treatments. This setback is
typically neurological in origin. A decrease during therapy is
usually transient and further improvement eventually occurs. Therefore, the trend of improvement is most important, not necessarily a single measurement of bladder capacity at a given time.
We observed that patients with neurogenic bladder demonstrate sustained improvement in bladder compliance after
intravesical electrotherapy. Although those with sacral level
lesions had the best sensation response, they did not have as
substantial an improvement in capacity and pressure as the
other subgroups. Patients with lumbar lesions had the best
response in regard to bladder compliance. The majority of
patients had myelomeningocele. The rest of the subgroups
were too small to determine a significant difference from this
larger group of patients.
Future studies should review the anticholinergic dose
throughout therapy to determine whether less medication is
needed with time. Also, there have been trials with home
therapy in Europe, which would improve the large time
commitment required with therapy. If similar improvements are seen with this therapy, we would consider its
usefulness at our institution.
CONCLUSIONS
Bladder stimulation is beneficial for improving bladder compliance in patients with neurogenic bladder dysfunction.
This technique is safe and effective for improving bladder
compliance. Some patients also have an improvement in
their ability to feel the bladder filling and be able to catheterize when feeling full rather than by the clock.

REFERENCES
1.
2.
Edelstein RA, Bauer SB, Kelly MD, Darbey MM, Peters CA,
Atala A et al: The long-term urological response of neonates
with myelodysplasia treated proactively with intermittent
catheterization and anticholinergic therapy. J Urol 1995;
154: 1500.
Kaefer M, Pabby A, Kelly M, Darbey M and Bauer SB: Improved bladder function after prophylactic treatment of the
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
15.
16.
17.
18.
19.
20.
high risk neurogenic bladder in newborns with

myelomeningocele. J Urol 1999; 162: 1068.
Wu HY, Baskin LS and Kogan BA: Neurogenic bladder dysfunction due to myelomeningocele: neonatal versus childhood treatment. J Urol 1997; 157: 2295.
Lendvay TS, Cowan CA, Mitchell MM, Joyner BD and Grady
RW: Augmentation cystoplasty rates at childrens hospitals
in the United States: a pediatric health information system
database study. J Urol 2006; 176: 1716.
Guven A, Onal B and Kogan BA: Spontaneous bladder perforations following augmentation cystoplasty in children.
Nat Clin Pract Urol 2006; 3: 584.
Metcalfe PD, Cain MP, Kaefer M, Gilley DA, Meldrum KK,
Misseri R et al: What is the need for additional bladder
surgery after bladder augmentation in childhood? J Urol
2006; 176: 1801.
Metcalfe PD, Casale AJ, Kaefer MA, Misseri R, Dussinger AM,
Meldrum KK et al: Spontaneous bladder perforations: a
report of 500 augmentations in children and analysis of
risk. J Urol 2006; 175: 1466.
Shekarriz B, Upadhyay J, Demirbilek S, Barthold JS and
Gonzalez R: Surgical complications of bladder augmentation: comparison between various enterocystoplasties in
133 patients. Urology 2000; 55: 123.
Soergel TM, Cain MP, Misseri R, Gardner TA, Koch MO and
Rink RC: Transitional cell carcinoma of the bladder following augmentation cystoplasty for the neuropathic bladder.
J Urol 2004; 172: 1649.
Kaplan WE and Richards I: Intravesical transurethral electrotherapy for the neurogenic bladder. J Urol 1986; 136: 243.
Kaplan WE and Richards I: Intravesical bladder stimulation in
myelodysplasia. J Urol 1988; 140: 1282.
Kaplan WE and Richards I: Intravesical transurethral bladder
stimulation to increase bladder capacity. J Urol 1989; 142:
600.
Cheng EY, Richards I and Kaplan WE: Use of bladder stimulation in high risk patients. J Urol 1996; 156: 749.
Cheng EY, Richards I, Balcom A, Steinhardt G, Diamond M,
Rich M et al: Bladder stimulation therapy improves bladder
compliance: results from a multi-institutional trial. J Urol
1996; 156: 761.
Katona F and Berenji M: Intravesical transurethral electrotherapy in meningomyelocele patients. Acta Paediatr Acad
Sci Hung 1975; 16: 363.
Madesbacher H, Pauer W and Reiner E: Rehabilitation of
micturition by transurethral electrostimulation of the bladder in patients with incomplete spinal cord lesions. Paraplegia 1982; 20: 191.
Janneck C: Electric stimulation of the bladder and the anal
sphinctera new way to treat the neurogenic bladder. Prog
Pediatr Surg 1976; 9: 119.
Boone TB, Roehrborn CG and Hurt G: Transurethral intravesical electrotherapy for neurogenic bladder dysfunction in
children with myelodysplasia: a prospective, randomized
clinical trial. J Urol 1992; 148: 550.
Lyne CJ and Bellinger MF: Early experience with transurethral electrical bladder stimulation. J Urol 1993; 150:
697.
Decter RM, Snyder P and Laudermilch C: Transurethral electrical bladder stimulation: a followup report. J Urol 1994;
152: 812.
EDITORIAL COMMENT
These authors highlight the experience at 1 institution with
treating a large group of children with neurogenic bladder
dysfunction using intravesical electrotherapy. It is clear from
the author long-term perspective that intravesical electrother-
1683
apy improves bladder capacity by greater than 20% in 77% of

the patients and it maintains acceptable detrusor pressure at
capacity along with minimal detrusor activity midway through
filling in approximately 74% to 81%. In addition, the sensation
of bladder fullness is attained and maintained in 62% of the
children which, as the authors point out, helps the child know
when to catheterize and empty the bladder.
These observations are important for determining the
efficacy of treatment in these children. However, the authors
have not addressed several other just as important issues.
Based on pretreatment testing and the author extensive
experience what children are not good candidates at all in
whom to even initiate this form of therapy? When do the
authors consider a patient refractory to stimulation sessions
and pursue an alternative course of treatment after 1 or
more therapy sessions? Conversely what response parameters encourage the authors to repeat the intravesical electrotherapy sessions after a period? How long are the effects
of therapy (detrusor muscle changes and sensory improvement) sustained and did each additional set of therapy sessions prolong or shorten these responses? After such an
extensive analysis of their patient population are there any
patients whom the authors would not consider likely to
respond to electrotherapy? Alternatively what patients
would they consider to be ideal candidates who would most
likely benefit from this treatment program?
In trying to advocate intravesical electrotherapy as a
means of avoiding augmentation cystoplasty it is paramount
that these questions be answered.
Stuart B. Bauer
Childrens Hospital Boston
Boston, Massachusetts
REPLY BY AUTHORS
In our long-term experience with intravesical electrotherapy
we have determined which patients may or may not respond
to therapy. The myelomeningocele population responds the
best to therapy, especially those with small, poorly compliant bladders. Other therapies have often failed in this group
who would otherwise require augmentation. Children
younger than 6 years typically have the most success with
sensation. Patients who have a poor response to therapy
and, therefore, are discouraged from this treatment modality include those with a large hypotonic bladder, severe
detrusor overactivity with frequent high pressure contractions, unstable spinal lesions, complete cord transections,
prior augmentations and no outlet resistance.
Responses are typically seen after 1 series of therapy. If
no response is seen on cystometrogram when the patients
return for the second series, continuation of the program is
discouraged. However, even a slight response is a predictor
of success. Complete success may take up to 3 years after
therapy has been finished.
Results are usually sustained for a lifetime. In patients
with close to 20 years of followup the improvements have not
subsided. These patients continue to have large capacity
compliant bladders with a degree of sensation. The exception is those who have neurological setbacks, and these
patients, if they responded previously, often improve with
another treatment series.
Predictive Factors of Early Spontaneous

Resolution in Children With Primary Vesicoureteral Reflux
Matthew J. Knudson, J. Christopher Austin, Zeb M. McMillan, Charles E. Hawtrey
and Christopher S. Cooper*
From the Division of Pediatric Urology, Department of Urology, University of Iowa, Iowa City, Iowa
Purpose: Questions regarding the efficacy and safety of long-term prophylactic antibiotics and the option of minimally
invasive therapy have increased clinical emphasis on predicting early resolution in patients with vesicoureteral reflux. We
reviewed multiple potential prognostic factors in addition to grade to enhance predictive ability regarding early vesicoureteral
reflux resolution.
Materials and Methods: Medical records and cystograms from 324 children (257 girls and 67 boys) with primary
vesicoureteral reflux were reviewed. Multiple factors were analyzed with respect to outcome and multivariable dependent
tables were constructed to enhance the prediction of vesicoureteral reflux resolution.
Results: Mean SD age at diagnosis was 2.3 2.1 years and followup was 1.7 2.8 years. Reflux grade was significantly
associated with resolution (p 0.001). Multivariate analysis stratifying by grade demonstrated that a bladder volume at
reflux onset on the initial cystogram of greater than 50% of predicted bladder capacity (p 0.001), age younger than 2 years
at diagnosis (p 0.003) and history of prenatal hydronephrosis (p 0.001) were significant factors predicting resolution
within 2 years. Multivariable tables using age, bladder volume at reflux onset and grade demonstrated that children younger
than 2 years at diagnosis and volume at onset greater than 50% of predicted bladder capacity had a higher early resolution
rate for grades 2 (p 0.012) and 3 (p 0.001) reflux.
Conclusions: Initial reflux grade, bladder volume at reflux onset, age at diagnosis and history of prenatal hydronephrosis
were shown to be independent factors affecting the resolution rates of vesicoureteral reflux. The multivariable tables permit
improved individualized prediction of resolution in patients with grades 2 and 3 reflux.
Key Words: bladder; vesico-ureteral reflux; surgical procedures, minimally invasive; ureter; antibiotic prophylaxis
ncreasing concern regarding the efficacy and side effects

of long-term antibiotic prophylaxis and the development
of minimally invasive treatments for reflux lead more
parents and physicians to consider early surgical treatment
when faced with the possibility that reflux may not spontaneously resolve for many years.1 Thus, an improved ability
to predict which children will achieve reflux resolution
within 1 or 2 years has become more important in the
clinical setting. Currently reflux grade, patient age and laterality are the major factors used to predict the probability
of spontaneous resolution.2 We previously reported an analysis of the records of 118 children, demonstrating bladder
volume at the onset of reflux as another independent factor
affecting resolution rates.3 To further assess predictive factors of early primary reflux resolution independent of grade
the records of 206 additional patients were analyzed.

We performed a retrospective chart review of 324 children,
including 67 males (21%) and 257 females (79%), who were
diagnosed with primary VUR at ages 0 months to 10 years
Supported by a Doris Duke Clinical Research Fellowship (MJK).

* Correspondence: Department of Urology, University of Iowa, 200
Hawkins Dr., 3 RCP, Iowa City, Iowa 52242-1089 (telephone: 319-3848299; FAX: 319-356-3900; e-mail: christopher-cooper@uiowa.edu).
0022-5347/07/1784-1684/0
and were evaluated from 1988 to 2004. Children were diagnosed by VCUG (300) or NCG (24). The first followup was
performed with VCUG in 82 cases and NCG in 218 at a
median SD of 1.1 0.9 years (range 0.1 to 7.7) after initial
diagnosis. End points were defined as spontaneous reflux
resolution or as unresolved if reflux was persistent at last
followup or the child underwent surgical correction.
Age, reflux grade, bladder volume at reflux onset and
laterality were recorded during the initial and second cystograms. A comparison of grade on the first and second cystograms was defined as worse, stable, improved or resolved.
For bilateral reflux to be considered improved each ureter
needed to decrease in grade. Reflux was graded in accordance with the International Reflux Study.4 Reflux on NCG
was recorded as being grade 1 confined to the ureter, grade
2reaching the renal pelvis without or with dilatation, and
grades 4/5massive dilatation of the ureter and pelvis.
Grades 4/5 were combined in our analysis because only 29
patients presented with this level of reflux and there was
difficulty distinguishing among them on NCG. In cases of
bilateral reflux of different grades the higher grade was used
for statistical calculations.
PBC was calculated for the age of each patient using the
formula, PBC [age 2] 30 ml. The volume at reflux
onset on the initial cystogram was known for 171 patients
and normalized by dividing by PBC to determine the percent
of PBC at reflux onset. Also recorded were presenting symp-
1684
Vol. 178, 1684-1688, October 2007

Printed in U.S.A.
DOI:10.1016/j.juro.2007.03.161
PREDICTIVE FACTORS OF EARLY REFLUX RESOLUTION

toms, including prenatal hydronephrosis in 67 patients, of
whom 66% were male, febrile UTI in 158, nonfebrile UTI in
105 and screening in 20, as well as the reason for the operation and the presence of breakthrough UTIs. Exclusion
criteria were ureterocele, ectopic ureter, megaureter, duplicated ureter, neurological or structural bladder abnormalities, a known history of voiding dysfunction or anticholinergic use. Children needed to have at least 1 year of followup
to be included in the study.
To identify factors that predict resolution data on the
initial visit were independently compared 1 and 2 years
after diagnosis as well as for overall outcome. Patients with
and without resolution were compared. To compensate for
variability in return followup visits we included any image
study within 0.4 years of the specified time frame. The
Fisher exact or chi-square test was used to compare categorical variables and the 2-sample t or Wilcoxon rank sum test
was used to compare continuous and ordinal variables. Cox
proportional hazard regression stratified by initial reflux
grade was performed to test for the association of multiple
variables with p 0.05 considered significant.
RESULTS
Median age at diagnosis for boys and girls was 0.2 (range 0
to 6.6) and 2.3 years (range 0 to 10.7), respectively. Of the
patients 193 (60%) achieved spontaneous resolution and 94
(29%) underwent surgical correction for reflux, while in 37
(11%) reflux persisted. Mean time from diagnosis to spontaneous resolution was 2.5 2.5 years vs 2.9 3.2 until
surgical correction. Of the 94 patients who underwent surgery 46 had a documented breakthrough infection with
other reasons for surgery, including parental preference/
failure to resolve in 18 and worsening renal function in 30.
Children with persistent reflux were followed an average of
4.6 2.7 years (range 1.0 to 11.5). Bilateral reflux was
initially present in 164 patients (51%).
The figure shows the resolution rate with time by reflux
grade. As expected, lower reflux grade significantly increased the chances of resolution at 1 and 2 years as well as
overall resolution (each p 0.001). Of 37 patients 33 (80%)
with grade 1, 113 of 165 (68%) with grade 2, 42 of 93 (45%)
1685
TABLE 1. Univariate analysis of resolved and unresolved reflux

stratified by grade
Vol at onset (greater vs less than

50% PBC):
No. pts
1-Yr outcome
2-Yr outcome
Overall
Prenatal hydronephrosis (vs UTI
or screening):
No. pts
1-Yr outcome
2-Yr outcome
Overall
Age per 1-year increase:
No. pts
1-Yr outcome
2-Yr outcome
Overall
Age (younger vs older than 2 yrs):
No. pts
1-Yr outcome
2-Yr outcome
Overall
Male (vs female):
No. pts
1-Yr outcome
2-Yr outcome
Overall
Unilat (vs bilat):
No. pts
1-Yr outcome
2-Yr outcome
Overall
HR (95% CI)
p Value
93
9.0 (2.829.3)
4.6
(2.68.2)
3.09 (2.004.78)
0.001*
0.001*
0.001*
41
3.8
(2.36.0)
2.9
(1.94.6)
2.39 (1.553.69)
0.001*
0.001*
0.001*
324
0.8
(0.70.9)
0.9
(0.81.0)
0.96 (0.891.03)
0.001*
0.003*
0.219
169
2.1
(1.43.2)
1.7
(1.22.4)
1.23 (0.921.64)
0.001*
0.003*
0.159
67
2.3
(1.53.5)
1.8
(12.64)
1.45 (1.022.07)
0.073
0.003*
0.040*
160
1.1
(0.71.7)
1.4
(1.02.0)
1.34 (0.991.80)
0.607
0.056
0.055
* Significant.
with grade 3 and 5 of 29 (17%) with grade 4/5 reflux had

resolution.
Table 1 demonstrates that an initial bladder volume at
reflux onset of greater than 50% PBC as well as presentation
with a history of prenatal hydronephrosis were the most
significant factors predicting resolution at all time points
(p 0.001). Age was also predictive of resolution at 1 and 2
years. We compared children younger vs older than 2 and
also compared age per 1-year increment (younger than 1, 1
to 2, 2 to 3, etc). Each result was significant for early resolution at 1 and 2 years (p 0.001 and 0.003, respectively).
Time to spontaneous resolution based on reflux grade
1686
However, age did not significantly affect overall resolution

outcomes in this analysis (p 0.219 and 0.159, respectively). Male gender was predictive at 2 years (p 0.003)
and overall (p 0.040), and it approached significance 1
year after diagnosis (p 0.073). Although unilateral reflux
was a strong independent factor for reflux resolution, it
approached but did not meet statistical significance for reflux resolution when stratified by grade at any time point (at
2 years p 0.056).
To further evaluate variables that were predictors at 1
and 2 years multivariate analysis was performed, controlling for grade. Volume at onset, male gender, age younger
than 2, prenatal hydronephrosis and unilateral reflux were
analyzed (table 2). Using the multivariate approach male
gender was no longer a significant predictor, while volume at
onset greater than 50% PBC, prenatal hydronephrosis and
age younger than 2 years remained significant predictors of
resolution 1 and 2 years after diagnosis.
The Wilcoxon rank sum test showed that the percent of
PBC at reflux onset as a continuous variable was predictive
of time to resolution (p 0.001). Volume at onset was placed
into observed quartiles (less than 30%, 30% to 50%, 50% to
90% and greater than 90% PBC). A significant difference
existed between the top 2 quartiles and the bottom quartile
(p 0.001). No difference existed between the bottom 2 quartiles (p 0.131) or between the top 2 quartiles (p 0.866).
Tables 3 and 4 show resolution rates for grades 2 and 3
reflux with respect to age at diagnosis and volume at reflux
onset. The percent of patients with resolution for each subcategory was compared to the corresponding resolution rate
based on grade alone (see figure). Statistically different resolution rates were noted. For grades 2 and 3 reflux children
younger than 2 years at diagnosis with a volume at onset of
greater than 50% PBC had higher resolution rates, while
children older than 2 years with less than 50% PBC had a
lower probability of resolution.
TABLE 2. Multivariate analysis of several factors to predict

resolution 1 and 2 years after diagnosis
Volume at onset (greater vs

less than 50% PBC):
No. pts
1-Yr outcome
2-Yr outcome
Age (younger vs older than
2 yrs):
No. pts
1-Yr outcome
2-Yr outcome
Male (vs female):
No. pts
1-Yr outcome
2-Yr outcome
(vs UTI or screening):
No. pts
1-Yr outcome
2-Yr outcome
Unilat (vs bilat):
No. pts
1-Yr outcome
2-Yr outcome
HR (95% CI)
p Value
93*
7.5 (2.324.7)
3.9 (2.17.0)
0.001*,
0.001
169
1.7 (1.02.7)
1.5 (1.02.2)
0.034
0.038
67
1.3 (0.72.3)
1.2 (0.81.9)
0.372
0.445
41
2.9 (1.75.1)
2.2 (1.33.6)
0.001
0.001
160
Not included
1.4 (1.02.0)
* In 171 patients with known bladder volume at onset.

Significant.
Not significant in
univariate
model
0.058
In 126 of 182 patients with persistent reflux on the second

cystogram bladder volume at reflux onset could be determined. Volume at onset as a percent of PBC on the second
cystogram was predictive of resolution in the 20 children
with greater than 75% PBC 2 years after the second study
compared to less than 75% (50% vs 19%, p 0.002) as well
as predictive of overall outcome (p 0.007). Unlike the first
cystogram, greater than 50% PBC at reflux onset on the
second cystogram did not attain statistical significance compared to less than 50% PBC (p 0.081). A change in the
percent of PBC at onset between the first and second cystograms was also compared but no significant difference was
noted (p 0.604). Finally, improvement in grade from the
first to the second cystogram occurred in 35 patients, while
109 were stable and 38 worsened. Patients with improvement in grade on the second cystogram had a significantly
higher resolution rate 2 years following the second image
(29% vs 14%, p 0.032). This approached significance for
overall outcome (p 0.058).
DISCUSSION
The effectiveness of minimally invasive surgery and concern
regarding prophylactic antibiotics has made early surgical
intervention more appealing to some clinicians and parents.5,6 However, with little evidence showing that surgery
promotes kidney health more than medical therapy and few
studies assessing long-term complications of injectable materials, promoting early surgical intervention for many patients does not seem prudent.712 Improved ability to predict
early reflux resolution on an individualized basis may assist
the parent and physician with VUR management decisions.
As expected, grade was significantly associated with resolution. When controlling for grade, volume at reflux onset,
patient age, gender and presentation with a history of prenatal hydronephrosis were independent predictive factors of
resolution. While these variables have been discussed in
previous literature, to our knowledge no investigators have
looked at all of them in the same set of patients and controlled for reflux grade.
Volume at onset was the strongest predictor of reflux
resolution at all time points after controlling for grade. Past
studies evaluated volume at onset with mixed results. In a
study of 133 children Mozley et al noted that those who did
not begin to experience reflux until the bladder was filled to
more than 60% of total bladder capacity were at smaller risk
of surgery than those who began to experience reflux at
smaller bladder volumes.13 Papachristou et al found that of
children younger than 1 year those with greater than 45%
bladder volume at onset had a better prognosis.14 Other
studies15,16 could not reproduce the conclusions of Mozley et
al.13 Our study uniquely used the initial cystogram determined volume at reflux onset normalized to age predicted
bladder capacity and controlled for grade of reflux.
We previously reported that volume at reflux onset relative to PBC was predictive of resolution at less than 75% of
PBC for grade 2 reflux.3 In the current study we observed
that volume at onset greater than 50% of PBC is statistically
significant for improved resolution compared to less than
50% of PBC for all reflux grades. Our data also show that, as
a continuous variable, volume at reflux onset is a positive
predictor of resolution.
1687
TABLE 3. Resolution rates and timing for grade 2 reflux based on age at diagnosis and volume at onset
Age at Diagnosis
Younger than 2 yrs (% resolution):
No. pts
1 Yr
2 Yrs
Overall
Older than 2 yrs:
No. pts
1 Yr
2 Yrs
Overall
Onset Vol
Greater Than 50% PBC
p Value vs Time/
Grade Specific Resolution
Onset Vol
Less Than 50% PBC
p Value vs Time/
46
65
77
92
0.011*
0.012*
0.013*
34
37
39
44
0.870
0.348
0.037
46
60
75
85
0.069
0.039*
0.137
34
7
7
47
0.013
0.001
0.080
* Significantly above average time/grade specific resolution rate (see figure).

Significantly below average time/grade specific resolution rate (see figure).
Age at diagnosis was previously shown to be a factor in

spontaneous resolution. The 1997 VUR guidelines summary
report broke down reflux by ages 0 to 2, 3 to 5 and 6 to 10
years, and showed that for grade 3 these age groups had a
significantly different chance of resolution with younger
groups spontaneously resolving at a higher rate.2 Greenfield
et al did not find that resolution rates correlate with age for
low grade VUR.17 We found that children younger than 2
years had a higher 1 and 2-year resolution rate than older
children when controlling for all grades. Likewise, increasing age by 1-year increments statistically decreases the
chance of early resolution. However, the overall resolution
outcome in younger children was not statistically different
than that in older children. The reason for this pattern is
unclear but we offer 2 hypotheses, including 1) the longer
reflux takes to resolve, the more likely it is that surgical bias
will impact the results and 2) children diagnosed at a young
age are more likely to have transient reflux that resolves
rapidly but after that early period they have a chance of
resolution similar to that in older children.
Improved VUR resolution presenting with prenatal hydronephrosis was previously reported.18 Herndon et al found
that even patients with high grade bilateral reflux have a
high resolution rate.19 Our results support the finding that
children presenting with a history of prenatal hydronephrosis achieve resolution at a higher rate than other children
even when controlling for grade as well as age.
Male gender is associated with improved resolution.20
However, our multivariate analysis incorporating bladder
volume at reflux onset, age, prenatal hydronephrosis and
laterality did not demonstrate male gender as independently associated with resolution (table 2). This may reflect
the fact that male patients on average are diagnosed at a

younger age and they present more frequently with prenatal
hydronephrosis. Thus, gender is not a significant factor affecting resolution when controlling for these factors.
Laterality was previously shown to be a predictor of resolution in grade 3 reflux.2 Schwab et al also found that
unilateral reflux resolved more rapidly when it was lower
grade20 but other reports do not support this finding.17
While unilateral reflux in our study resolved at a higher
rate, unilateral reflux approached but did not attain statistical significance when controlling for grade. Our data suggest that laterality does not have as great an impact on
resolution rates as other factors.
By combining grade, age and volume at reflux onset we
noted significantly different resolution rates than by using
grade alone (tables 3 and 4). Using these additional factors
permits more individualized predictions for the chance of
early and overall resolution than grade alone. Due to a
limited number of patients neither history of prenatal hydronephrosis nor laterality was incorporated into the tables
but they are factors that should be considered when assessing the chance of reflux resolution in a patient.
Data on the second imaging study show that volume and
to a limited extent grade improvement predict the timing of
resolution. The 2 factors predicted increased resolution 2
years after the second imaging study. This information
would be useful when counseling patients regarding the
chance for subsequent resolution. Often the first imaging
study is VCUG and volume at onset is not recorded but it is
noted on followup NCG.
The retrospective nature of our study created limitations.
Since the decision regarding surgical intervention was based
TABLE 4. Resolution rates and timing for grade 3 reflux based on age at diagnosis and volume at onset
Age at Diagnosis
Younger than 2 yrs (% resolution):
No. pts
1 Yr
2 Yrs
Overall
Older than 2 yrs:
No. pts
1 Yr
2 Yrs
Overall
Onset Vol
Greater Than 50% PBC
p Value vs Time/
Onset Vol
Less Than 50% PBC
p Value vs Time/
18
80
80
90
0.001*
0.001*
0.007*
29
7
14
29
0.304
0.344
0.234
18
25
63
75
0.644
0.030*
0.113
29
0
0
27
0.072
0.025
0.172
* Significantly above average time/grade specific resolution rate (see figure).

Significantly below average time/grade specific resolution rate (see figure).
1688
on physician and parent choice, some patients underwent

surgery and they may have spontaneously achieved resolution given more time. The relatively large number of patients in our series who were followed by urologists and
nephrologists may have decreased this potential surgeon
bias. Average time to operation was longer than the average
resolution time, which may also have decreased this potential bias. A potential factor affecting reflux resolution is
voiding dysfunction. This factor was not routinely screened
with urodynamics. We attempted to screen for voiding dysfunction during our medical record review by excluding patients with a known history of voiding dysfunction, anticholinergic use or wetting beyond the expected age of
continence. However, it remains possible that some patients
with abnormal bladder function were not recognized and
included in our analysis. Another potential limitation relates to the fact that in most patients the initial study was
VCUG and the followup studies were NCGs. The ability to
accurately grade reflux on NCG is not as precise, given the
decreased anatomical detail, which may interfere with the
accurate determination of small degrees of improvement in
reflux.
4.
5.
6.
7.
8.
9.
10.
11.
CONCLUSIONS
Given the treatment options now available, predicting early
reflux resolution has become increasingly important. At
VUR diagnosis using a combination of grade, age, volume at
reflux onset and presentation with a history of prenatal
hydronephrosis improves predictive ability regarding early
reflux resolution. Combining these individual factors may
improve the decision making process regarding reflux management.
12.
13.
14.

NCG
PBC
UTI
VCUG
VUR
nuclear cystogram
predicted bladder capacity
urinary tract infection
voiding cystourethrogram
15.
16.
REFERENCES
1.
Garin EH, Olavarria F, Garcia Nieto V, Valenciano B, Campos

A and Young L: Clinical significance of primary vesicoureteral reflux and urinary antibiotic prophylaxis after
acute pyelonephritis: a multicenter, randomized, controlled
study. Pediatrics 2006; 117: 626.
2. Elder JS, Peters CA, Arant BS Jr, Ewalt DH, Hawtrey CE,
Hurwitz RS et al: Pediatric Vesicoureteral Reflux Guidelines Panel summary report on the management of primary
vesicoureteral reflux in children. J Urol 1997; 157: 1846.
3. McMillan ZM, Austin JC, Hawtrey CE, Knudson MJ and
Cooper CS: Bladder volume at onset of reflux on initial
cystogram predicts spontaneous resolution. J Urol 2006;
176: 1838.
17.
18.
19.
20.
Lebowitz RL, Olbing H, Parkkulainen KV, Smellie JM and

Tamminen-Mobius TE: International system of radiographic grading of vesicoureteric reflux. International Reflux Study in Children. Pediatr Radiol 1985; 15: 105.
Capozza N, Lais A, Nappo S and Caione P: The role of endoscopic treatment of vesicoureteral reflux: a 17-year experience. J Urol 2004; 172: 1626.
Lackgren G, Wahlin N, Skolenberg E and Stenberg A: Longterm followup of children treated with dextranomer/hyaluronic acid copolymer for vesicoureteral reflux. J Urol 2001;
166: 1887.
A prospective trial of operative versus nonoperative treatment
of severe vesico-ureteric reflux: 2 years observation in 96
children. Birmingham Reflux Study Group. Contrib Nephrol 1984; 39: 169.
Prospective trial of operative versus non-operative treatment
of severe vesicoureteric reflux in children: five years observation. Birmingham Reflux Study Group. Br Med J 1987;
295: 237.
Duckett JW, Walker RD and Weiss R: Surgical results: International Reflux Study in ChildrenUnited States branch.
J Urol 1992; 148: 1674.
Smellie JM: Commentary: management of children with severe vesicoureteral reflux. J Urol 1992; 148: 1676.
Jodal U, Koskimies O, Hanson E, Lohr G, Olbing H, Smellie J
et al: Infection pattern in children with vesicoureteral reflux randomly allocated to operation or long-term antibiotic
prophylaxis. The International Reflux Study in Children.
J Urol 1992; 148: 1650.
Knudson MJ, Cooper CS, Block CA, Hawtrey CE and Austin
JC: Calcification of glutaraldehyde cross-linked collagen in
bladder neck injections in children with incontinence: a
long-term complication. J Urol 2006; 176: 1143.
Mozley PD, Heyman S, Duckett JW, Turek P, Snyder H, Maislin G et al: Direct vesicoureteral scintigraphy: quantifying
early outcome predictors in children with primary reflux.
J Nucl Med 1994; 35: 1602.
Papachristou F, Printza N, Doumas A and Koliakos G: Urinary
bladder volume and pressure at reflux as prognostic factor
of vesicoureteral reflux outcome. Pediatr Radiol 2004; 34:
556.
Barthold JS, Martin-Crespo R, Kryger JV and Gonzalez R:
Quantitative nuclear cystography does not predict outcome
in patients with primary vesicoureteral reflux. J Urol 1999;
162: 1193.
Cooper CS, Madsen MT, Austin JC, Hawtrey CE, Gerard LL
and Graham MM: Bladder pressure at the onset of vesicoureteral reflux determined by nuclear cystometrogram.
J Urol 2003; 170: 1537.
Greenfield SP, Ng M and Wan J: Resolution rates of low grade
vesicoureteral reflux stratified by patient age at presentation. J Urol 1997; 157: 1414.
Brophy MN, Austin PE, Yan Y and Coplen DE: Vesicoureteral
reflux and clinical outcomes in infants with prenatally detected hydronephrosis. J Urol 2002; 168: 1716.
Herndon CD, McKenna PH, Kolon TF, Gonzales ET, Baker LA
and Docimo SG: A multicenter outcomes analysis of patients with neonatal reflux presenting with prenatal hydronephrosis. J Urol 1999; 162: 1203.
Schwab CW Jr, Wu HY, Selman H, Smith GH, Synder HM 3rd
and Canning DA: Spontaneous resolution of vesicoureteral
reflux: a 15-year perspective. J Urol 2002; 168: 2594.
Ipsilateral Ureteroureterostomy in the Surgical

Management of the Severely Dilated Ureter in Ureteral Duplication
Job K. Chacko, Martin A. Koyle, Gerald C. Mingin and Peter D. Furness, III*
From the Department of Pediatric Urology, Childrens Hospital and Department of Surgery, University of Colorado Health Science Center,
Denver, Colorado
Purpose: Ipsilateral ureteroureterostomy for the surgical management of severely dilated ureter in ureteral duplication is
well supported in the surgical literature but often not done. We evaluated our institutional experience with ureteroureterostomy in duplication anomalies to assess the feasibility and success of this procedure.
Materials and Methods: An 8-year retrospective review of the records of all patients with complete renal duplex anomalies
was evaluated. Anatomical presentations, and operative and nonoperative treatment of these patients were evaluated.
Results: A total of 193 patients were identified with complete renal duplication. Associated anomalies included ureterocele
in 24 patients, ectopic ureter in 38 and vesicoureteral reflux in 57. Of 193 patients 160 (83%) with duplex anomalies
underwent surgical intervention with a total of 41 ureteroureterostomies performed in 39 patients with dilated donor ureters.
A total of 11 ureteroureterostomies were performed primarily and 30 were performed in conjunction with ipsilateral ureteral
reimplantation of the distal common segment below the ureteroureterostomy. Ten of the 39 patients had the contralateral
side reimplanted for vesicoureteral reflux. In all children with ureteroureterostomy the anastomosis between the 2 ureters
remained patent. Two of the 11 children who underwent ureteroureterostomy alone had de novo ipsilateral vesicoureteral
reflux (1), which was treated with ureteral reimplantation, and subureteral injection (1). Two children who underwent
concomitant ureteroureterostomy and reimplantation without indwelling stents had transient postoperative urinomas that
required subsequent drainage. Additionally, 3 patients had persistent ipsilateral vesicoureteral reflux, which was treated
with subureteral injection in 1 and observation in 2. One patient presented with transient ipsilateral urinary obstruction,
which required percutaneous drainage and resolved spontaneously.
Conclusions: In cases of ureteral duplication with a severely dilated ureter requiring surgical intervention ipsilateral
ureteroureterostomy is a viable option for reflux and/or obstruction. The procedure is rapid and technically feasible, and it
offers excellent cosmesis. In addition, ureteroureterostomy has minimal morbidity and it facilitates early hospital discharge.
Key Words: abnormalities, ureter, stomas, ureterocele, vesico-ureteral reflux
ignificant hydronephrosis associated with ureteral dilatation is not an uncommon congenital anomaly, especially with the use of prenatal ultrasonography.1
Often the finding of ureteral dilatation is associated with
duplicated systems with ureteral obstruction and/or VUR.
The anomaly of complete ureteral duplication occurs in approximately 1/125 individuals.2 Obstructive causes of ureteral dilatation are most often related to ureterocele or ectopic ureteral insertion anomalies. Reflux as a cause of
ureteral dilatation stems from primary or secondary VUR.
Surgery for severe ureteral dilatation may be required for
recurrent urinary tract infections, renal obstruction or urinary incontinence due to ureteral ectopia. There are various
treatment options available for the severely dilated ureter
associated with a duplicated system, including heminephrectomy, pyelopyelostomy, ureteropyelostomy, ipsilateral
U-U and common sheath reimplantation.35 In the past
there was some concern about performing U-U with resultant donor ureter-to-recipient ureter drainage patterns, the
* Correspondence: Department of Pediatric Urology, Childrens

Hospital and University of Colorado, School of Medicine, 1056 East
19th Ave., B-463, Denver, Colorado 80218 (FAX: 303-864-5572;
e-mail: Furness.Peter@tchden.org).
0022-5347/07/1784-1689/0
so-called yo-yo effect. It was speculated that this yo-yo effect

could place kidney function at risk and it might not ensure
adequate drainage. However, this has not proved to be clinically significant.4,6 8 We present our experience with ipsilateral U-U for the severely dilated ureter in complete ureteral duplication anomalies.

A retrospective review was performed to identify patients
with complete renal duplex anomalies who were evaluated
at our institution between January 1998 and November
2005. Modes of presentation, associated anomalies and
treatment modalities were noted. We then focused our review on patients with ureteral duplication anomalies with
ureteral dilatation and grade 3 4 hydronephrosis who underwent U-U. Surgical outcomes in all children following
U-U were assessed with RUS with or without nuclear scan,
or with excretory urogram.
The figure shows the technique of ipsilateral U-U. The
patient is placed supine on the operating table. Most cases
can be performed through a modified Gibson or Pfannenstiel
incision. Based on imaging if there is preoperative suspicion
of 2 dilated ureters, cystoscopic stent placement is done
1689
Vol. 178, 1689-1692, October 2007

Printed in U.S.A.
DOI:10.1016/j.juro.2007.05.098
1690
URETEROURETEROSTOMY FOR DILATED URETER IN URETERAL DUPLICATION
A, ureteral complex is identified through modified Gibson or Pfannenstiel incision. B, donor ureter is ligated and excess distal ureter is
excised. C, recipient ureter is incised to correspond with donor ureter diameter. D, stay sutures are placed. E, U-U anastomosis is performed.
F, completed U-U.
before incision to help identify the UP and LP ureters. In

cases of only 1 enlarged ureter an intraoperative stent is
placed before completing the anastomosis. The stent is usually placed across the anastomosis.
Ureteral complex identification is facilitated by focal dissection along the bladder wall just inferior to the obliterated
umbilical artery. Each ureter is identified and then isolated
above the common distal blood supply. The ureteral length
needed to perform the anastomosis is obtained proximal on
the ureters to minimize perivesical dissection and its associated autonomic innervation. The dilated donor ureter is
then transected diagonally and ligated distally. The ureteral
stump is partially excised as much as possible with care
taken not to interrupt the common distal blood supply. The
recipient ureter is then opened lengthwise to match the
diameter of the donor ureter. The ureteral anastomosis is
performed end to side with 7-zero polydioxanone absorbable
suture. Extravesical ureterocele excision and recipient ureter reimplantation can be performed through this incision
since surgery is at the bladder level. The ureterocele is
dissected from the detrusor by dissecting along the ureteral
wall until the confluence of the ureterocele is identified with
the intravesical trigonal/bladder neck mucosa. The entire
ureteral stump and associated ureterocele can then be excised and the mucosa defect can be reconstructed to receive
a UP ureter for reimplantation. With the described careful
dissection extravesical ureteral reimplantation can be performed to avoid anterior cystotomy. If a Penrose drain is
placed, it is usually brought out inferior to the incision.
RESULTS
During the study period 193 patients were diagnosed with
complete renal duplex anomalies, of whom 142 (73%) with
duplex anomalies were treated surgically, including 39 of
160 (24%) who underwent a total of 41 U-Us. There were 8
males and 31 females with a mean age of 31 months (range
3 months to 15 years). Mean SD operative time was 124
48.1 minutes. Mean hospital stay was 1 day for all patients.
Other surgical procedures performed in patients with ureteral duplication without concurrent U-U were common
sheath ureteral reimplantation in 64, UP heminephrectomy
in 8, LP heminephrectomy in 2, pyelopyelostomy in 6, ure-
terocele puncture only in 12 and subureteral injection of the

refluxing duplex system in 12.
In the 39 patients identified a total of 41 U-Us were
performed. Two patients were treated with bilateral U-Us. A
total of 11 U-Us alone were performed. Five patients had an
ectopic UP ureter that underwent UP-to-LP U-U. Three
patients had a UP ureterocele, for which UP-to-LP U-U was
done. In 1 patient the ureterocele was incised before U-U.
The remaining 3 patients underwent LP-UP U-U for LP
VUR, including 2 who also underwent contralateral reimplantation for VUR.
The remaining 28 patients underwent a total of 30 ipsilateral U-Us with ipsilateral reimplantation. Six patients
underwent ureterocele incision with postoperative VUR and
were treated with U-U and ipsilateral reimplantation. One
of these patients underwent bilateral U-U with reimplantation. Eight patients with ectopic UP and VUR of the LP
underwent UP-LP U-U with reimplantation. In 1 patient
ectopic UP and LP VUR were treated with LP-UP U-U with
reimplantation. Six patients had an UP ureterocele and LP
reflux, which were treated with UP-to-LP U-U with ipsilateral reimplantation. Of note, 1 patient received bilateral
treatment with U-U and ipsilateral reimplantation. Five
patients had VUR of the UP and LP, which was treated with
U-U and ipsilateral reimplantation. Ten of these 27 patients
had contralateral VUR, which was treated with concomitant
contralateral reimplantation (see table).
Two of the 11 children with U-U alone showed de novo
ipsilateral VUR, which was treated with ureteral reimplantation and subureteral injection in 1 each. Of the children
who underwent concomitant U-U and reimplantation 2
without indwelling stents had transient postoperative urinomas, which required subsequent drainage. Additionally, 2
Diagnosis
Ureterocele
Ureterocele incision
Ectopic UP
LP VUR
UP LP VUR
Ectopic UP LP VUR
Totals
Contralat reimplantation
No. U-U
No. U-U Reimplantation
2
1
5
3
0
0
7
9
0
0
3
11
11
10
30
2

patients had persistent ipsilateral VUR, which was treated
with subureteral injection and observation in 1 each. One
patient presented with transient ipsilateral urinary obstruction, which required percutaneous drainage and resolved
spontaneously. Mean followup following U-U was 12 months
(range 3 to 34). In all children with U-U the anastomosis
between the 2 ureters remained patent. All children showed
improvement of dilatation on postoperative RUS.
DISCUSSION
Ureteral duplication occurs in approximately 1/112 individuals in the general population with a slight predominance in
females. Unilateral duplication of the ureter is 3 times more
common than bilateral duplication.2 There is a 12% incidence in the progeny of an affected parent.9 It was suggested
that this anomaly is associated with an autosomal dominant
gene with incomplete penetrance.10 Ureteral duplication is
often diagnosed in children investigated for urinary tract
infection or in females who present with urinary incontinence.11 It may also be discovered incidentally during abdominal US, including routine maternal US.1 VUR in the LP
segment in ureteral duplication is the most common problem
observed.12 The terminal location of the UP ureter may be
ectopic or associated with a ureterocele. Either anomaly may
cause reflux and/or obstruction, and may require surgical
intervention, particularly in cases of recurrent urinary tract
infection or renal function loss.
Conventional treatment in patients with ureteral duplication with an obstructed nonfunctioning UP renal moiety is
UP heminephrectomy. The argument for this procedure is
based on autopsy findings of dysplasia in these units, as
described by Mackie and Stephens.13 However, Monfort et al
noted only inflammatory and obstructive changes without
dysplasia in 57% of UP kidney specimens.3 These findings
were further confirmed by Smith et al, who also found no
dysplasia in 57% of patients, of whom 28% had almost normal renal tissue with mild or minimal inflammation.14
Concern was raised that, if the nonfunctioning moiety is
not excised, it would lead to an increased risk of hypertension, proteinuria and malignancy, leading to the historical
recommendation for heminephrectomy.15 However, the upper tract ablative procedure is not without potential morbidity since there is a 5% incidence of loss of LP moiety function
in complete renal duplication following UP heminephrectomy for ureteroceles.16 Additionally, recent long-term followup studies after surgery on the upper tract show that
64% to 70% of cases ultimately require repeat major lower
tract surgery, most commonly due to persistent reflux, when
associated with ureteroceles.17,18
Based on these observations Gran et al suggested that
upper tract ablative surgery does not always provide definitive cure of the original lower urinary tract pathology that
gave rise to the nonfunctional segment because there is
associated LP reflux in 75% of cases.19 They suggested a
lower tract approach to address these issues and recommended single or staged lower tract reconstruction, consisting of ureterocele excision, bladder neck reconstruction with
ureteral tapering and reimplantation, leaving the nonfunctioning upper moiety in situ.19 Our approach uses a similar
rationale, in that severely dilated ureter pathology in ureteral duplication anomalies exists at the lower tract level.
Our technique differs since one can avoid significant ure-
1691
teral tailoring at a single corrective procedure. The challenging ureteral tapering with bulky duplex reimplantation and
necessary additional anterior cystotomy are avoided by our
approach. U-U can be performed regardless of ureteral disparity with extravesical reimplantation through the detrusor incision made for the ureterocele excision, if present.
Ipsilateral U-U in children was first reported by Buchtel
in 1965 for single ureteral disease in a duplex system.20
Initially this procedure was not popular because there was
concern about creating ureteroureteral yo-yo reflux into the
normal ureter and, thus, theoretically putting the recipient
renal segment at risk.4,21 However, more recently groups
recommended U-U for duplex system.5 8 Our series would
support this idea, in that in all successful U-U surgeries all
donor ureters had a decreased hydroureter and there was no
evidence of recipient ureteral dilatation on followup abdominal US.
In our experience ipsilateral U-U can be easily performed
through a less than 4 cm Pfannenstiel or modified Gibson
incision. We believe that flank incision or a larger Pfannenstiel
incision in not necessary to perform U-U and our approach
potentially decreases postoperative discomfort and morbidity. The majority of our patients were discharged home following an overnight stay in the hospital. To facilitate
perivesical dissection and the location of the pathological
ureter a ureteral stent can be endoscopically placed in the
recipient ureter before incision. We have not noted a luminal
disparity issue, which was proposed as a problematic finding
when performing U-U.20 We have found that a luminal
disparity as high as 4 to 6-fold could be successfully addressed with U-U. However, we would have a low threshold
to place a stent in cases of a large luminal disparity.
We have not observed any complications related to U-U
per se, which is comparable to other series in which the U-U
success rate is between 94% and 98%.4 8 Followup investigations, including RUS and excretory urogram, have shown
resolution of hydronephrosis without any stasis at the U-U.
Some groups would recommend performing ureteropyelostomy or UP heminephrectomy in similar situations through
the flank incision but we believe that the small Pfannenstiel
incision causes less discomfort and morbidity. It also allows
us to change our intraoperative decision, including concurrent ureteral reimplantation for VUR or incidentally found
vesicoureteral junction obstruction of the recipient ureter.
CONCLUSIONS
Ipsilateral U-U can be done for various conditions associated
with dilated ureteral duplication, including LP reflux, an
obstructed UP with or without reflux and ectopic UP ureteral insertion. A small Gibson or Pfannenstiel incision
causes minimal morbidity and postoperative discomfort.
U-U is a single surgical procedure for the severely dilated
ureter in ureteral duplication anomalies. U-U allows one to
avoid significant ureteral tailoring and bulky duplex reimplantation as well as the additional anterior cystotomy used
in other techniques. In our experience ipsilateral U-U is a
successful and relatively simple technique for treating the
severely dilated ureter in ureteral obstruction or reflux in
children with ureteral duplication. Concerns about the discrepancy in size of the ectopic and recipient ureters, and
ureteroureteral yo-yo reflux have not been substantiated.
1692

LP
RUS
UP
US
U-U
VUR
lower pole
renal US
upper pole
ultrasound
ureteroureterostomy
9.
10.
11.
12.
REFERENCES
Jee LD, Rickwood AM, Williams MP and Anderson PA: Experience with duplex system anomalies detected by prenatal
ultrasonography. J Urol 1993; 149: 808.
2. Nation EF: Duplication of the kidney and ureter: a statistical
study of 230 new cases. J Urol 1944; 51: 456.
3. Monfort G, Guys JM, Coquet M, Roth K, Louis C and Bocciardi
A: Surgical management of duplex ureteroceles. J Pediatr
Surg 1992; 27: 634.
4. Huisman TK, Kaplan GW, Brock WA and Packer MG: Ipsilateral ureteroureterostomy and pyeloureterostomy: a review
of 15 years of experience with 25 patients. J Urol 1987; 138:
1207.
5. Choi H and Oh SJ: The management of children with complete
ureteric duplication: the use of ureteroureterostomy as a
primary and salvage procedure. BJU Int 2000; 86: 508.
6. Bochrath JM, Maizels M and Firlit CF: The use of ipsilateral
ureteroureterostomy to treat vesicoureteral reflux or obstruction in children with duplex ureters. J Urol 1983; 129:
543.
7. Bieri M, Smith CK, Smith AY and Borden TA: Ipsilateral
ureteroureterostomy for single ureteral reflux or obstruction in a duplicate system. J Urol 1998; 159: 1016.
8. Lashley DB, McAleer IM and Kaplan GW: Ipsilateral ureteroureterostomy for the treatment of vesicoureteral reflux or
13.
1.
14.
15.
16.
17.
18.
19.
20.
21.
obstruction associated with complete ureteral duplication.

J Urol 2001; 165: 552.
Whitaker J and Danks DM: A study of the inheritance of
duplication of the kidneys and ureters. J Urol 1966; 95: 176.
Atwell JD, Cook PL and Howell CJ: Familial incidence of bifid
and double ureters. Arch Dis Child 1974; 49: 390.
Williams DI: The ectopic ureter: diagnostic problems. Br J Urol
1980; 44: 267.
Kaplan WE, Nasrallah P and King LR: Reflux in complete
duplication in children. J Urol 1978; 120: 220.
Mackie GG and Stephens FD: Duplex kidneys: a correlation of
renal dysplasia with position of the ureteral orifice. J Urol
1975; 114: 274.
Smith FL, Ritchie EL, Maizels M, Zaontz MR, Hseuh W,
Kaplan WE et al: Surgery for duplex kidneys with ectopic
ureters: ipsilateral uretero-ureterostomy versus polar nephrectomy. J Urol 1989; 142: 532.
Husmann DA: Renal dysplasia: the risks and consequences of
leaving dysplastic tissue in situ. Urology 1998; 52: 533.
Decter RM, Roth DR and Gonzales ET: Individualized treatment of ureteroceles. J Urol 1989; 142: 535.
Husmann D, Strand B, Ewalt D, Clement M, Kramer S and
Allen T: Management of ectopic ureterocele associated with
renal duplication: a comparison of partial nephrectomy and
endoscopic decompression. J Urol 1999; 162: 1406.
Caldamone AA, Snyder HM III and Duckett JW: Ureteroceles
in children: followup of management with upper tract approach. J Urol 1984; 131: 1130.
Gran CD, Kropp BP, Cheng EY and Kropp KA: Primary lower
urinary tract reconstruction for nonfunctioning renal moieties associated with obstructing ureteroceles. J Urol 2005;
173: 198.
Buchtel HA: Uretero-ureterostomy. J Urol 1965; 93: 153.
Lenaghan D: Bifid ureters in children: an anatomical, physiological and clinical study. J Urol 1962; 87: 808.
Penile Curvature
Outcome Analysis of Severe Chordee Correction Using
Tunica Vaginalis as a Flap in Boys With Proximal Hypospadias
Luis H. P. Braga, Joao L. Pippi Salle, Sumit Dave, Darius J. Bagli, Armando J. Lorenzo and
Antoine E. Khoury*
From the Division of Urology, Hospital for Sick Children, Toronto, Ontario, Canada
Purpose: There is ongoing controversy regarding optimal treatment for severe ventral curvature. It has been suggested that
ventral corporeal lengthening may be associated with recurrent curvature and erectile dysfunction. To further assess these
issues we reviewed our experience with ventral penile lengthening for correcting the severe ventral curvature associated with
proximal hypospadias.
Materials and Methods: We reviewed the records of 38 boys with severe hypospadias and congenital ventral curvature
greater than 45 degrees who were treated at our institution from 1995 to 2004 with placement of a flap or graft in the
corporeal bodies to straighten the phallus. Of the patients 21 had perineal and 17 had penoscrotal hypospadias, including 22
with associated penoscrotal transposition and/or bifid scrotum and 6 with ambiguous genitalia. Testosterone stimulation
before surgery was given in 11 children at surgeon discretion.
Results: Median age at surgery was 15 months. The urethral plate was divided in 94.7% of patients. A tunica vaginalis flap
was used alone in 23 cases and associated with dura, pericardium or small intestinal submucosa in 8, 2 and 1, respectively.
The remaining 4 patients underwent ventral grafting alone, including lyophilized dura in 1, pericardium in 1 and dermis in
1. Urethral reconstruction was achieved by the transverse island flap technique or 1 of its modifications in 34 children. Four
boys underwent a 2-stage procedure. Followup available on 35 of 38 patients was 1 to 11 years (median 5.3). Recurrent ventral
curvature in 5 of 35 patients was mild in 1 and clinically significant, requiring re-intervention, in 4. Four of 9 patients (44.4%)
who underwent corporeal grafting with lyophilized dura had recurrent ventral curvature vs 1 of 23 (4.3%) who had a tunica
vaginalis flap (chi-square 5.14, p 0.02). At last followup straight erections were documented by patients and/or parents in
30 of 35 children (85.7%).
Conclusions: The short-term outcome of ventral penile lengthening using tunica vaginalis flap alone for correcting severe
chordee is favorable with a 95% success rate. Dural grafts were associated with a higher risk of recurrent ventral curvature
compared to tunica vaginalis flaps. Although most of our patients were not yet adults, when chordee and erectile dysfunction
may become apparent, we believe that tunica vaginalis flap repair is a good option for correcting severe ventral curvature.
Key Words: penis, ureter, hypospadias, transplants, abnormalities, surgical flaps
orrecting VC is an important step in the surgical

management of hypospadias. Several surgical techniques have been described to correct VC, including
dorsal plication, corporeal rotation, urethral plate mobilization and ventral penile lengthening.13 Different materials
can be grafted to achieve ventral penile lengthening, such as
dermis, SIS, tunica vaginalis, dura and pericardium.4 6
TVF is an attractive alternative for bridging the corporeal
defect and straightening the phallus due to its easy availability and better vascular supply.7
The decision to repair severe VC with placement of a graft
or flap in the corpora cavernosa is based on the concept that
there is a significant disproportion between the ventral and
dorsal aspects of the corporeal bodies. It seems logical to
lengthen the ventral aspect of the corpora rather than reduce the size of an already small phallus by dorsal plication.
However, prior reports have suggested that placing grafts in
the corpora to straighten the penis is associated with a high
Study received institutional research ethics board approval.

* Correspondence: Division of Urology, Hospital for Sick Children,
Toronto, Ontario, Canada M5G 1X8 (e-mail: tony.khoury@sickkids.ca).
0022-5347/07/1784-1693/0
failure rate secondary to recurrent curvature.8 Furthermore, although it is not well established, concern regarding
erectile dysfunction has been raised. These potential complications prompted us to review our experience with ventral
penile lengthening for correcting the severe VC associated
with penoscrotal or perineal hypospadias.
A total of 2,534 boys with hypospadias were treated at our
institution between 1995 and 2004. We identified 38 children with penoscrotal or perineal hypospadias associated
with severe VC who had a graft or flap placed in the corpora
to straighten the phallus during this 10-year span. Operative notes and ambulatory clinical charts were retrospectively reviewed after obtaining permission from the research
ethics board at our hospital. Severe VC was defined intraoperatively by the erection test9 as curvature greater than
45 degrees (fig. 1). A total of 11 patients received preoperative testosterone stimulation when penile size was too small
according to surgeon judgment and preference.
VC correction was performed with ventral corporeal
lengthening procedures using TVF alone or associated with
1693
Vol. 178, 1693-1697, October 2007

Printed in U.S.A.
DOI:10.1016/j.juro.2007.03.166
1694
TUNICA VAGINALIS FLAP FOR SEVERE CHORDEE
FIG. 1. Erection test shows 90-degree ventral curvature

FIG. 3. Procedures used to lengthen ventral corporeal aspect
grafts. Different materials were used for grafting, such as

dermis, pericardium, SIS and dura. Grafts were used alone
in patients who had undergone prior bilateral orchiopexy
because tunica vaginalis was unavailable. In the other cases
grafts were covered by a TVF. Briefly, the urethral plate was
completely mobilized off of the underlying corpora cavernosa. A transverse incision was made at the point of greatest
curvature through the tunica albuginea, extending from the
3 to the 9 oclock position (fig. 2). Careful distal and proximal
dissections were carried out to lift the tunica albuginea off of
the underlying delicate vascular tissue. By doing so the
linear incision expanded to become an ellipsoid defect. The
defect was closed with a graft alone (pericardium, dermis,
SIS or dura), with a graft covered by a TVF or with a TVF
alone (fig. 3). When using a graft, attention was given to
harvesting a patch at least 20% to 30% larger than the defect
to prevent contracture induced recurrent curvature. If a
TVF was used, the tunica vaginalis was opened transversely
on its distal and anterior aspect. Two longitudinal incisions
were made on each side of the initial transverse incision to
create a rectangular flap, preserving the cord and the posteriorly based blood supply originating from the cremasteric
artery (fig. 4). The flap or graft was then anastomosed to the
tunica albuginea to cover the underlying corporeal defect
with a running, locking, double armed 5-zero polydioxanone
suture.
Urethroplasty was done using the transverse island flap
technique or 1 of its modifications depending on whether the
plate was or was not divided. Glanuloplasty was performed
FIG. 2. Incision in tunica albuginea of corpora cavernosa extends

from 3 to 9 oclock positions.
with interrupted sutures using 6-zero Monocryl. Skin closure was achieved with a midline closure or with a double
face onlay flap10 when ventral skin deficiency was present.
All patients were given intravenous cephazolin before
skin incision. Oral antibiotics were continued as long as a
urethral stent or suprapubic tube remained in place. A compressive Coban dressing was wrapped around the penis for
24 to 48 hours.
Information regarding recurrent VC and erections was
obtained from patient charts. Chordee was assessed by direct history via children or parent reports and physical examination, which was only possible if the child had a reflex
erection during inspection. Straight erections were documented by parent reports, patient self-assessment or physician observation during physical examination.
Descriptive statistics were done with SPSS, version
11.0. Fischers exact test was applied to compare the outcome of grafts vs TVFs for chordee correction.
RESULTS
Of the 38 patients assessed 21 had perineal and 17 had
penoscrotal hypospadias. A total of 22 patients had associ-
FIG. 4. Tunica vaginalis flap is dissected and ready to be anastomosed to corporeal defect.

ated penoscrotal transposition and/or bifid scrotum, and 6
presented with ambiguous genitalia at birth.
Median age at surgery was 15 months (range 7 to 132).
The urethral plate was divided in 36 patients (94.7%) and
lifted in 2, allowing placement of the graft or flap on the
corpora. A TVF was used alone in 23 cases and associated
with lyophilized dura, pericardium and SIS in 8, 2 and 1,
respectively, when the TVF was used as a second layer to
cover the graft. The remaining 4 boys, who had bilateral
cryptorchidism (therefore, tunica vaginalis was unavailable), underwent ventral grafting, including 1 with dura, 1
with pericardium and 2 with dermis (fig. 5). Urethral reconstruction was achieved by the transverse island flap technique or 1 of its modifications in 34 boys. Four patients
underwent a 2-stage procedure using buccal mucosa graft. A
double face transverse island flap was used to cover the
ventral skin deficiency in 22 patients and midline closure
was applied in 16. Followup was available on 35 of 38 children (range 1 to 11 years, median 5.3).
Postoperatively 15 complications occurred in 11 children
(31.4%), including urethrocutaneous fistula in 5 (14.2%),
recurrent ventral curvature in 5 (14.2%), urethral diverticulum in 2 (5.7%), meatal stenosis in 2 (5.7%) and urethral
stricture in 1 (2.8%). Two boys with a fistula at the
penoscrotal junction had associated recurrent VC due to
ventral skin tethering. In the 2 patients with meatal stenosis a urethral diverticulum subsequently developed. The
only case of urethral stricture occurred at the site of onlay
tube anastomosis.
Recurrent VC greater than 30 degrees was found in 5 of
35 children at a median followup of 5.3 years. Re-intervention to correct chordee was performed in 4 cases. The only
child who did not undergo surgical intervention had mild VC
due to skin tethering from midline ventral skin closure. In 3
of the 4 patients who underwent reoperation a dural graft
had been used to resurface the corporeal defect. These patients had recurrent chordee due to graft contraction with
associated skin tethering, which was evident on clinical
examination of an indurated segment palpable on the ventral surface of the penis. The remaining patient had undergone correction of penoscrotal transposition and a tunica
vaginalis flap for chordee correction. He had recurrent VC
due to considerable scrotal skin tethering. The dural graft
was significantly associated with more recurrent VC than
the TVF. Four of the 9 boys who had a dural graft showed
recurrent VC vs 1 of 23 who had a TVF (44.4% vs 4.3%,
chi-square 5.14, p 0.02, fig. 6).
FIG. 5. Techniques used for ventral corporeal lengthening
1695
FIG. 6. Incidence of recurrent chordee for tunica vaginalis flap vs

dural grafts for ventral corporeal lengthening.
At last followup 22 patients had a slit-like meatus and 13

had an oval meatus. The meatus was orthotopic in 26 children and at the corona in 9. Straight erections were documented in 30 of 35 children (85.7%), including 4 who were
pubertal (Tanner 3 and 4, 11 to 13 years old) and 1 who was
postpubertal (Tanner 5, 17.5 years old). No aneurysmal dilatation of the penile shaft has been observed in any patient
with TVF repair to date.
DISCUSSION
Severe chordee (greater than 45 degrees) in hypospadias is
usually secondary to corporeal disproportion with a shorter
ventral surface of the phallus. The 2 main approaches to
correct severe penile VC are shortening the dorsal aspect of
the corpora cavernosa by plication or lengthening the ventral part by patching the tunica albuginea with several types
of materials.5,1113 Options for correcting severe VC associated with penoscrotal or perineal hypospadias remain a
controversial topic, as shown by a recent survey of an international cohort of pediatric urologists. According to this
multicenter evaluation 69% of the respondents would repair
severe chordee using a ventral approach and 31% would
choose dorsal plication.14
We have adopted a philosophy of lengthening the ventral
surface of the corporeal bodies instead of shortening the
dorsal aspect of the phallus by plication for cases of VC
greater than 45 degrees.15 The available literature shows
that a variety of materials may be used as a graft to
straighten the phallus, such as tunica vaginalis, SIS and
dermis.5,1113 TVF was first used for correcting penile curvature in adults with Peyronies disease.16 It has subsequently been applied for chordee correction in children with
severe hypospadias.11 However, the reported results of this
technique have been contradictory. Caesar and Caldamone
reported on 28 patients undergoing primary and secondary
repairs for chordee.8 They found that 60% of their patients
treated with a TVF had recurrent VC. Additionally, Vandersteen and Husmann identified late onset recurrent chordee
after prior successful hypospadias repair using tunica vaginalis graft.17 On the other hand, Ritchey and Ribbeck reported excellent results with tunica vaginalis grafting of the
corpora. In only 1 of their 25 patients with scrotal or perineal
hypospadias was there evidence of recurrent VC at a median
followup of 9 months during second stage repair.4
There has been some favorable experience with single
layer SIS to correct severe VC.18 We used SIS for corporeal
grafting in only 1 case but the Indiana group reported sig-
1696
nificant complications in 20% of their patients, such as foreign body inflammatory reaction leading to palpable fibrosis
and recurrent ventral chordee.6 Dermal corporeal grafts are
another option that have been widely used with acceptable
results to date.5,19 They are our preferred option in cases of
bilateral cryptorchidism or bilateral inguinal hernia repair
where the TVF is not available as a vascularized flap.
To date there are no published reports of tunica vaginalis
used as a flap for ventral penile lengthening.4,11 Our technique has evolved with time, changing from grafts covered
by TVF to TVF alone to repair severe VC. Our choice is
supported by a previous experimental study in rabbits showing that grafts were associated with necrosis and contracture in 42% of cases, while no contraction was observed in
any TVF specimens.20 We currently consider TVF our material of choice due to its better vascular supply, availability
and easy harvest.
Data on long-term outcomes of VC correction are sparse
and complications have been described after ventral lengthening procedures as well as dorsal plications.17 As reported
by Vandersteen and Husmann, late onset of recurrent penile
chordee has been noted after tunica vaginalis graft use in 3
boys as well as dorsal plication in 19 of 22 with proximal
hypospadias.17 The recurrent VC observed in most of those
children might have been complicated by urethral fibrosis,
leading to inability of the urethra to grow at puberty, rather
than to inadequate correction at the ventral lengthening
procedure. However, more than 80% of these patients had
undergone free graft tube urethroplasty, which may have
contributed to recurrent VC, possibly due to contraction or
lack of growth. The development of chordee in their series
was reported to occur at ages 12 to 18 years as puberty
progressed. It was hypothesized that VC was due to an
intrinsically abnormal growth pattern of the ventral corporeal bodies compared to the dorsal penile surface, which
grew normally during puberty.17
Our observation was different because 22 of 23 patients
(95.6%) who had a TVF to resurface the corporeal defect
demonstrated no recurrent chordee at a median followup of
5.3 years. Although the number of patients with TVF attaining puberty was small in our series, we documented straight
erections with no recurrent VC in 4 pubertal boys 11 to 13
years old and in 1 postpubertal patient who was 17.5 years
old, showing that chordee did not recur after pubertal penile
growth spurt. Based on these findings we speculate that
TVF may result in a better long-term outcome than that of
currently available grafts for ventral corporeal lengthening.
Early in our experience we used grafts covered with TVF
to bridge the corporeal defect in children with severe chordee. As a result, these patients had longer followup than the
boys with TVF alone (7.6 vs 3.8 years). The higher rate of
recurrent VC in patients with grafts could have been related
to extended followup since chordee might have progressed
with time. Despite the shorter followup in patients with TVF
we believe that the outcome achieved with this technique is
superior to that of other procedures because we could document induration in the ventral aspect of the phallus at the
site of the lyophilized dural graft in 3 of 4 children who
required reoperation for recurrent VC. For this reason we
abandoned the use of dural grafts for ventral penile lengthening.
Another potential complication of corporeal grafting is
the risk of erectile dysfunction related to venous leakage
phenomena or aneurysmal dilatation of the corpora. We

acknowledge the limitation of our retrospective study to
obtain reliable information on straight erections, although it
was obtained from patients, and/or parent or physician observations. Since the mean age at last followup was 7.1
years, most boys or their families have witnessed an erection
and been able to detect any abnormality. We also recognize
that our short followup may not have revealed complications
that would become more apparent in adulthood, such as the
real extent of penile chordee and aneurysmal dilatation of
the corpora. However, our previously published results in
the rabbit model demonstrated that histologically normal
tunica albuginea developed under the TVF.20 Therefore,
only prospective studies of the long-term outcome of ventral
lengthening for correcting severe VC will be able to address
those issues.
CONCLUSIONS
The short-term outcome of ventral penile lengthening using
a TVF alone is favorable with a 95% success rate. Dural
grafts appeared to be associated with a higher risk of recurrent VC. Although most of our patients have not yet attained
adulthood, when chordee and erectile dysfunction may become apparent, we believe that TVF repair is a good option
for correcting severe VC.

SIS small intestinal submucosa
TVF tunica vaginalis flap
VC ventral curvature
REFERENCES
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
Perovic SV, Djordjevic ML and Djakovic NG: A new approach

to the treatment of penile curvature. J Urol 1998; 160:
1123.
Mollard P and Castagnola C: Hypospadias: the release of chordee without dividing the urethral plate and onlay island
flap (92 cases). J Urol 1994; 152: 1238.
Kass EJ: Dorsal corporeal rotation: an alternative technique
for the management of severe chordee. J Urol 1993; 150:
635.
Ritchey ML and Ribbeck M: Successful use of tunica vaginalis
grafts for treatment of severe penile chordee in children.
J Urol 2003; 170: 1574.
Soergel TM, Cain MP, Kaefer M, Gitlin J, Casale AJ, Davis
MM et al: Complications of small intestinal submucosa for
corporal body grafting for proximal hypospadias. J Urol
2003; 170: 1577.
Upadhyay J, Shekarriz B and Khoury AE: Midshaft hypospadias. Urol Clin North Am 2002; 29: 299.
Caesar RE and Caldamone AA: The use of free grafts for
correcting penile chordee. J Urol 2000; 164: 1691.
Gittes RF and McLaughlin AP 3rd: Injection technique to
induce penile erection. Urology 1974; 4: 473.
Barroso U Jr, Jednak R, Spencer Barthold J and Gonzalez R:
Further experience with the double onlay preputial flap for
hypospadias repair. J Urol 2000; 164: 998.

11.
12.
13.
14.
15.
16.
17.
18.
19.
20.
Perlmutter AD, Montgomery BT and Steinhardt GF: Tunica

vaginalis free graft for the correction of chordee. J Urol
1985; 134: 311.
Pope JCt, Kropp BP, McLaughlin KP, Adams MC, Rink RC,
Keating MA et al: Penile orthoplasty using dermal grafts in
the outpatient setting. Urology 1996; 48: 124.
Kropp BP, Cheng EY, Pope JCT, Brock JW 3rd, Koyle MA,
Furness PD et al: Use of small intestinal submucosa for
corporal body grafting in cases of severe penile curvature
J Urol 2002; 168: 1742.
Cook A, Khoury AE, Neville C, Bagli DJ, Farhat WA and Pippi
Salle JL: A multicenter evaluation of technical preferences
for primary hypospadias repair. J Urol 2005; 174: 2354.
Geenfield JM, Lucas S and Levine LA: Factors affecting the
loss of length associated with tunica albuginea plication for
correction of penile curvature. J Urol 2006; 175: 238.
Das S: Peyronies disease: excision and autografting with
tunica vaginalis J Urol 1980; 124: 818.
Vandersteen DR and Husmann DA: Late onset recurrent penile chordee after successful correction at hypospadias repair. J Urol 1998; 160: 1131.
Weiser AC, Franco I, Herz DB, Silver RI and Reda EF: Single
layered small intestinal submucosa in the repair of severe
chordee and complicated hypospadias. J Urol 2003; 170: 1593.
Horton CE Jr, Gearhart JP and Jeffs RD: Dermal grafts for
correction of severe chordee associated with hypospadias.
J Urol 1993; 150: 452.
Hafez AT, Smith CR, McLorie GA, El-Ghoneimi A, Herz DB,
Bagli DJ et al: Tunica vaginalis for correcting penile chordee in a rabbit model: is there a difference in flap versus
graft? J Urol 2001; 166: 1429.
EDITORIAL COMMENT
These authors report their results with ventral grafting for
correcting congenital penile chordee and they cite dorsal
plication as the alternative. Unfortunately the approach
recommended by Devine, that is correcting chordee with
multiple small ventral incisions that he termed fairy cuts,
seems to have escaped notice.1 I have applied this latter
approach to chordee correction for the majority of my 30
years in Washington, having never placed a ventral graft.
The plate is elevated from the penile fascia from the corona
proximal, multiple transverse relaxing incisions are made
through the fascia down to erectile tissue and onlay urethroplasty is performed. If urethral transaction becomes necessary, the neourethra is constructed as a transverse preputial
pedicle tube. Rarely dorsal plication has also been necessary
to achieve a straight penis. Fairy cuts produce the same
effect as the single ventral incision recommended by the
authors and they do not require a graft.
Not surprising in this retrospective review, only 5 of 35
patients were followed into prepuberty or puberty. Therefore, one cannot conclude from this report that the ultimate
results of ventral grafting are known. We have also unsuccessfully attempted to obtain objective long-term results on
boys who underwent chordee correction by the method advocated by Devine.1 Although a scientific conclusion cannot
be reached, in all these years of practicing at 1 locale no
postpubertal patients have presented with complaints of
erectile dysfunction, although they may have gone to Baltimore or Philadelphia.
1697
The overriding questions that we all must answer are

how straight a penis must be for normal function and to
what extent we must go to achieve this. Grafting has never
appealed to me for urethral formation or for chordee correction due its risk of contracture.
A. Barry Belman
Childrens National Medical Center
Washington, D. C.
1.
Devine CJ Jr: Chordee in hypospadias. In: Urologic Surgery, 3rd

ed. Edited by JF Glenn. Philadelphia: JB Lippincott Co 1983.
REPLY BY AUTHORS
We believe some theoretical aspects regarding the use of
multiple transverse corporeal relaxing incisions need further discussion. We agree that long-term results of ventral
lengthening with TVF and fairy cuts are difficult to obtain
and, therefore, to some extent unknown, reflecting the obstacles encountered by many authors when following patients with genital problems past adolescence. Although Belman has not observed cases of erectile dysfunction after
performing fairy cuts in many children during his practice,
one must keep in mind that VC may recur as a result of
scarring in the sites where multiple cuts were made. However, this is speculative because the healing process may
occur by regeneration of the tunica albuginea or by fibrosis.
Although limited by the analysis of a select group of boys in
a retrospective fashion, the fact that in the present series
postpubertal evaluations demonstrated good long-term results for ventral lengthening supports the use of this technique as a reasonable management option compared to other
procedures.
We would like to highlight that 23 of 38 boys underwent
correction of ventral curvature with placement of a tunica
vaginalis flap and not a graft. By preserving tissue blood
supply, and not solely relying on neovascularization at the
surgery site, the risk of ischemia and subsequent contraction is diminished. We have previously demonstrated in an
experimental study in rabbits that tunica vaginalis flap
specimens remained viable with regeneration of a normal
tunica albuginea on the ventral aspect of the corpora cavernosa in contrast to grafts which were associated with a
significant amount of scar formation and contraction (reference 20 in article). We believe a similar phenomenon occurs
in humans, as VC recurred in only 1 of 23 boys who underwent placement of a TVF. The degree of curvature in that
particular case was mild and caused by skin tethering, and
not by contraction of the flap, which was verified at reoperation.
The long-term evaluation of children who undergo complex reconstructive procedures before puberty is important. Relying on the assumption that they do not seem to
be having problems during adulthood because there are
scant reports in the literature or concerns have not been
raised by reconstructive surgeons who see adults on a
regular basis is not sufficient evidence. We must continue
to improve the evaluation of our patients as they graduate
from pediatric urology practices since long-term followup
is clearly lacking.
Small Intestinal Submucosa for Corporeal Body Grafting

in Severe Hypospadias Requiring Division of the Urethral Plate
James M. Elmore, Andrew J. Kirsch,* Hal C. Scherz* and Edwin A. Smith
From the Healthcare of Atlanta and Emory University School of Medicine, Atlanta, Georgia
Purpose: Severe chordee associated with proximal hypospadias is a challenging problem. Occasionally division of the
urethral plate and corporeal body grafting are necessary to achieve a straight phallus. We report our experience with small
intestinal submucosa for corporeal body grafting to correct severe chordee associated with proximal hypospadias.
Materials and Methods: We reviewed the charts of all 28 patients who underwent small intestinal submucosa corporeal
body grafting to correct severe chordee during stage 1 of planned 2-stage proximal hypospadias repair from January 2001 to
June 2006. After mobilization and division of the urethral plate ventral horizontal corporotomies were performed at the site
of maximum curvature. Four and 1 ply small intestinal submucosa in 21 and 7 cases, respectively, was then fashioned to size
and secured into the defect. Preoperative, operative and postoperative clinic visit records were reviewed to assess outcomes.
Results: Preoperatively the meatal position was penoscrotal in 18 patients, scrotal in 5 and perineal in 5. In all patients, who
had a mean age of 16 months, the urethral plate was divided and the corporeal bodies were grafted during stage 1 surgery.
Vascularized preputial flaps were rotated ventrally for subsequent urethroplasty in 25 patients and buccal mucosa was used
for urethroplasty in the remaining 3. A total of 26 patients underwent stage 2 repair with a mean followup of 38 months
(range 6 to 61). Overall there were 6 complications, including epidermal retention cysts in 2 cases, partial urethral dehiscence
in 2 and urethral diverticulum requiring reduction urethroplasty in 1, while a perimeatal granuloma developed in 1 following
stage 1 repair, which required excision. Except the possibility of this latter patient there were no complications related to the
small intestinal submucosa graft and a straight phallus without fibrosis was achieved in all 26.
Conclusions: Our experience suggests that small intestinal submucosa is a suitable material for corporeal body grafting in
2-stage hypospadias repairs when there is associated severe chordee. In rare instances in which other maneuvers are
inadequate to correct chordee small intestinal submucosa grafting is a reasonable option and a low complication rate can be
anticipated.
Key Words: penis; abnormalities; hypospadias; transplants; small intestine
roximal hypospadias associated with severe chordee is

an infrequent but challenging problem. A goal of hypospadias repair is the achievement of a straight phallus, which must be accomplished before urethroplasty. Maneuvers to address ventral penile chordee are usually
attempted sequentially from less to more aggressive. These
steps involve skin release, excision of ventral fibrotic fascia,
dissection of the spongiosum and dorsal plication.1 However,
occasionally these maneuvers are inadequate, and the degree of corporeal body disproportion and foreshortening of
the spongiosum necessitates division of the urethral plate
and corporeal body grafting. Material options for corporeal
reconstruction traditionally include free dermal or tunica
vaginalis grafts.25
SIS was recently been described for corporeal body grafting in children with severe chordee.6,7 This material is a
commercially available, acellular, freeze-dried and sterilized
matrix from porcine small intestine, which is composed principally of type 1 collagen.6 It was reported to be safe and

* Financial interest and/or other relationship with Q-Med Scandinavia.
Correspondence: Georgia Urology, P. A., 5445 Meridian Mark
Plaza, Suite 430, Atlanta, Georgia 30340 (telephone: 404 252-5206;
FAX: 404 252-1268; e-mail: smithea@aol.com).
0022-5347/07/1784-1698/0
easy to use, and it has a significant advantage over dermal

grafts, in that it does not require a separate incision.8
SIS has been successful in animal models and in adults
with Peyronies disease.9 11 However, currently there are
only a few reports in the literature of SIS for corporeal body
grafting in children with severe penile curvature. Furthermore, some reports indicate the possibility that SIS may
elicit a foreign body giant cell granulomatous reaction that
may lead to inflammatory fibrotic nodules or recurrent chordee.6 Consequently there is some debate regarding the use of
SIS for corporeal body grafting.
In the last several years we have used SIS to correct the
severe chordee associated with proximal hypospadias. We
have used this material because it is readily available and
easy to handle. Our initial experience has been encouraging.

We reviewed the charts of all 28 patients who underwent
SIS corporeal body grafting to correct severe chordee as
stage 1 of planned 2-stage proximal hypospadias repair between January 2001 and June 2006. Corporeal body grafting
was only performed when there were more than 30 degrees
of persistent curvature after the penis was degloved, any
chordee tissue was excised and the urethral plate was
1698
Vol. 178, 1698-1701, October 2007

Printed in U.S.A.
DOI:10.1016/j.juro.2007.05.090
SMALL INTESTINAL SUBMUCOSA FOR CORPOREAL BODY GRAFTING

transected. These patients were judged not to be candidates
for dorsal plication alone because of the unacceptable degree
of penile shortening that would have resulted.
After completely degloving the penis the urethral plate
was mobilized and divided. Artificial erection was then performed and the point of maximum curvature was identified.
Ventral horizontal corporotomies were performed at the site
directly opposite that of maximum curvature, producing an
elliptical defect. Four and 1 ply SIS in 21 and 7 patients,
respectively, was then fashioned to size and secured into the
defect with 4 interrupted 6-zero polyglactin sutures at the
corners and subsequently with 2 to 4 running 6-zero polyglactin sutures (see figure). The type of SIS used was based
only on operating surgeon preference. The thickness of the
corporeal body, the degree of chordee or the size of the
corporotomy defect did not influence this decision. To allow
for postoperative contraction the graft was fashioned 10% to
20% larger than the defect.
Vascularized preputial flaps were then rotated ventrally
distal to the transected urethral meatus for subsequent urethroplasty in 25 patients, while in the remaining 3 an onlay
buccal mucosal graft was used. Graft coverage was performed with well vascularized penile skin. If the quality of
the ventrally rotated skin flaps appeared questionable, a
dartos flap was used to supply additional vascularized support to the graft. All patients received intravenous antimicrobials intraoperatively, usually cefazolin. They were continued on prophylactic doses of antibiotics until the urethral
stent was removed, usually between postoperative days 5
and 10.
Success of the initial surgical procedure was determined
during a preoperative clinic visit before stage 2 repair, by an
artificial penile erection test intraoperatively during the
stage 2 procedure and by physical examination in the office
postoperatively after the stage 2 procedure. Preoperative,
operative and postoperative clinic visit records were reviewed to assess outcomes.
RESULTS
Preoperatively the meatal position was penoscrotal in 18
patients, scrotal in 5 and perineal in 5. In all patients, who
had a mean age of 16 months, the urethral plate was divided
and the corporeal bodies were grafted during stage 1. Vascularized preputial flaps were rotated ventral for subsequent urethroplasty in 25 patients, while a buccal mucosal
onlay graft was performed in the remaining 3. Stage 2 ure-
1699
throplasty was performed 6 to 12 months following the stage

1 procedure using the Thiersch-Duplay technique.
Currently 26 patients have undergone stage 2 repair with
a mean followup of 38 months (range 6 to 61). Overall there
were 6 complications, including epidermal retention cysts in
2 patients, partial urethral dehiscence in 2 and a urethral
diverticulum requiring reduction urethroplasty in 1, while
in 1 a perimeatal granuloma developed following the initial
surgery, which subsequently required excision. With the
possible exception of this last patient, in whom 1 ply SIS was
used, there were no complications related to the SIS graft. A
straight phallus without fibrosis was achieved in all 26 patients.
DISCUSSION
Proximal hypospadias with associated chordee that is severe
enough to warrant division of the urethral plate and corporeal body grafting is uncommon. During the 5-year study
period corporeal body grafting was necessary in only 28
patients. During the same period we have performed more
than 1,000 hypospadias repairs in which penile straightening was unnecessary or could be achieved without corporeal
reconstruction.
Typical maneuvers used to correct chordee include (in the
usual order of application) skin release, excision of ventral
fibrotic fascia (chordee tissue), mobilization of the urethral
plate with radical bulbar dissection,12 division of the urethral plate and, finally, ventral corporeal grafting. As described by Baskin and Duckett,13 dorsal plication is frequently performed to correct milder degrees of chordee that
may persist after any of these maneuvers. However, in cases
of more severe residual chordee this technique can result in
penile shortening. Particularly when the phallus is already
short, plication may result in a squat tuna can penis.14
Corporeal grafting was done in our series when the degree of
residual chordee was in the range of 30 degrees after performing all other release maneuvers, including division of
the urethral plate. When necessary, corporeal body grafting
may be performed using various materials. As first described in 1975 by Horton and Devine,15 free dermal grafts
to repair chordee, were reported to have a low complication
rate.3 The disadvantage of this technique is the additional
time requirement and the potential morbidity from graft
harvest. One study also suggests that a second dermal graft
and/or dorsal plication may be beneficial in more than 40%
of cases.14
SIS corporoplasty. A, 3 of 4 quadrants have been sewn. B, before stage 2 procedure. C, after stage 2 procedure
1700
The other commonly used autologous material for corporeal reconstruction is tunica vaginalis. It is easy to harvest
but reported outcomes vary in terms of long-term success. In
the initial report by Perlmutter et al excellent results were
achieved in 10 of 11 patients.4 Although similar results were
reported by others, less favorable outcomes have likewise
been reported. Caesar and Caldamone noted recurrent chordee in 3 of 5 patients after tunica vaginalis corporoplasties.16 Vandersteen and Husman also observed recurrent
curvature using tunica vaginalis.17 Despite advantages that
include the ease of the procedure and low morbidity discrepancies in reports regarding efficacy make this technique less
attractive.
Initial studies in animal models suggesting that SIS is
safe and effective for corporeal body reconstruction in animal models provided the impetus for its use in humans.9
Groups at several centers reported their experience using
SIS in adults with penile curvature resulting from Peyronies disease. Except for a recently published series of 4
patients,18 most of the literature suggests that this material
is safe and efficacious in this setting.10,11 Obviously Peyronies disease is a entirely different entity than congenital
chordee and caution should be used when attempting to
extrapolate success or failure with SIS in the Peyronies
population to children with congenital chordee. Disease progression or the completeness of plaque excision are confounders that do not have a role in congenital curvature.
However, the literature on SIS for chordee in children is
less ample. Only 3 series with a total of 36 patients of SIS in
this scenario have been reported in the literature. These
studies suggest that 1 ply SIS may be safe in this setting,
while the 4 ply variety may be prone to complications. Kropp
et al reported their experience with correcting chordee in
patients with hypospadias or epispadias using 1 ply SIS.8
Using 1 ply SIS there were no complications related to SIS at
a mean followup of 14 months and the penis was straight in
all patients. Weiser et al used 1 ply SIS for corporeal body
grafting in a series of 9 boys with proximal hypospadias and
severe chordee.7 Followup was 16 to 21 months and again
there were no complications related to SIS. In contrast,
Soergle et al reported a complication rate of 33% in a series
of 12 patients who underwent corporeal body grafting.6 In
this series all grafts were performed using the 4 ply version
of SIS. A relatively high complication rate prompted that
group to abandon the use of SIS for corporeal reconstruction.
The discrepancy among the outcomes of these reports raises
the question of whether SIS should be considered a safe and
effective option for corporeal reconstruction, more specifically whether equivalent outcomes could be expected with
the 1 and 4 ply versions of SIS.
In our study 2 of us preferred 4 over 1 ply SIS and, as a
result, 21 of our 28 patients underwent reconstruction with
4 ply SIS. The motivation for using 4 rather than 1 ply SIS
was similar to that described by Soergel et al.6 The 4 ply
version of SIS handles easily and it more closely approximates the thickness and consistency of the corpora. Therefore, it seems to be a closer substitute to a dermal graft. In
addition, 4 ply SIS does not bulge as severely at artificial
erection and a watertight anastomosis may be easier to
achieve than with the 1 ply variety. Most importantly initial
results using 4 ply SIS were satisfactory and, thus, its use
was continued during the study interval. Indeed, in the
current series at a mean followup of more than 3 years there
were no complications related to 4 ply SIS. Although unlikely, the single complication that was potentially related to
SIS was a meatal granuloma, which occurred in a patient
who received 1 ply SIS. However, given that the graft and
granuloma were not in proximity to each other, SIS was
likely not causative in this case. There is evidence in the
literature suggesting that SIS can cause chronic inflammation, fibrosis and even calcification but in an animal model
this occurred with 4 ply SIS and it did not occur when 1 ply
SIS was used.19 As noted, clinical reports of the use of 1 ply
SIS for pediatric corporeal grafting have been uniformly
favorable.7,8
Although our complication rate was low and at a reasonable mean followup of more than 3 years, it is possible that
late complications could occur, namely recurrent chordee or
erectile dysfunction. Hypothetically recurrent of chordee
could manifest years later at puberty when the grafted corpora does not grow normally. This could lead to corporeal
body disproportion and subsequent recurrent chordee.
Longer term followup is necessary to determine the true risk
of this condition.
Vascular ingrowth and subsequent replacement of the
SIS graft by host tissues occurs through the processes of
inosculation and imbibition. These processes may occur
more easily and dependably with 1 rather than 4 ply SIS. If
these processes do not successfully occur, SIS could be seen
as foreign and an inflammatory reaction may ensue. This is
a plausible explanation of why animal models and prior
clinical experience with the thicker 4 ply SIS resulted in
fibrosis. The reason for our low complication rate using 4 ply
SIS is unclear. We speculate that the graft bed as well as the
tissue superficial to the graft is required for a good outcome.
For this reason if the skin overlying the SIS appeared compromised, a dartos flap was harvested and placed over the
SIS. Although this maneuver was required in only 5 patients, it may have prevented complications that would have
significantly altered our outcomes. It does not appear as
though dartos flaps were used by Soergel et al,6 which may
explain their higher complication rate. Lastly the inflammatory reaction and fibrosis caused by SIS may be related to
the amount of SIS used and not just to graft thickness. This
could also explain why the 4 ply version is associated with
chronic inflammation and fibrosis in animals. We fashioned
our grafts to be 10% to 20% larger than the corporeal defect
to allow for postoperative contracture. In the series by
Soergel et al the grafts were fashioned 25% larger than the
defect. Although the difference is small, it is another possible explanation for their higher complication rate.
CONCLUSIONS
Our experience suggests that SIS is a suitable material for
corporeal body grafting in 2-stage hypospadias repairs with
severe chordee. In rare instances in which other maneuvers
are inadequate to correct chordee SIS grafting is a reasonable option and a low complication rate can be anticipated.
Although we have had excellent outcomes with 1 and 4 ply
SIS, the 1 ply variety is associated with fewer complications
when the literature as a whole is reviewed. Given that
animal studies also suggest that fibrosis may be more common with the 4 ply variety, we believe that 1 ply SIS may be
the best choice and we now use this exclusively in our
practice.

SIS small intestinal submucosa
REFERENCES
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
15.
16.
17.
18.
19.
Donnahoo KK, Cain MP, Pope JC, Casale AJ, Keating MA,
Adams MC et al: Etiology, management and surgical complications of congenital chordee without hypospadias.
J Urol 1998; 160: 1120.
Devine CJ Jr and Horton CE: Use of dermal graft to correct
chordee. J Urol 1975; 113: 56.
Kogan SJ, Reda EF, Smey PL and Levitt SB: Dermal graft
correction of extraordinary chordee. J Urol 1983; 130: 952.
Perlmutter AD, Montgomery BT and Steinhardt GF: Tunica
vaginalis free graft for the correction of chordee. J Urol
1985; 134: 311.
Snow BW, Cartwright PC and Unger K: Tunica vaginalis blanket wrap to prevent urethrocutaneous fistula: an 8-year
experience. J Urol 1995; 153: 472.
Soergel TM, Cain MP, Kaefer M, Gitlin J, Casale AJ, Davis
MM et al: Complications of small intestinal submucosa for
corporeal body grafting for proximal hypospadias. J Urol
2003; 170: 1577.
Weiser AC, Franco I, Herz DB, Silver RI and Reda EF: Single
layered small intestinal submucosa in the repair of severe
chordee and complicated hypospadias. J Urol 2003; 170: 1593.
Kropp BP, Cheng EY, Pope JC, Brock JW, Koyle MA, Furness
PD et al: The use of small intestinal submucosa (SIS) for
corporeal body grafting in cases of severe penile curvature.
J Urol 2002; 168: 1742.
Kropp BP, Ludow JK, Spicer D, Rippy MK, Badylak SF, Adams
MC et al: Rabbit urethral regeneration using small intestinal submucosa onlay grafts. Urology 1998; 52: 138.
Knoll LD: Use of porcine small intestinal submucosal graft in
the surgical management of Peyronies disease. Urology
2001; 57: 753.
Knoll LD: Use of porcine small intestinal submucosal graft in
the surgical management of tunical deficiencies with penile
prosthetic surgery. Urology 2002; 59: 758.
Baker LA, Mathews RI and Docimo SG: Radical bulbar dissection to correct severe chordee and proximal hypospadias.
J Urol 2000; 164: 1347.
Baskin LS and Duckett JW: Dorsal tunica albuginea plication
for hypospadias curvature. J Urol 1994; 151: 1668.
Devine CJ Jr and Horton CE: Use of dermal graft to correct
chordee. J Urol 1975; 113: 56.
Caesar RE and Caldamone AA: The use of free grafts for
correcting penile chordee. J Urol 2000; 164: 1691.
Vandersteen DR and Husmann DA: Late onset recurrent penile chordee after successful correction at hypospadias repair. J Urol 1998; 160: 1131.
John T, Bandi G and Santucci R: Porcine small intestinal
submucosa is not an ideal graft material for Peyronies
disease surgery. J Urol 2006; 176: 1025.
Hafez AT, El-Assmy A and El-Hamid MA: 4 Layer versus 1
layer small intestinal submucosa for correction of penile
chordee: experimental study in a rabbit model. J Urol 2004;
171: 2489.
division of the urethral plate and incision of the ventral

corpora. This technique was previously reported in the pediatric urological literature and this is the fourth article to
date discussing the validity of choosing 1 vs 4 ply SIS as the
optimal construct for corporeal grafting (references 6 to 8 in
article). The authors discuss their experience and excellent
results with 4 ply SIS and touch on the concept of revascularization as a possible influence on outcomes. I would agree
and further state that the concept of revascularization is
probably what dictates success vs less than favorable results
in corporeal grafting. The author provides the data that
support this concept, in that they placed a vascularized
dartos flap over the corporeal graft in 5 patients in whom the
corpora and covering skin were thin and dysplastic. This
maneuver likely allowed prompt inosculation and imbibition
of the graft, thus, avoiding possible chronic inflammation,
fibrosis and failure.
The authors reported experience with corporeal grafting
occurred in children who had with proximal hypospadias. As
all pediatric urological surgeons have experienced too often,
the severe hypospadiac penis shows a lack of healthy ventral
tissue to start and any reconstructive process benefits take
the form of providing adjacent or rotational flaps of vascularized tissue. From the large body of literature on hypospadias we can draw a parallel conclusion that providing
healthy vascularized tissue to the healing process, whether
by urethroplasty or corporeal graft healing, certainly helps
overall surgical success.1 As with all tissue engineering,
whether it be in constructs for bladder augmentation or
constructs used for corporeal grafting, the key to tissue engineering is revascularization of the donor tissue via the host.
The larger the graft, the larger the task, and the greater the
chance of failure. The poorer the tissues, the more the need to
provide adjacent healthy (vascularized) tissue, thus, optimizing the chance of success.
The cases reported requiring corporeal grafting are not
common practice. In the last 8 years at our institution my
partner and I have needed to perform corporeal grafting in
fewer than 10 children in more than 2,000 hypospadias
repairs. I personally have used 1 ply SIS and routinely cover
the graft with a healthy dartos flap, if possible with 2 flaps,
because of this concept of providing vascularized tissue for
the optimal healing process. I have not found in an infant
that 1 ply is disadvantageous to use or handle and I also use
a graft that is approximately 25% larger than the defect
being grafted. In our experience we have not noted complications when using 1 ply SIS as the corporeal graft (reference 8 in article). I am encouraged that the authors have had
good results with 4 ply and I will not hesitate to considering
using this in an older patients as long as I can provide
adequate dartos flaps for revascularization. Tissue revascularization is the current roadblock that must be overcome in
all tissue engineering.
Peter D. Furness, III
University of Colorado Health Science Center
The Childrens Hospital
Denver, Colorado
EDITORIAL COMMENT
1.
These authors report a single institutional experience with

SIS for corporeal grafting in severe hypospadias requiring
1701
Furness PD III: Successful hypospadias repair with a ventral

based vascular dartos pedicle (VVP) for urethral coverage.
J Urol 2003; 169: 1825.
Vesicoureteral Reflux II
Multivariate Comparison of the Efficacy of Intraureteral Versus
Subtrigonal Techniques of Dextranomer/Hyaluronic Acid Injection
Jonathan C. Routh,* Yuri Reinberg, Richard A. Ashley, Brant A. Inman, James J. Wolpert,
David R. Vandersteen, Douglas A. Husmann and Stephen A. Kramer
From the Department of Urology, Mayo Clinic, Rochester and Division of Urology, Pediatric Surgical Associates (YR, JJW, DRV),
Minneapolis, Minnesota
Purpose: Numerous factors have been postulated to increase success rates for dextranomer/hyaluronic acid injection for
vesicoureteral reflux. Ureteral hydrodistention combined with intraureteral injection reportedly improves injection success
rates. We combined the results of 5 pediatric urologists to evaluate the efficacy of this technique compared to that of
subtrigonal-only injection in relation to other factors.
Materials and Methods: Patients with primary vesicoureteral reflux undergoing dextranomer/hyaluronic acid injection
from April 2002 to December 2005 at 2 institutions were eligible. Only patients with primary vesicoureteral reflux were
included in the study. Injection success was defined as the complete absence of reflux on followup voiding cystourethrogram
or radionuclide cystogram. Predictors of a successful outcome were analyzed statistically with logistic regression. Factors
included in our analysis were gender, age, vesicoureteral reflux grade, dysfunctional voiding, amount of injected dextranomer/
hyaluronic acid, injection technique (intraureteral vs subureteral) and surgeon.
Results: A total of 301 patients (453 ureters) with a median age of 5.5 years met inclusion criteria, of whom 199 (66%) were
cured at 3 months of followup. Of the patients 145 (48%) underwent subureteral injection and 156 (52%) underwent ureteral
hydrodistention combined with intraureteral injection. On multivariate analysis only vesicoureteral reflux grade (p 0.001)
and surgeon (p 0.01) were significantly predictive of injection success. There was a trend toward significance with ureteral
hydrodistention combined with intraureteral injection (p 0.056).
Conclusions: In our multivariate model only vesicoureteral reflux grade and surgeon were independently predictive of
injection success in patients with primary, uncomplicated vesicoureteral reflux. There was a trend toward improved results
with ureteral hydrodistention combined with intraureteral injection, although this did not achieve statistical significance.
Key Words: ureter, bladder, vesico-ureteral reflux, hyaluronic acid, injections
esicoureteral reflux is a common and controversial

problem in pediatric urology, affecting approximately
1% of children in the United States.1 Since its introduction to the United States in September 2001, the use
of Dx/HA copolymer (Deflux) for endoscopic correction of
VUR has increased greatly. The minimal morbidity of
Dx/HA injection prompted some investigators to recommend
its use as a first line alternative to traditional therapeutic
standards of antibiotic prophylaxis or open ureteroneocystostomy.2 However, the reported efficacy of Dx/HA injection
is variable with case series showing a success rate of 64% to
100%.37 It is difficult to directly compare these series because they were performed in heterogeneous patient populations using different injection techniques and varying definitions of success and failure.
Kirsch et al recently advocated the use of the HIT technique, which increased their success rates.8 To our knowledge these results have not been validated at other institutions. We present a multi-institutional cohort study of
Dx/HA injection for the treatment of children with VUR, in
Study received approval from the institutional review boards.

* Correspondence: 200 1st St Southwest, Rochester, Minnesota
55901 (telephone: 507-284-2511; FAX: 507-284-4951; e-mail: routh.
jonathan@mayo.edu).
0022-5347/07/1784-1702/0
which we evaluated the importance of injection technique as

a predictor of treatment success.

Patient Selection and Clinical Features
After obtaining approval from the appropriate Institutional
Review Boards we identified 467 patients who underwent
Dx/HA injection at the 2 study institutions between April
2002 and December 2005. All patients were determined to
have VUR by VCUG or RNC. Patients and families were
presented with a detailed explanation of currently accepted
management options for VUR, including: antibiotic prophylaxis, Dx/HA injection and open ureteroneocystostomy. Patients electing Dx/HA injection were tracked in a database
maintained by study nurses and 1 of us (JCR). Only the first
Dx/HA injection in a patient was considered in this analysis.
Radiological followup consisted of cyclic voiding studies
(VCUG or RNC) and renal ultrasound performed 3 months
postoperatively. Treatment success was defined as the absence of VUR and obstruction on these imaging modalities.
Exclusion criteria included paraureteral diverticula in 20
patients, ureteral duplication in 42, prior bladder surgery in
17, neurogenic bladder in 14 and lack of adequate radiographic followup at 3 months in 73. Only ureters with pre-
1702
Vol. 178, 1702-1706, October 2007

Printed in U.S.A.
DOI:10.1016/j.juro.2007.03.174
INTRAURETERAL VERSUS SUBTRIGONAL DEXTRANOMER/HYALURONIC ACID INJECTION

operative radiographically proven VUR were analyzed. Contralateral ureters that were not initially injected but later
developed de novo VUR were not included in our analysis.
Injection Technique
The method of injection technique was determined by the
attending surgeon. We defined the HIT and subureteral
injection techniques in accordance with published descriptions.8,9 The figure shows these techniques.
Statistical Methods
Univariate logistic regression models were constructed to
predict treatment success at 3 months. Covariates analyzed
were gender, age, VUR grade, dysfunctional voiding,
amount of injected Dx/HA, injection technique and surgeon.
All covariates with univariate p 0.2 were included in a
multivariate logistic regression model. Model residuals revealed no violation of regression assumptions. Statistical analyses were performed using SPSS, version 14.0. All tests were
2-sided and p 0.05 was considered statistically significant.
RESULTS
A total of 301 patients (453 ureters) with primary VUR were
included in our analysis. Median patient age was 5.3 years
(95% CI 0.4-11.8). The cohort was predominantly female
(262 of 301 or 87%) and 111 ureters (24.5%) had persistent
VUR at 3 months of followup. Subureteral injection was
performed in 204 ureters (45%) and it was successful in 141
(69%). In contrast, HIT was performed in 249 ureters (55%)
and it was successful in 201 (81%). Interestingly there was
no significant difference in the time frame of injections between the HIT and STING subgroups, as manifested by
median injection dates and the range of injection dates.
Success rates decreased with increasing VUR, including
83% (25 of 30 cases), 82% (206 of 250), 66% (106 of 161) and
53% (8 of 15) for grades I to IV/V. The average amount of
Dx/HA injected was 0.93 ml (range 0.2 to 3.5).
On univariate analysis we found that gender, age, VUR
grade, injection technique and surgeon were significant predictors of outcome (see table). On multivariate analysis only
1703
VUR grade and surgeon remained significant predictors of

outcome, although there was a trend toward significance for
injection technique (see table). There was a definite association between surgeon and injection technique (p 0.001),
which explained much but not all of the difference in injection failure rates. Similarly there was a significant association between surgeon and VUR grade (p 0.001). Despite
adjustment for these differences by multivariate analysis surgeon remained an important predictor of outcome (p 0.01).
DISCUSSION
In 1995 Stenberg and Lackgren introduced their experience
with Dx/HA in 75 children, reporting a 68% cure rate at 3
months of followup.9 Since that time, multiple series have
shown highly divergent cure rates of 64% to 100%.37,9 Unfortunately these series are difficult to compare due to their
heterogeneous patient populations and differing study end
points, eg grade 0 vs grade I VUR. In 2004 Kirsch et al
described the use of the modified STING (or HIT) technique
in the setting of Dx/HA injection, reporting a per patient
cure rate of 89% across all VUR grades.8 While the use of
this modification was encouraging, these results have not
been validated or reproduced. Nonetheless, results such as
these coupled with the minimally invasive nature of injection therapy led some investigators to conclude that Dx/HA
injection may be an acceptable first line modality for the
treatment of uncomplicated VUR in some children.2
We determined the reproducibility of success reported
with the HIT technique. We combined the results of 5 fellowship trained pediatric urologists at 2 tertiary pediatric
hospitals, of whom all are experienced with Dx/HA injection
therapy. Our study population was limited to only patients
with primary VUR who received a single Dx/HA injection.
We excluded all patients with an anatomically or functionally complex bladder to avoid potential confounding factors.
Multivariate regression was done to adjust for important
covariates such as VUR grade, which we would reasonably
expect to impact our results.
Our findings are notable for various reasons. We noted a
trend toward improved outcome with the HIT injection tech-
Two injection techniques
1704

Per ureter logistic regression analysis of factors predicting success of Dx/HA injection
Univariate
Sex:
F (referent)
M
Age (yrs):
Younger than 6 (referent)
6 or Older
VUR grade:
I (referent)
II
III
IV/V
Injection technique:
HIT (referent)
STING
Surgeon:
1 (referent)
2
3
4
5
Dysfunctional voiding
No (referent)
Yes
Dx/HA injected (ml):
Less than 1 (referent)
1 or Greater
No. Pts
OR (95% CI)
388
65
1
1.87 (1.073.27)
275
176
1
0.71 (0.320.82)
30
248
161
14
1
1.08 (0.392.97)
2.59 (0.947.15)
5.00 (1.2120.71)
249
204
1
1.87 (1.212.89)
51
160
103
87
50
1
2.03 (0.844.86)
1.71 (0.684.31)
1.64 (0.634.25)
6.29 (2.3816.60)
315
138
1
0.76 (0.471.22)
241
212
1
1.16 (0.751.78)
Multivariate
p Value
OR (95% CI)
0.029
p Value
0.192
1
1.51 (0.812.79)
0.018
0.001
0.244
1
0.57 (0.340.95)
0.001
1
0.99 (0.342.86)
2.46 (0.857.12)
7.27 (1.5134.94)
0.005
0.056
1
1.89 (0.983.53)
0.001
0.010
1
2.89 (1.077.79)
2.20 (0.786.21)
1.56 (0.475.20)
5.17 (1.5117.74)
0.254
0.253
Total sample size was 453 ureters.
nique, although our results did not replicate the success

reported by Kirsch et al.8 Even with the HIT technique our
per patient success rate was only 71% across all VUR grades
and only 1 surgeon was able to achieve a success rate of
greater than 80%. This is particularly intriguing in light of
the fact that 61% of our cohort had nondilating VUR compared to 54% of the cohort of Kirsch et al. Similarly only 3%
of patients in our series had severe (grade IV/V) VUR compared to 9% in the series by Kirsch et al.8 By our own
analysis this should predispose our series to higher success
rates than those of the Atlanta group, and yet we did not find
this to be the case. However, it is important to note that our
aggregate success rate is well within the CI provided by
Elder et al in their recent meta-analysis of endoscopic VUR
correction.10 Thus, despite a cumulative injection experience
of more than 450 patients and 850 ureters we cannot conclude that a per ureter cure rate of 90% or higher is reliably
reproducible even at high volume centers.
A second finding of our study is that surgeon was an
important predictor of success. This may represent evidence
for a Dx/HA specific learning curve because surgeons 1 and
5 with significant deviation from the remainder of the group
also had the lowest numbers of injected ureters (see table).
Interestingly we saw no other evidence of a demonstrable
learning curve in our analyses. However, all 5 attending
surgeons had previously performed endoscopic injections
with other materials before Dx/HA. Further studies are necessary to clarify the role of surgeon as a determinant of
injection success. It should be noted that the lack of randomization in our study implies that selection of the injection
technique and of patients for each technique may have been
inherently biased. The importance of VUR grade in our
analysis theoretically means that grade may have unduly
influenced the surgeon choice of technique. However, as mentioned, this factor was carefully addressed in our multivariate
analysis, correcting any possible bias insofar as possible.
In contrast to previous reports, we saw no correlation

between the amount of injected Dx/HA and success
rates.11,12 It is possible that study surgeons tended to use
higher amounts of Dx/HA in patients with more severe VUR.
Because these patients were predisposed to failure due to
the severity of VUR, it is possible that this effectively nullified any benefit from a larger amount of Dx/HA.
Several limitations of the current study merit discussion.
The retrospective, nonrandomized nature of our study design introduced inherent limitations to our analyses, particularly the possibility of confounding bias due to unidentified
factors. We attempted to decrease the probability of confounding by excluding from our analyses any patient with an
anatomically or functionally complex bladder, including ureteral duplication, neurogenic bladder or previous bladder
surgery, among others. A potential confounding factor for
which we could not account is the number of times that the
urothelium was punctured during injection. This factor was
previously shown to influence recurrence rates in at least 1
series.4 Thus, this could have skewed our results, particularly if patients with HIT or STING were disproportionately
more likely to receive multiple punctures. Unfortunately
based on our data this factor was not evaluable.
We did not exclude patients with clinical voiding dysfunction because this diagnosis did not predict treatment failure
in prior series.4 However, it is worth noting that this parameter was previously observed by some groups to be predictive
of injection failure.13 Our analyses did not reveal that voiding dysfunction had any role in our success or failures and it
was not disproportionately present in either technique cohort. Therefore, we conclude that it is unlikely that voiding
dysfunction represents a likely confounder. While it is possible that other confounding variables were present and
unaccounted for in our study, we believe that our strict
exclusion criteria and careful multivariate analyses minimized this risk insofar as possible.

Followup in this study was short at only 3 months. Previous publications showed that with continued followup a
small percent of patients who are cured by Dx/HA injection
will later experience treatment failure.7,14 We anticipate
that with time our cure rate would decrease slightly. However, due to the discomfort and radiation associated with
radiographic assessment of reflux few patients in this cohort
consented to 1-year followup imaging in the setting of a normal
3-month study. Longitudinal studies are clearly required to
better evaluate the long-term efficacy of Dx/HA in children.
As mentioned, resident urologists participated in the majority of injections. While the argument can be made that
our success rates are limited only due to resident involvement, it should be noted that the level of resident involvement was consistent among all attending surgeons. Indeed,
surgeons 1 and 5 with a 14% and 50% failure rate, respectively, each had residents perform most injections under
their direct supervision. Because the resident variable remained constant throughout the series with quite divergent
results, this argument appears less important than might
otherwise be assumed.
CONCLUSIONS
In our multivariate model only VUR grade and surgeon were
independently predictive of injection success in patients
with primary, uncomplicated VUR. There was a trend toward improved results with the HIT injection technique,
although this did not achieve statistical significance. Larger
prospective studies are required to confirm these findings.
ACKNOWLEDGMENTS
Heidi Pfefferle and Sue Rathbun assisted with this study.
6.
7.
8.
9.
10.
11.
12.
13.
14.
1705
of dextranomer/hyaluronic acid copolymer: preliminary

results. J Urol 2003; 170: 1541.
Routh JC, Vandersteen DR, Pfefferle H, Wolpert JJ and Reinberg Y: Single-center experience with endoscopic management of vesicoureteral reflux in children. J Urol 2006; 175:
1898.
Lackgren G, Wahlin N, Skoldenberg E and Stenberg A: Longterm followup of children treated with dextranomer/hyaluronic acid copolymer for vesicoureteral reflux. J Urol 2001;
166: 1887.
Kirsch AJ, Perez-Brayfield M, Smith EA and Scherz HC: The
modified STING procedure to correct vesicoureteral reflux:
improved results with submucosal implantation within the
intramural ureter. J Urol 2004; 171: 2413.
Stenberg A and Lackgren G: A new bioimplant for the endoscopic treatment of vesicoureteral reflux: experimental and
short-term clinical results. J Urol 1995; 154: 800.
Elder JS, Diaz M, Caldamone AA, Cendron M, Greenfield S,
Hurwitz R et al: Endoscopic therapy for vesicoureteral reflux: a meta-analysis. I. Reflux resolution and urinary tract
infection. J Urol 2006; 175: 716.
Kirsch AJ: Intraluminal ureteral injection 2005. Presented at
annual meeting of Society for Pediatric Urology, San Antonio, Texas, May 21, 2005.
Elmore JM, Scherz HC and Kirsch AJ: Dextranomer/hyaluronic acid for vesicoureteral reflux: success rates after
initial treatment failure. J Urol 2006; 175: 712.
Capozza N, Lais A, Nappo S and Caione P: The role of endoscopic treatment of vesicoureteral reflux: a 17-year experience. J Urol 2004; 172: 1626.
Oswald J, Riccabona M, Lusuardi L, Bartsch G and Radmayr
C: Prospective comparison and 1-year follow-up of a single endoscopic subureteral polydimethylsiloxane versus
dextranomer/hyaluronic acid copolymer injection for
treatment of vesicoureteral reflux in children. Urology
2002; 60: 894.
EDITORIAL COMMENTS
Dx/HA
HIT
RNC
STING
dextranomer/hyaluronic acid
hydrodistention-intraureteral injection
radionuclide cystogram
subureteral polytetrafluoroethylene
injection
REFERENCES
1.
2.
3.
4.
5.
Elder JS, Peters CA, Arant BS Jr, Ewalt DH, Hawtrey CE,
Stenberg A, Hensle TW and Lackgren G: Vesicoureteral reflux: a new treatment algorithm. Curr Urol Rep 2002; 3:
107.
Kirsch AJ, Perez-Brayfield MR and Scherz HC: Minimally
invasive treatment of vesicoureteral reflux with endoscopic
injection of dextranomer/hyaluronic acid copolymer: the
Childrens Hospitals of Atlanta experience. J Urol 2003;
170: 211.
Lavelle MT, Conlin MJ and Skoog SJ: Subureteral injection of
Deflux for correction of reflux: analysis of factors predicting
success. Urology 2005; 65: 564.
Puri P, Chertin B, Velayudham M, Dass L and Colhoun E:
Treatment of vesicoureteral reflux by endoscopic injection
This study is a statistically valid analysis comparing the

intraureteral (HIT) technique with the subtrigonal injection
technique for the endoscopic correction of VUR across the
experiences of 5 pediatric urologists. What is unique about
this analysis is that all injections were performed in a teaching hospital setting with resident participation. It is important to note that the pediatric urologists had residents working with them.
The results of the study indicate a trend towards improved results with the HIT technique. As the authors state,
one must accept that this is not a randomized study of the 2
injection techniques and, therefore, surgeon bias must have
had a role in the type of technique used. In addition, since
the description of the HIT technique followed the more traditional subtrigonal injection technique, it is likely that each
surgeon had already accumulated a variable degree of experience with endoscopic injections before using the HIT approach, which may account for the trend towards better
results with the HIT technique.
Another interesting finding is that the cumulative success rate was significantly lower than reported in other
similar sized series, as the authors note. This is not attributable to the severity of reflux. It could be attributable to
surgeon experience or the participation of residents in the
procedures because other studies did not address this important variable.
1706
The question of the role of voiding dysfunction in the

success or failure of injectable therapy for VUR remains
unresolved. Studies have demonstrated a correlation between dysfunctional voiding established on clinical grounds
with failure of a single injection, whereas in other studies no
such correlation has been found. Our multicenter study of
failed injectable therapy for reflux did not reveal a correlation with dysfunctional voiding but it demonstrated a lower
success rate with a second injection in children with symptoms of dysfunctional voiding.1 Lackgren et al found that
children with dysfunctional voiding and VUR who were successfully treated with dextranomer microspheres had a high
rate of resolution of voiding symptoms without specific therapy directed at dysfunctional voiding.2
This raises concern about the relationship between dysfunctional voiding and VUR. However, what is needed is a
controlled, randomized study of VUR, dysfunctional voiding
and endoscopic correction.
The authors add to our understanding of endoscopic correction of reflux, but also generate more questions to be
answered. Further studies such as this one should help
define the role of endoscopic correction in our armamentarium of reflux correction.
Anthony A. Caldamone
Brown Medical School
Providence, Rhode Island
1.
Higham-Kessler J, Reinert SE, Snodgrass WT, Hensle TW,

Koyle MA, Hurwitz RS et al: A review of failures of endo-
2.
scopic treatment of vesicoureteral reflux with dextranomer

microspheres. J Urol 2007; 177: 710.
Lackgren G: Unpublished data.
These authors address the impact of resident trainees on the

outcome. This issue should probably be addressed in every
study of surgical technique, of which most come from centers
where residents are trained. One is puzzled by the lack of
correlation between voiding dysfunction and outcome in
light of the consistent association between them in other
studies of Deflux injection, spontaneous resolution or open
surgical correction of reflux. If true, this conclusion would be
welcome news but it can only be substantiated if the study
includes data on the frequency and severity of voiding dysfunction in the study population and the efficacy of perioperative efforts to manage it.
The trend toward superior results with HIT could result from greater surgeon experience because it occurred
later in the series. The confusion could have been resolved
by a prospective, randomized study with appropriate
stratification for certain variables. Each inadequate, nonrandomized study that we publish raises more questions
than it answers and leaves the practitioner scratching his
head and saying, How does this affect the cases I am
doing next week?
William F. Tarry
Department of Urology and Pediatrics
West Virginia University
Morgantown, West Virginia
Histopathological Changes Associated With Dextranomer/

Hyaluronic Acid Injection for Pediatric Vesicoureteral Reflux
Jonathan C. Routh, Richard A. Ashley, Thomas J. Sebo, David R. Vandersteen, Jeffrey Slezak
and Yuri Reinberg*
From the Departments of Urology (JCR, RAA), Pathology (TJS) and Biostatistics (JS), Mayo Clinic, Rochester and Division of Urology,
Pediatric Surgical Associates (DRV, YR), Minneapolis, Minnesota
Purpose: Few studies have examined the medium and long-term histological changes associated with periureteral injection
of dextranomer/hyaluronic acid copolymer (Deflux). We present the results of a histological review of a series of distal
ureteral excisions in patients undergoing ureteroneocystostomy after failed dextranomer/hyaluronic acid injection.
Materials and Methods: All patients undergoing ureteroneocystostomy after failed dextranomer/hyaluronic acid injection(s) at 1 institution were eligible for this study. Excised ureteral segments were histologically examined by a single
urological pathologist. An immunohistochemical battery was used for each specimen, including hematoxylin and eosin, CD3,
CD20, MIB-1 and trichrome stains. Pathological criteria included the presence, location and intensity of fibrosis, giant cell
reaction, chronic inflammation, free dextranomer/hyaluronic acid, and CD3, CD20 and MIB-1 staining. Pathological features
were correlated with the time from injection to surgical excision.
Results: The ureters of 16 children with a mean age of 4.5 years were examined. Median time from injection to implant
excision was 8 months. Giant cell reaction was present in 94% of patients and it was typically located in the serosa. No
histological or immunophenotypical feature correlated with the duration of implantation except CD3 and CD20 lymphocyte counts, which increased with time from injection (p 0.06 and 0.02, respectively).
Conclusions: Dextranomer/hyaluronic acid appears to be stable and safe for use after 3 to 22 months of followup of
subureteral injection. The periureteral inflammatory reaction increases with time, although no increases in nuclear turnover
or fibrosis were detected.
Key Words: ureter, vesico-ureteral reflux, hyaluronic acid, inflammation, fibrosis
esicoureteral reflux is a common and controversial

problem in pediatric urology, affecting up to 1% of all
children in the United States.1 Traditionally patients
with VUR underwent prolonged courses of antibiotic prophylaxis or open ureteral reimplantation. However, this dogma
was recently challenged by the advent of newer endoscopic
treatment methods and injectable materials.2 4 Specifically
the use of Dx/HA copolymer has greatly increased in this country in the last decade. Since its approval by the Food and Drug
Administration in 2001, multiple series of Dx/HA use have
been published, showing varying success rates.4 7 Few problems or complications have been reported due to Dx/HA use.8,9
However, relatively few studies have explored the histopathological effects of the implantation of Dx/HA in pediatric bladders.10,11
According to the United States Census Bureau the average American life span is now 77.6 years. Because injected
Dx/HA is expected to remain in vivo for the remaining decades of a patient life, it would appear that the long-term
effects of these substances are significantly under studied.
Therefore, we present our experience with patients undergoing subureteral and/or intraureteral Dx/HA injection, and
subsequent histopathological analysis.
Submitted for publication October 22, 2006.

* Correspondence: 2545 Chicago Ave. South, Suite 104, Minneapolis, Minnesota 55404 (telephone: 612-813-8000; FAX: 612-8138005; e-mail: yurir@yahoo.com).
0022-5347/07/1784-1707/0

Patient Selection and Clinical Features
Following approval by the institutional review board we
reviewed the charts of all patients undergoing open reimplantation after failed Dx/HA injection(s) at a single pediatric institution. Each child undergoing Dx/HA injection at
this institution is enrolled in a database maintained by
study nurses and 1 of us (JCR). The database includes preoperative and postoperative voiding cystourethrogram results, age, gender, surgeon, injection technique, date of procedure(s), presence of dysfunctional voiding, neurogenic
bladder, pre-injection surgical interventions, perioperative
urinary tract infection(s), ureteral duplication anomalies,
periureteral (Hutch) diverticula, amount of Dx/HA injected
per ureter and postoperative complications, including obstruction. Followup voiding cystourethrogram and renal ultrasound are performed 3 months following injection. Cases
are defined as failing injection therapy if they show no
change bilaterally or a worse grade of reflux in either ureter
on postoperative imaging.
At followup consultation after failed injection we discuss the
available options with all patients and families. This includes
observation with continued antibiotics, observation without
continued antibiotics, repeat Dx/HA injection and open surgical reimplantation. Patients choosing this latter option were
eligible for inclusion in this study.
1707
Vol. 178, 1707-1710, October 2007

Printed in U.S.A.
DOI:10.1016/j.juro.2007.03.165
1708
HISTOLOGICAL CHANGES AFTER DEXTRANOMER/HYALURONIC ACID INJECTION
Pathological Features
Excised distal ureteral segments with periureteral tissues
were examined by a single genitourinary pathologist (TJS). A
histological and immunohistochemical battery of stains was
used on each tissue block, including hematoxylin and eosin for
routine morphology, CD3 for T lymphocytes, CD20 for B lymphocytes, MIB-1 as a marker for cell turnover and trichrome to
assess for fibrosis. The pathological criteria evaluated were the
presence, location and intensity of morphological abnormalities, fibrosis, GCR, chronic inflammation, eosinophilic infiltration, free Dx/HA, and CD3, CD20 and MIB-1 immunostaining.
Free Dx/HA was defined as the presence of Dx/HA globules
without a surrounding pseudocapsule of giant cells and fibrous
tissue. CD3, CD20 and MIB-1 cell counts were defined semiquantitatively as the percent (in 5% intervals) of the total cell
population present per high power field and qualitatively as
none, focal, moderate or severe staining. Pathological features
were then correlated with the time from injection to surgical
excision. For patients undergoing more than 1 injection the date
of the first injection was used for calculating implant duration.
Statistical analysis was performed using SAS software.
Temporal relationships were evaluated using Spearmans
correlation tests. All p values were 2-sided and due to the
relatively small number of patients assessed p 0.10 was
considered statistically significant.
RESULTS
Clinical Results
Distal ureteral segments of 1 boy and 15 girls (total of 30
ureters) were examined. Mean patient age was 4.5 years
(range 0.4 to 9.7). Median time from injection to implant excision was 8 months (range 3 to 22). VUR grade was unilateral
grade I to bilateral grade IV with no patient having grade V
VUR. Three ureters were grade I VUR, 8 were grade II, 14
were grade III and 5 were grade IV. Three patients had anatomical anomalies, including a Hutch diverticulum and complete ureteral duplication in 1 each, and the 2 conditions in the
remaining patient. Dysfunctional voiding was present in 5
patients (31%). The mean injected volume of Dx/HA was 1.0 cc
per ureter. Of the patients 14 (88%) underwent bilateral injection, 15 (94%) underwent 1 Dx/HA injection and the remaining
patient received 2 injections before proceeding to open surgery.
No patient experienced any perioperative complications as a
result of Dx/HA implantation or ureteroneocystostomy. No
clinical covariate (age, gender, VUR grade, amount of Dx/HA
injected, voiding dysfunction or anatomical anomaly) correlated with the duration of Dx/HA implantation or histological
findings.
Histological and Immunohistochemical Results
The most common histological finding, GCR, was present in 15
patients (94%) and it was located in the serosa in 13 (87%).
Figure 1 shows a representative GCR. Free Dx/HA was noted
in 4 patients (25%) and a pseudocapsule surrounded the
Dx/HA microspheres in 12 (75%). Focal chronic inflammation
was present on hematoxylin and eosin staining in 8 patients
(50%), while moderate inflammation was present in only 1
(6%). No patient demonstrated severe inflammation on hematoxylin and eosin examination. No patients showed significant
FIG. 1. Intense periureteral giant cell reaction encapsulating dextranomer microspheres (arrow). Reduced from 200.
morphological atypia or eosinophilic infiltrate. None of the

histological features listed correlated with the duration of
Dx/HA implantation or with clinical covariates.
MIB-1 staining was present in 0% to 20% of cells per patient. Only 1 patient demonstrated no staining throughout all
slides reviewed and the remaining patients showed focal to
moderate staining. MIB-1 staining did not correlate with the
duration of implantation (p 0.6). As assessed by trichrome
staining, fibrosis was present in all specimens and in 40% to
80% of each excised specimen (fig. 2). Fibrosis levels did not
appear to change with the duration of implantation (p 0.8).
CD3 T cells and CD20 B cells were present in all patients
at the ureteral lumen (fig. 3). The mean CD3 T-cell count was
14% (range 1% to 50%) and the mean CD20 B-cell count was
2.9% (range 0% to 10%). The 2 parameters ranged from no
staining to moderate staining with no patient showing a strong
lymphocyte presence. CD3 and CD20 lymphocyte counts
increased concurrent with time from injection (p 0.06 and
0.02, respectively).
DISCUSSION
In 1995 Stenberg and Lackgren introduced their experience
with Dx/HA in animal models and in 75 children, reporting
a 68% cure rate for grades IIIIV VUR at 3 months.12 In
2001 Dx/HA was approved by the United States Food and
Drug Administration for use in children with primary
grades IIV VUR. Multiple series since that time have
shown good to excellent success rates of 64% to 92% depending on the injection technique used and the population studied.4 7 Indeed, an increasing number of clinicians are now
choosing primary endoscopic management of VUR in children over the more traditional strategies of antibiotic prophylaxis or ureteroneocystostomy.2,3
It is reasonable to assume that children with VUR undergoing Dx/HA injection should have a normal projected
life span, which is currently estimated to be 77.6 years in the
United States.13 The average age at injection is reported to
be about 6 years,14 which implies that injected Dx/HA
should remain in vivo more than 7 decades. Clearly it is
incumbent on pediatric urologists to prove the long-term
FIG. 2. CD3 lymphocytes 3 (A) and 8 (B) months after Dx/HA

injection, and CD20 lymphocytes 3 (C) and 8 (D) months after
injection. Reduced from 200.
safety of an injected material that is expected to remain in

place for several decades.
However, few reports exist of histopathological investigations after Dx/HA injection. Stenberg and Lackgren performed
a histopathological review of Dx/HA injections in animals during a 16-month time span.10 No worrisome histopathological
changes were noted in or around the implants. Specifically no
eosinophilia or nuclear changes suggesting mutagenesis were
seen. The major histological effect observed was the early ingrowth of fibroblasts and inflammatory cells, and the subsequent deposition of collagen fibers with concurrent GCR. They
then expanded these findings by examining a series of distal
ureteral stumps from 13 children undergoing ureteroneocystostomy following failed primary Dx/HA injection, comparing
these results to those in a cohort of patients undergoing reimplantation who did not receive Dx/HA injection.11 In these
patients mild chronic periureteral inflammation (33%), pseudoencapsulation of dextranomer microspheres (43%) and prominent GCR (100%) were noted.
Several novel aspects of the current study bear mention.
To our knowledge we present the first use of immunohistochemical methods to evaluate Dx/HA injection sites for lymphocyte infiltration, as assessed by CD3 and CD20 stains,
and nuclear turnover, as assessed by MIB-1 staining. Our
observed low cell turnover rates (MIB-1) correlate well with
the universally diploid periureteral tissues noted by Stenberg and Lackgren.10 This reinforces the premise that
Dx/HA injection, at least in the short-term to medium-term,
does not carry a significant mutagenic or carcinogenic potential. This is further bolstered by the observation that
lymphocyte infiltration is only minimally present after
Dx/HA injection, as demonstrated by CD3 and CD20 counts.
A second novel aspect of this study is the observed increasing T and B-cell inflammatory response with time, which is the
only pathological feature that correlated with the duration of
implantation. A similar effect was noted in the past with polytetrafluoroethylene.15 This effect has led some groups to speculate that the inflammatory increase could theoretically serve
as a nidus for neoplastic transformation. As Virchow observed
in 1863, chronic inflammation is a cancer promoter and this
concept has since been proved in numerous malignancies, no-
1709
tably including bladder cancer.16 Although inflammatory levels increased in our patients with time, it should be reiterated
that only low levels of inflammation were present and our level
of followup was not adequate to allow speculation on the longterm risk of carcinogenesis. Further long-term studies to clarify this point are clearly required. However, currently we cannot conclude that there is any evidence of future carcinogenesis
due to Dx/HA injection.
The primary limitation of this study is that we only
examined children in whom Dx/HA injection had failed.
Thus, it is impossible to say with certainty that our observations of periureteral inflammation were indeed a result of
Dx/HA injection and not a result of the underlying VUR in
patients, which was resistant to endoscopic management.
However, our results correlate with and are validated by a
previous series by Stenberg et al,11 whose study included a
control group of refluxing, noninjected ureters and who did
not observe any inflammatory changes in noninjected ureters. Thus, it appears likely that our results represent a
Dx/HA mediated phenomenon. To our knowledge this finding has not been previously reported with Dx/HA or other
submucosal injectable materials. However, our correlations
with the duration of implantation must be interpreted with
caution since they do not represent a true longitudinal study
of histological change. To accomplish this, repeat biopsies
would be needed from the injection sites in each child and
obviously this is neither ethically nor practically feasible.
Thus, although our findings are limited, to our knowledge
they represent the only histopathological assessment performed to date of human Dx/HA injection sites, correlated
with the duration of implantation.
The small sample size of our cohort prevented the use of
multivariate analysis to examine the relative impact of various
factors on histological findings. Therefore, it is possible that
voiding dysfunction or increasing VUR grade may explain part
or all of the observed increase in periureteral inflammation,
rather than increased time from Dx/HA injection. However, no
association was observed between these factors on analysis.
The lack of short-term or long-term followup represents
another noteworthy limitation. We cannot comment on any
changes that may occur at less than 3 months or at more
than 2 years because all study patients were histologically
FIG. 3. Trichrome stain 3 months after Dx/HA injection. Reduced

from 200.
1710
assessed only during that period. This is important in terms

of our observations of Dx/HA related fibrosis. It is clear that
fibrosis increases early in the post-injection process as the
sodium hyaluronan solution disperses, and fibroblast ingrowth and collagen deposition occurs and later stabilizes.
This scenario was previously reported by Stenberg and
Lackgren.10 Likewise although the lack of change in MIB-1
values with time is reassuring, it should not be interpreted
as a true long-term assessment of the potential of Dx/HA to
induce proliferative or potentially neoplastic changes in pediatric bladders as they age. Such a phenomenon can only be
assessed using long-term longitudinal studies and, since this
product has only been approved for use in the United States
for 5 years, it would be imprudent to state that no further
changes could occur with time. Nonetheless, our findings
support prior studies showing no evidence of a carcinogenic
or mutagenic potential for Dx/HA. Further studies are required to better evaluate our findings in regard to periureteral lymphocytic infiltrate and its long-term significance.
2.
3.
4.
5.
6.
7.
CONCLUSIONS
Dx/HA appears to be histologically and immunohistochemically stable in the short to medium term following subureteral injection. Periureteral inflammatory reaction
slightly increases with time, although no increases in nuclear turnover or fibrosis were detected. Although the longterm safety of Dx/HA could not be definitively assessed by
this study, we detected no worrisome histological or immunohistochemical features. Further studies are required to
assess the meaning of the observed periureteral lymphocytic
infiltrate and its changes with time.
8.
9.
10.
11.
ACKNOWLEDGMENTS
Dr. Steven Schatz contributed to this project.
12.
13.

Dx/HA dextranomer/hyaluronic acid copolymer
GCR giant cell reaction
REFERENCES
1.
Hurwitz RS et al: Pediatric Vesicoureteral Reflux Guide-
14.
15.
16.
lines Panel summary report on the management of primary

Puri P, Pirker M, Mohanan N, Dawrant M, Dass L and
Colhoun E: Subureteral dextranomer/hyaluronic acid injection as first line treatment in the management of high
grade vesicoureteral reflux. J Urol 2006; 176: 1856.
Stenberg A, Hensle TW and Lackgren G: Vesicoureteral reflux:
a new treatment algorithm. Cur Urol Rep 2002; 3: 107.
Kirsch AJ, Perez-Brayfield M, Smith EA and Scherz HC: The
modified sting procedure to correct vesicoureteral reflux:
improved results with submucosal implantation within the
Lackgren G, Wahlin N, Skoldenberg E and Stenberg A: Longterm followup of children treated with dextranomer/hyaluronic acid copolymer for vesicoureteral reflux. J Urol 2001;
166: 1887.
Kirsch AJ, Perez-Brayfield MR and Scherz HC: Minimally
170: 211.
Routh JC, Vandersteen DR, Pfefferle H, Wolpert JJ and
Reinberg Y: Single center experience with endoscopic
management of vesicoureteral reflux in children. J Urol
2006; 175: 1889.
Bedir S, Kilciler M, Ozgok Y, Deveci G and Erduran D: Longterm complication due to dextranomer based implant: granuloma causing urinary obstruction. J Urol 2004; 172: 247.
Snodgrass WT: Obstruction of a dysmorphic ureter following
dextranomer/hyaluronic acid copolymer. J Urol 2004; 171:
395.
Stenberg A, Larsson E, Lindholm A, Ronneus B and Lackgren
G: Injectable dextranomer-based implant: histopathology,
volume changes and DNA-analysis. Scand J Urol Nephrol
1999; 33: 355.
Stenberg A, Larsson E and Lackgren G: Endoscopic treatment
with dextranomer-hyaluronic acid for vesicoureteral reflux:
histological findings. J Urol 2003; 169: 1109.
Stenberg A and Lackgren G: A new bioimplant for the endoscopic treatment of vesicoureteral reflux: experimental and
short-term clinical results. J Urol 1995; 154: 800.
Hoyert DL, Kung HC and Smith BL: Deaths: Preliminary Data
for 2003. 15. Atlanta: Centers for Disease Control 2005; pp
1 48.
Reinberg Y: Single center experience with endoscopic
management of vesicoureteral reflux in children. J Urol
2006; 175: 1889.
Dewan PA: Is injected polytetrafluoroethylene (polytef) carcinogenic? Br J Urol 1992; 69: 29.
Balkwill F and Mantovani A: Inflammation and cancer: back to
Virchow? Lancet 2001; 357: 539.
New Contralateral Vesicoureteral Reflux

After Endoscopic Correction of Unilateral Reflux
Is Routine Contralateral Injection Indicated at Initial Treatment?
Maria Menezes, Nochiparambil Mohanan, Juliana Haroun, Eric Colhoun and Prem Puri*
From the National Childrens Hospital, Dublin, Childrens Research Centre, Our Ladys Childrens Hospital and University College
Hospital, Dublin, Ireland
Purpose: As more and more pediatric urologists use endoscopic therapy as a primary treatment option for vesicoureteral
reflux, newer indications for this procedure are being considered. Recently it was suggested that contralateral nonrefluxing
ureters should be treated prophylactically in patients undergoing unilateral endoscopic correction of vesicoureteral reflux. We
analyzed the incidence of newly diagnosed contralateral reflux after endoscopic correction of unilateral reflux and identified
possible risk factors for its development.
Materials and Methods: Between 1996 and 2004, 662 patients underwent endoscopic correction of unilateral grades II to
V vesicoureteral reflux. Of the ureters 97% had grades III to V reflux. The tissue augmenting substance used for endoscopic
treatment was polytetrafluoroethylene from 1996 to 2000 and dextranomer/hyaluronic acid from 2001 to 2004. There were
203 males (30.7%) and 459 females (69.3%) with an age at endoscopic treatment of 2 months to 11 years. Voiding
cystourethrograms performed 3 months after endoscopic treatment of unilateral vesicoureteral reflux were analyzed in all
patients to document newly diagnosed contralateral reflux.
Results: A total of 67 children (10.1%), including 18 boys and 49 girls, showed new contralateral reflux on voiding cystourethrogram after endoscopic correction of unilateral reflux. Contralateral VUR was grades I to IV in 16 (23.9%), 17 (25.4%), 27 (40.3%)
and 7 patients (10.5%), respectively. There was no correlation between the severity of ipsilateral reflux and the development of
contralateral reflux. Patient age and gender did not influence the development of new contralateral reflux.
Conclusions: The low incidence and lower grade of newly diagnosed contralateral vesicoureteral reflux after endoscopic
correction of unilateral reflux does not support prophylactic treatment of nonrefluxing contralateral ureters.
Key Words: ureter, vesico-ureteral reflux, endoscopy, hyaluronic acid
s increasingly more pediatric urologists use endoscopic therapy as a primary treatment option for
VUR, newer indications for this procedure are being
considered. The minimally invasive nature of the procedure
and favorable results make it an attractive option for many
patients. The development of NCVUR has been reported to
occur in 5% to 25% of patients undergoing unilateral reimplantation.1 4 However, few studies have reported the incidence of new contralateral VUR following endoscopic correction of unilateral VUR.5 8 Recently it was suggested that
the contralateral nonrefluxing ureter should be treated prophylactically in patients undergoing unilateral endoscopic
correction of VUR, particularly in some patients, such as
girls younger than 5 years.5 We analyzed the incidence of
newly diagnosed contralateral VUR after endoscopic correction of unilateral VUR and identified possible risk factors for
its development.
96.7% of ureters had grades III to V reflux. Patients with

secondary reflux due to posterior urethral valves, neurogenic bladder and bladder exstrophy were excluded from
study. Reflux was graded according to the international
classification. The tissue augmenting substance used for
endoscopic treatment was polytetrafluoroethylene from
1996 to 2000 and dextranomer/hyaluronic acid from 2001 to
2004. The paste was injected using a disposable 4Fr Puri
catheter through a cystoscope, as previously described.9 The
needle was introduced under the bladder mucosa 2 to 3 mm
below the affected ureteral orifice and the paste was injected
to create a volcanic bulge at the orifice, giving the orifice a
slit-like appearance. There were 203 males (30.74%) and 459
females (69.3%) with an age at endoscopic treatment of 2
months to 11 years. VCUGs performed 3 months after endoscopic treatment of unilateral VUR were analyzed in all
patients to assess the resolution of unilateral VUR and
document newly diagnosed contralateral VUR. The chisquare test of significance was used to assess the various
risk factors for NCVUR.
Between 1996 and 2004, 662 patients underwent endoscopic

correction of unilateral grades II to V VUR. In these cases
RESULTS
* Correspondence: Childrens Research Centre, Our Ladys
Childrens Hospital, Crumlin, Dublin-12, Ireland (telephone:
0035314096420; e-mail: prem.puri@ucd.ie).
0022-5347/07/1784-1711/0
Ipsilateral VUR was corrected after the first, second and

third injection in 547 (82.6%), 93 (14.1%) and 22 patients
(3.3%), respectively. A total of 67 children (10.1%), including
1711
Vol. 178, 1711-1713, October 2007

Printed in U.S.A.
DOI:10.1016/j.juro.2007.03.177
1712
CONTRALATERAL REFLUX AFTER ENDOSCOPIC CORRECTION OF UNILATERAL REFLUX
18 boys and 49 girls, showed new contralateral VUR on

VCUG after endoscopic correction of unilateral reflux. Contralateral VUR was grade I to IV in 16 (23.9%), 17 (25.4%),
27 (40.3%) and 7 patients (10.5%), respectively. There was
no correlation between the severity of ipsilateral reflux and
the development of contralateral reflux (table 1, p 0.05).
Patient age and gender did not influence the development of
new contralateral VUR (table 2, p 0.05). NCVUR was
treated conservatively in 32 of 33 patients with grade I or II
contralateral VUR. One patient with grade II contralateral
VUR had recurrent urinary tract infections and was treated
endoscopically, while 34 with grade III or IV contralateral
VUR were treated endoscopically.
DISCUSSION
The incidence of NCVUR following endoscopic correction of
unilateral VUR is reported to be 4.5% to 13%.5 8 In our
study the incidence of NCVUR following unilateral correction of VUR was 10.1%. In comparison, the incidence of
NCVUR following unilateral open reimplantation of the ureter is reported to be 5% to 21%.1 4
The exact etiology of NCVUR is not fully understood.
Various theories have been put forward about the development of contralateral VUR. Some investigators suggested
that surgical distortion of the contralateral trigone during
unilateral reimplantation may result in new onset contralateral reflux. However, since there is no distortion of the
trigone in endoscopically treated patients, this mechanism is
unlikely to be responsible for NCVUR. Diamond et al suggested that the elimination of a pop-off mechanism could be
a cause of NCVUR following open reimplantation based on
the supposition that VUR acts as a pressure relief mechanism and correcting it predisposes the contralateral ureter
to new VUR.1 Furthermore, they found a significant trend
toward NCVUR in patients with higher grades of ipsilateral
reflux. In our study there was no correlation between the
severity of ipsilateral reflux and the development of contralateral reflux. The low incidence of contralateral reflux
after endoscopic correction of even high grade reflux suggests that it is unlikely that a pop-off mechanism is the
cause of new contralateral reflux. Sparr et al found that new
contralateral reflux developed in a third of the patients
initially diagnosed with unilateral disease while they were
being followed nonoperatively, suggesting that the postoperative development of contralateral reflux may be a manifestation of the natural history of unilateral disease, rather
than a result of the surgical procedure.10 In addition, it is
possible that contralateral reflux may be intermittent and,
hence, it could be missed on initial VCUG. Liu et al suggested that contralateral reflux may be preexistent and contralateral renal scarring on preoperative dimercapto succinic acid scan might predict the patients who would be
TABLE 1. Ipsilateral VUR severity and development of

contralateral VUR
Ipsilat VUR Grade
No. Cases
No. Contralat VUR (%)
2
3
4
5
29
404
208
21
3 (10.4)
41 (10.2)
21 (10.1)
2 (9.5)
TABLE 2. Age as risk factor for NCVUR

Age Group (yrs)
No. Ipsilat VUR Correction
No. NCVUR (%)
Younger than 1
15
510
Older than 10
92
291
118
18
10 (10.8)
25 (8.6)
9 (7.6)
likely to show contralateral VUR.3 However, all patients in

that study underwent open reimplantation.
In our study we also determined possible risk factors for
the development of NCVUR. Some groups suggested that
patient age and gender could be risk factors and girls
younger than 5 years had the highest incidence of NCVUR.5
However, in our study there was no statistically significant
difference in patient gender or in the various age groups
with respect to NCVUR development. In addition, the preoperative grade of ipsilateral VUR did not appear to influence the development of NCVUR.
It was recently suggested that contralateral nonrefluxing
ureters should be treated prophylactically in certain groups
of patients undergoing unilateral endoscopic correction of
VUR, such as girls younger than 5 years.5 However, this
would certainly be associated with increased cost and it
would also result in the unnecessary treatment of a large
number of normal nonrefluxing contralateral ureters. The
majority of the new contralateral refluxing ureters had
lower grades of reflux in our series. Only half of the new
contralateral refluxing ureters had grade III/IV reflux, requiring further endoscopic treatment. In view of the low
incidence of NCVUR in this series prophylactic treatment of
nonrefluxing contralateral ureters seems unnecessary.
CONCLUSIONS
We were unable to identify any particular risk factors for
NCVUR development. The low incidence and lower grade of
newly diagnosed contralateral VUR after endoscopic correction of unilateral reflux does not support prophylactic treatment of nonrefluxing contralateral ureters.

NCVUR new contralateral VUR
REFERENCES
1.
Diamond DA, Rabinowitz R, Hoenig D and Caldamone AA: The

mechanism of new onset contralateral reflux following unilateral ureteroneocystostomy. J Urol 1996; 156: 665.
2. Hoenig DM, Diamond DA, Rabinowitz R and Caldamone AA:
Contralateral reflux after unilateral ureteral reimplantation. J Urol 1996; 156: 196.
3. Liu C, Chin T and Wei C: Contralateral reflux after unilateral
ureteral reimplantationpreexistent rather than newonset reflux. J Pediatr Surg 1999; 34: 1661.
4. Noe HN: The risk and risk factors of contralateral reflux following repair of simple and unilateral primary reflux.
J Urol 1998; 160: 849.
5. Elmore JM, Kirsch AJ, Lyles RH, Perez-Brayfield MR and
Scherz HC: New contralateral vesicoureteral reflux follow-
CONTRALATERAL REFLUX AFTER ENDOSCOPIC CORRECTION OF UNILATERAL REFLUX
6.
7.
ing dextranomer/hyaluronic acid implantation: incidence

and identification of a high risk group. J Urol 2006; 175:
1097.
Kumar R and Puri P: Newly diagnosed contralateral reflux
after successful unilateral endoscopic correction: is it due to
the pop-off mechanism? J Urol 1997; 158: 1213.
Kirsch AJ, Perez-Brayfield MR and Scherz HC.: Minimally
170: 211.
8.
1713

Reinberg Y: Single center experience with endoscopic management of vesicoureteral reflux in children. J Urol 2006;
175: 1889.
9. Puri P: Endoscopic correction of primary vesicoureteral reflux
by subureteric injection of polytetrafluoroethylene. Lancet
1990; 335: 1320.
10. Sparr KE, Balcom AH and Mesrobian HO: Incidence and natural history of contralateral vesicoureteric reflux in patients presenting with unilateral disease. J Urol 1998; 160:
1023.
Endoscopic Treatment of Moderate and High Grade

Vesicoureteral Reflux in Infants Using Dextranomer/Hyaluronic Acid
Prem Puri,* Nochiparambil Mohanan, Maria Menezes and Eric Colhoun
From the National Childrens Hospital and Childrens Research Centre, Our Ladys Childrens Hospital, University College Dublin,
Dublin, Ireland
Purpose: Renal parenchymal injury in vesicoureteral reflux occurs early, in most patients before age 3 years. It is generally
believed that early prevention of urinary tract infections may decrease the amount of renal parenchymal damage. Endoscopic
treatment for vesicoureteral reflux has become an established alternative to long-term antibiotic prophylaxis and ureteral
reimplantation. We prospectively evaluated the effectiveness and safety of dextranomer/hyaluronic acid copolymer for high
grade vesicoureteral reflux in infants.
Materials and Methods: A total of 276 infants, including 124 boys and 152 girls, with a median age of 7 months (range 2
to 12) underwent endoscopic treatment for vesicoureteral reflux with dextranomer/hyaluronic acid. Reflux was detected in
225 infants following investigation for urinary tract infection, in 46 following screening for sibling vesicoureteral reflux and
in 5 for prenatally diagnosed hydronephrosis. Vesicoureteral reflux was unilateral in 85 infants and bilateral in 191 (467
refluxing ureters). Reflux was grade II in 14 ureters, grade III in 188, grade IV in 248 and grade V in 17. Dimercapto-succinic
acid scan was performed in 250 infants and it demonstrated renal scarring in 43 (17.2%).
Results: Vesicoureteral reflux completely resolved in 373 ureters (79.9%) after a single injection of dextranomer/hyaluronic
acid and it was downgraded to grade I in 21 (4.4%). Of the ureters 65 (14%) required a second injection and 8 (1.7%) required
a third injection to resolve reflux. All patients underwent endoscopic treatment on an outpatient basis. There were no
complications except in 1 patient, who was readmitted to the hospital the next day with acute pyelonephritis. Three children
had urinary tract infections during followup, of whom 1 was found to have recurrent reflux on investigation.
Conclusions: Endoscopic treatment with dextranomer/hyaluronic acid is safe and highly effective for eradicating high grade
vesicoureteral reflux in infants. Early intervention in infants with high grade reflux may change its natural history and
protect against renal scarring.
Key Words: ureter, hyaluronic acid, vesico-ureteral reflux, kidney, cicatrix
esicoureteral reflux is a common urological abnormality discovered in 29% to 50% of children investigated for
UTI.1 The association of VUR, UTI and renal scarring is
well recognized and reflux nephropathy is a major cause of
childhood hypertension and chronic renal failure.2 Long-term
antibiotic prophylaxis is generally recommended for low grade
VUR (grades I to II) to prevent UTI, while awaiting spontaneous resolution. High grade VUR has a much lower tendency
toward spontaneous resolution. In the European arm of the
International Reflux Study at 10 years of followup VUR resolved in only 52% of children with grades III to IV reflux
treated with antibiotic prophylaxis.3 Of those with bilateral
grade III or IV reflux it resolved in only 39%. The American
Urological Association reflux guidelines indicate that greater
than 50% of unilateral grade IV VUR and up to 90% of bilateral
VUR persist after 5 years of prophylaxis.4
Surgical treatment has been the gold standard for high
grade VUR. However, surgery is not free of complications,
particularly in infants and in high grade VUR with dilated
ureters. Elder et al reported persistent VUR in 19.3% of patients after ureteral reimplantation for grade V reflux.4 Moreover, based on analysis of 33 studies it was reported that the
rate of obstruction after ureteral reimplantation that required

reoperation was 0.3% to 9.1%.
Since 1984, minimally invasive endoscopic treatment has
been successfully used throughout the world to correct VUR in
infants and children as an alternative to surgical intervention
and observational management.5,6 Particularly since Deflux
was approved by the Food and Drug Administration in 2001 as
an acceptable tissue augmenting substance for subureteral
injection, endoscopic treatment has become increasingly popular for managing VUR.
Renal parenchymal injury in VUR occurs early, in most
patients before age 3 years. Most renal scars are present when
reflux is discovered at initial evaluation for UTI. It generally
believed that early prevention of UTI is important to prevent
the occurrence or further progression of renal damage. During
the years we have increasingly performed endoscopic correction of high grade VUR in infants as our first line treatment
because it provides early resolution of reflux at a time when
children are at higher risk for renal scar formation.7 We prospectively evaluated the effectiveness and safety of subureteral
Deflux injection for managing high grade VUR in infants.
* Correspondence: Childrens Research Centre, Our Ladys Hospital for Sick Children, Dublin, Ireland (telephone: 00353 1 4096420;
FAX: 00353 1 4550201; e-mail: prem.puri@ucd.ie).
0022-5347/07/1784-1714/0
Between June 2001 and October 2005, 276 infants underwent endoscopic subureteral injection with DX/HA (Deflux)
1714
Vol. 178, 1714-1717, October 2007

Printed in U.S.A.
DOI:10.1016/j.juro.2007.03.172
REFLUX TREATMENT USING DEXTRANOMER/HYALURONIC ACID

for high grade VUR. Patients with neurogenic bladder, posterior urethral valves or ureterocele were excluded from
study. Of the 276 infants 225 presented with UTI, 46 were
detected on screening because of sibling VUR and 5 were
diagnosed with VUR on investigation for prenatally diagnosed hydronephrosis. There were 124 boys (45%) and 152
girls (55%). Median age at diagnosis was 7 months (range 2
to 12).
VUR was diagnosed by VCUG and reflux was graded
according to the international system of radiographic grading of VUR. Reflux was unilateral in 85 infants and bilateral
in 191 (467 refluxing ureters). Reflux was grade II to V in 14
(3%), 188 (40.3%), 248 (53.1%) and 17 ureters (3.6%), respectively. There were 19 patients with a duplex system, which
was unilateral in 18 and bilateral in 1. Ten ureters had
grade III, 8 had grade IV and 2 had grade V reflux.
The primary indication for treatment was grade III to V.
Infants with grade II reflux were given endoscopic treatment
only if they presented with febrile UTI necessitating hospitalization. In this study only 3 infants (4 refluxing ureters)
with isolated grade II VUR were offered endoscopic treatment. The remaining 10 grade II refluxing ureters were
associated with high grade VUR on the contralateral side.
A 99mtechnetium DMSA scan was performed in 250 children to determine renal scarring. Renal scarring was classified as mildfocal defects with relative uptake between
40% and 45%, moderaterelative uptake between 20% and
40%, and severea shrunken kidney with relative uptake
less than 20%.
Endoscopic treatment was done on an outpatient basis
according to a previously described technique.8 VCUG and
renal ultrasound were performed 3 months after the endoscopic procedure. If VUR had not resolved, the infant was
considered for further endoscopic treatment. Antibiotic prophylaxis was continued until an accepted grade of reflux
resolution was achieved (grade 0 to 1). These children were
followed with renal ultrasound at 1 year and every 2 years
thereafter. VCUG was performed only if they presented with
breakthrough UTIs.
RESULTS
Reflux resolved in 373 ureters (79.9%) after a single endoscopic injection and it was downgraded to grade I in 21
(4.4%). Of the ureters 65 (14%) required a second injection and
8 (1.7%) required a third injection to resolve VUR. Table 1
shows the result of resolution for each VUR grade. Table 2
shows the results of endoscopic treatment in duplex systems
according to VUR grade. The volume of injected Deflux per
ureter in 373 ureters in which VUR resolved completely
after a single injection was 0.2 to 1.1 ml.
TABLE 2. Results of endoscopic VUR treatment in duplex systems

by grade
VUR Grade
No. 1
Injection
No. 2
Injections
No. 3
Injections
Total No.
8
4
2
2
3
0
0
1
0
10
8
2
14
20
III
IV
V
Totals
A total of 12 cases of new contralateral reflux were found,

which was grades I to IV in 3, 4, 3 and 2, respectively. In this
group VUR was corrected endoscopically in 5 children with
grade III/IV VUR. No treatment was given in 7 children with
grades I and II VUR.
Three children had UTI during followup, of whom 1 had
recurrent VUR on investigation. The other 2 patients had
incomplete bladder emptying, as confirmed by ultrasound
showing significant post-void residual urine. There were no
complications except in 1 patient who was readmitted to the
hospital the next day with acute pyelonephritis.
DMSA scan in 250 children revealed renal scarring in 43
(48 kidneys). Table 3 shows the grade of reflux and degree of
renal scarring.
DISCUSSION
The potential causes of renal scar formation are considered
to be 1) reflux of infected urine with interstitial inflammation, 2) sterile reflux, which is usually high grade and may
damage the kidney through mechanical or immunological
mechanisms and 3) abnormal embryological development
with subsequent renal dysplasia.9 The first 2 causes of renal
damage can be prevented by early detection of VUR and
prompt treatment. Currently the damage due to congenital
dysplasia cannot be prevented. However, it is mandatory to
detect VUR early to prevent renal exposure to UTI and avoid
the progression of renal parenchymal damage.9
The various treatment options available for managing
VUR are 1) long-term antibiotic treatment, 2) surgical treatment and 3) minimally invasive endoscopic treatment.
Breakthrough UTIs leading to renal damage, noncompliance
with medications and long treatment duration are the disadvantages of antibiotic treatment. It is widely acknowledged that the likelihood of VUR resolution depends on the
grade of reflux. The American Urological Association reflux
guidelines report showed that grades I and II reflux resolved
in 90% and 80% of children, respectively, at 5 years of
followup.4 However, in patients with unilateral grade IV
reflux the resolution rate was 45% and in those with bilateral grade IV reflux it was less than 10%.
TABLE 3. Reflux grade and renal scarring
TABLE 1. VUR Resolution for each reflux grade

No. Pts (%)
VUR Grade
1 Injection
2 Injections
3 Injections
Total No.
II
III
IV
V
14 (100)
175 (93.1)
194 (78.2)
11 (64.7)
0
12 (6.4)
48 (19.4)
5 (29.4)
0
1 (0.5)
6 (2.4)
1 (5.9)
14
188
248
17
394
65
467
Totals
1715
VUR Grade
II
III
IV
V
Totals
No. Scarring
No. No
Scarring
Mild
Moderate
Severe
Total No.
(% scarring)
11
158
201
7
0
6
17
5
0
0
14
2
0
0
1
3
0
6 (3.7)
32 (13.7)
10 (59)
377
28
16
48 (11.2)
1716
In their study of infants with grades IV and V reflux

Sjstrm et al noted a spontaneous resolution rate of 29% in
boys and 4% in girls in the first year, and an annual resolution rate of only 9% in boys and girls thereafter.10 Despite
antibiotic prophylaxis 50% of children had breakthrough
UTIs, primarily in the infant year and more frequently in
girls than in boys. The high incidence of breakthrough UTIs
in boys can be attributable to uncircumcised males, who are
typical in European cohorts, but this does not explain the
high incidence of UTI in girls. A randomized study comparing antibiotic prophylaxis with endoscopic treatment using
DX/HA copolymer showed success rate of 38% for antibiotic
treatment and 69% for endoscopic treatment.11
Surgical treatment has been the gold standard of treatment for high grade VUR. Ureteral reimplantation is a major surgical procedure necessitating a few days of hospitalization. Surgery is not free of complications and it has a
failure rate of about 20% for grade V VUR. Moreover, a 0.3%
to 9.1% rate of obstruction after reimplantation requiring
reoperation has been reported.4
The concept of endoscopic treatment was introduced by
Puri and ODonnell in 1984 as a minimally invasive treatment for VUR following a successful experimental study in
piglets.12 Following Food and Drug Administration approval
to use DX/HA as a tissue augmenting substance in 2001
endoscopic treatment has become accepted first line of treatment for high grade VUR. Endoscopic treatment with Deflux
has proved to be a safe, effective and well tolerated procedure.13 We recently reported the results of subureteral polytetrafluoroethylene injection as first line treatment for high
grade VUR. The overall success rate for resolving VUR after
the first endoscopic injection of Deflux was 87%.14 Similar
success rates were reported by others.15
Our study population consisted of 124 boys (45%) and 152
girls (55%) with high grade refluxing ureters. This is contrary to most other studies, in which male infants were
predominant and accounted for majority with high grade
reflux. VUR resolved in 100% ureters with grade II, 93.1%
with grade III, 78.2% with grade IV and 64.7% with grade V
reflux after the first subureteral Deflux injection. The overall success rate after a single injection was 84.4%. An important advantage of endoscopic treatment is repeatability
in cases of failure after the first injection. Of 65 ureters 57
that received further endoscopic treatment after initial
treatment failure were cured by a second subureteral injection. The various advantages of endoscopic treatment do not
remain unnoticed by the general population. A recent study
showed that 80% of parents preferred endoscopic treatment
to antibiotic prophylaxis and open surgery when given the
option of any of the 3 treatment modalities after detailed
information was provided.16
It is well recognized that age at diagnosis, high grade
VUR and a history of UTI are the most important risk
factors for renal scarring. Therefore, it seems appropriate to
offer minimally invasive endoscopic treatment to infants
with high grade VUR before further UTIs develop with the
subsequent risk of renal scarring.
infants may provide protection against the renal damage

associated with reflux. We recommend endoscopic treatment
in infants with high grade VUR to prevent possible renal
damage and further complications.

DMSA
DX/HA
UTI
VCUG
VUR
dimercapto-succinic acid
dextranomer/hyaluronic acid
REFERENCES
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
15.
CONCLUSIONS
Endoscopic treatment with Deflux for VUR in infants is a
safe, effective, minimally invasive procedure that can be
performed on an outpatient basis. Early correction of VUR in
16.
Report of the International Reflux Study Committee: medical

versus surgical treatment of primary vesicoureteral reflux:
a prospective international reflux study in children. J Urol
1981; 125: 277.
Marra G, Oppezzo C, Ardission G, Dacco V, Testa S, Avilio
L et al: severe vesicoureteral reflux and chronic renal failure: a condition peculiar to male gender? Data from the
Italkid Project. J Pediatr 2004; 144: 677.
Smellie JM, Jodal U, Hildegard L, Tamminen Mbius T, Hirche
H and Olbing H: Outcome at 10 years of severe vesicoureteric
reflux managed medically: report of the International Reflux
Study in Children. J Pediatr 2001; 139: 656.
Elder JS, Peters CA, Arant BS, Ewalt DH, Hawtrey CE, Hurwitz
RS et al: Pediatric vesicoureteral Reflux Guidelines Panel.
Summary report on the management of primary vesicoureteral reflux in children. J Urol 1997; 157: 1846.
Puri P and Granata C: Multicentre survey of endoscopic treatment of vesicoureteral reflux using polytetrafluoroethylene.
J Urol 1998; 160: 1007.
Kirsch AJ, Perez-Brayfield MR and Scherz HC: Minimally invasive treatment of vesicoureteral reflux with endoscopic injection of dextranomer/hyaluronic acid copolymer: the childrens
hospital of Atlanta experience. J Urol 2003; 170: 211.
Winberg J: Commentary: progressive renal damage from infection with or without reflux. J Urol 1992; 148: 1733.
Puri P: Endoscopic correction of vesicoureteral reflux. Curr
Opin Urol 2000; 10: 593.
Chertin B and Puri P: Familial vesicoureteral reflux. J Urol
2003; 169: 1804.
Sjstrm S, Silln U, Bachelard M, Hanson S and Stokland E:
Spontaneous resolution of high grade infantile vesicoureteral reflux. J Urol 2004; 172: 694.
Cappozza N and Caione P: Dextranomer/hyaluronic acid copolymer implantation for vesicoureteral reflux: a randomised comparison with antibiotic prophylaxis. J Pediatr 2002; 140: 230.
Puri P and ODonnell B: Correction of experimentally produced
vesicoureteric reflux in the piglet by the intra-vesical injection of Teflon. Br Med J 1984; 89: 5.
Lackgren G, Wahlin N, Skoldenberg E and Stenberg A: Long
term follow up of children treated with dextranomer/hyaluronic
acid copolymer for vesicoureteral reflux. J Urol 2001; 166: 1887.
Puri P, Pirker M, Mohanan N, Dawrant M, Dass L and
Colhoun E: Subureteral dextranomer/hyaluronic acid injection as first line of treatment in the management of high
grade vesicoureteral reflux. J Urol 2006; 176: 1856.
Kirsch AJ, Perez-Brayfield M, Smith EA and Scherz HC: Modified STING procedure to correct vesicoureteral reflux: improved results with submucosal implantation within the
Cappozza N, Lais A, Matarazzo E, Nappo S, Patricolo M and
Caione P: Treatment of vesicoureteric reflux: a new algorithm based on parental preference. BJU Int 2003; 92: 285.

EDITORIAL COMMENTS
3.
The introduction of DX/HA has changed our treatment of

VUR. These authors are at the forefront of this technique.
However, have we changed the desired outcome? The authors present their experience using DX/HA for the correction of VUR in children down to age 2 months. We know that
we can but we still do not know if we should.
The authors report no complications related to treatment
age. However, the margin of error decreases as we push the
age envelope. Treating patients down to 48 weeks of age
raises issues related to anesthesia,1 equipment size, technique and liability. They report almost a 6-fold volume range
(0.2 to 1.1 ml) of DX/HA in these smallest of ureters. Surgeons and parents must know that the product package
labeling specifically states that safety and effectiveness for
treating children younger than 1 year has not been established.
The distinction between dysplasia and acquired scarring
continues to be debated, as are the long-term significance of
persistent VUR2 and the cost-effectiveness of treating it.3
Evidenced based medicine should be our goal. The American
Urological Association Pediatric Vesicoureteral Reflux
Guideline Panel meta-analysis in 1997 determined that
there is no evidence that intervention vs prophylaxis improves renal status, the risk of UTI or the ultimate outcome
(reference 4 in article). Has the advent of DX/HA as a new
treatment modality changed that underlying premise? Perhaps the time has arrived for new guidelines.
M. David Bomalaski
Elmendorf Air Force Base
Anchorage, Alaska
1.
American Academy of Pediatrics Section on Anesthesiology

Guidelines for the Pediatric Perioperative Anesthesia Environment. Pediatrics 1999; 103: 512.
2. Cooper, CS, Chung, BI, Kirsch, AJ, Canning, DA and Snyder,
HM III: The outcome of stopping prophylactic antibiotics in
older children with vesicoureteral reflux. J Urol 2000; 163:
269.
1717
Benoit, RM, Peele PB, Cannon GM Jr and Docimo SG: The

cost-effectiveness of dextranomer/hyaluronic acid copolymer
for the management of vesicoureteral reflux. 2. Reflux correction at the time of diagnosis as a substitute for traditional
management. J Urol 2006: 176: 2649.
These authors report that endoscopic treatment for medium

and high grade VUR with DX/HA copolymer in infants is
safe and effective but they have not presented a convincing
argument that early intervention during the first year of life
is necessary or desirable.
Although up front treatment has the advantages of minimizing antibiotics, followup visits and VCUG the author
assertion that early treatment prevents renal scarring is not
supported. Sterile VUR, the foundation on which medical
management is based, is highly unlikely to cause renal deterioration in primary reflux.
The practice of urgently eliminating all middle and high
grade reflux in infants to prevent possible renal damage
does not seem warranted since scars that are already
present at diagnosis rarely progress and new scars rarely
form. Routine infant treatment results in a lot of procedures
in exchange for minimal parenchymal benefit in a minority
of patients.
While they are not common, adverse anesthetic events
happen, particularly in infants younger than 6 months. I
was reminded of this recently when severe bronchospasm
occurred following endoscopic treatment, creating a crisis at
the head of the table.
Endoscopic correction of VUR with DX/HA copolymer is a
useful intervention that, like all others, should be used with
discretion. Indications continue to be controversial, subjective and influenced by parental concerns and physician bias.
In the absence of medical failure I believe that spontaneous
resolution deserves a chance, especially in the first year of
life.
Richard S. Hurwitz
Kaiser Permanente Medical Center
Los Angeles, California
Testis
Scrotal Exploration for Unilateral Nonpalpable Testis
Warren T. Snodgrass,* Selcuk Yucel and Ali Ziada
From the Pediatric Urology Section, University of Texas Southwestern Medical Center at Dallas and Childrens Medical Center Dallas,
Dallas, Texas
Purpose: We previously reported initial scrotal exploration for unilateral nonpalpable testis, followed by laparoscopy when
a viable testis was not found. Although we concluded that scrotal exploration could often make laparoscopy unnecessary, the
knowledge that patients with presumed nubbins would undergo laparoscopy meant that the surgeon did not have to decide
whether findings were definitive. In a second series of consecutive patients laparoscopy was used only when it appeared
indicated.
Materials and Methods: Boys with unilateral nonpalpable testis underwent initial scrotal exploration. Laparoscopy was
only performed when neither a nubbin nor a testis was found, when there was a patent processus vaginalis and/or when gross
findings suggesting a nubbin were thought inconclusive.
Results: There were 43 patients, including 30 (70%) with monorchism and 13 (30%) with an extra-abdominal (7) or
intra-abdominal (6) testis. Of 30 patients found to have monorchism laparoscopy was performed only in 7 (23%), including 2
with scrotal nubbins with a patent processus vaginalis, 1 with a questionable nubbin, 1 with a definitive nubbin but
preoperative magnetic resonance imaging suggesting a testis and 3 with an empty scrotum (intra-abdominal vanished testis
in 1 and inguinal nubbins in 2). The finding of monorchism was strongly predicted by a contralateral descended testicular
length of 1.8 cm or greater.
Conclusions: Scrotal exploration is potentially definitive for the diagnosis and management of extra-abdominal testes and
nubbins in patients with unilateral nonpalpable testis. Laparoscopy is only definitive when an intra-abdominal testis is
found, which occurs in a minority of cases.
Key Words: testis, abnormalities, cryptorchidism, laparoscopy
boy with unilateral nonpalpable cryptorchidism will

be found to have 1 of 4 possible conditions, including
an intra-abdominal testis, an extra-abdominal testis,
an intra-abdominal vanished testis or an extra-abdominal
nubbin. Reported incidences vary but a review of the laparoscopic literature by Merguerian et al of more than 1,500
procedures, including cases of bilateral nonpalpable testes,
showed an intra-abdominal testis in 34%, an extra-abdominal testis in 20% and monorchism in the remaining 46%.1
Exploration is required to establish the correct diagnosis
and reposition a viable undescended testis into the scrotum.
Traditionally surgery was performed through an inguinal
incision but during the last 15 years laparoscopy has been
increasingly used for the diagnosis and management of intraabdominal testes, when present. Otherwise laparoscopy may
reveal a blind-ending vas and vessels above a closed internal
ring, which has been presumed to indicate intra-abdominal
testicular loss, obviating the need for extra-abdominal exploration. Other possible findings are a vas and visibly normal or atretic vessels entering an internal ring that may be
patent or closed, often resulting in inguinal exploration to
distinguish an extra-abdominal testis from a testicular nubbin.
* Correspondence: Pediatric Urology Section, Department of Urology, Childrens Medical Center and University of Texas Southwestern Medical Center, 6300 Harry Hines Blvd., Suite 1401, Dallas,
Texas 75235 (telephone: 214-456-2481; FAX: 214-456-8803; e-mail:
warren.snodgrass@childrens.com).
0022-5347/07/1784-1718/0
Elder noted that extra-abdominal nubbins were found at or

beyond the external ring,2 implying that they were lost from
torsion after descent. Belman and Rushton subsequently reviewed their cases of monorchism explored through scrotal or
inguinal incisions and reported that 52 of 54 had a scrotal
nubbin.3 Laparoscopy or abdominal exploration was performed in only half of these patients, raising concerns that
an abdominal testis could remain undetected using this
approach. Consequently we performed a prospective study
in which consecutive boys with unilateral nonpalpable testis
underwent scrotal exploration first, followed by laparoscopy
in all in whom a viable testis was not found.4 Of 23 cases of
monorchism a scrotal nubbin was diagnosed in 22, while the
remaining case showed an empty scrotum and laparoscopy
demonstrated blind ending vas and vessels above the internal ring.
Although our routine use of laparoscopy when a scrotal
nubbin was detected validated the reliability of scrotal findings, from the surgeon perspective knowing that laparoscopy
would be performed made it unnecessary to decide whether
scrotal findings were conclusive. Similarly Belman and
Rushton stated that they relied on laparoscopy in 55% of
their patients when there was any question as to the findings on extra-abdominal exploration.3 Therefore, in the next
series of boys presenting to us with unilateral nonpalpable
cryptorchidism we began each operation with scrotal exploration and only used laparoscopy for specific indications, such as
a patent processus vaginalis, no nubbin or extra-abdominal
testis, or inconclusive findings. We now report results using
1718
Vol. 178, 1718-1721, October 2007

Printed in U.S.A.
DOI:10.1016/j.juro.2007.05.089
SCROTAL EXPLORATION FOR UNILATERAL NONPALPABLE TESTIS
1719
Of 30 monorchism cases scrotal nubbins occurred in 26 and

inguinal nubbins occurred in 2. Of these 28 extra-abdominal
nubbins scrotal exploration detected 27 with 1 mid inguinal
nubbin initially overlooked. A patent processus vaginalis was
noted in association with 2 scrotal nubbins. Laparoscopy showing an open and closed internal ring in 1 each.
Altogether laparoscopy was performed in 13 cases (30%),
including all 6 of an intra-abdominal testis and 7 of 30 (23%) of
monorchism. Laparoscopy was done in 3 patients with monorchism with negative scrotal exploration, including 2 with intraabdominal vanished testis and 1 subsequently found to have a
mid inguinal nubbin. As mentioned, another 2 boys with scrotal nubbins and a patent processus vaginalis underwent laparoscopy. In another patient with a scrotal nubbin preoperative
magnetic resonance imaging obtained by the referring physician revealed an intra-abdominal testis, and so laparoscopy
was used to further confirm an absent viable testis. Only the
second case in this series was thought to have inconclusive
scrotal findings for diagnosing a nubbin, although the processus vaginalis was closed and laparoscopy confirmed the vas
and vessels entering a closed internal ring.
In 6 of 24 nubbins (25%) initially detected during scrotal
exploration visible hemosiderin (5) or calcification (1) was
noted. Histology in all 28 nubbins showed calcification
and/or hemosiderin in 17, seminiferous tubules in 6, and
vasal and/or epididymal remnants in 17. Therefore, evidence
(calcification, hemosiderin and/or germinal epithelium) that
the specimen represented a testicular remnant was noted in
21 nubbins. Of the other 7 patients with only vaso-epididymal remnants associated with the fibrous nodule laparoscopy was done in 3, confirming no intra-abdominal testis.
Compensatory hypertrophy could be evaluated in 40 of
these 43 patients. The remaining 3 patients were pubertal,
including 2 with monorchism and 1 with a testis. In prepubertal patients with monorchism the greatest measured length of
the descended testis was 2 cm or greater in 17, 1.8 to 1.9 cm in
9 and less than 1.8 cm in 2. Similar measurements in those
with a viable testis were 2 cm or greater in 3, 1.8 to 1.9 cm in
1 and less than 1.8 cm in 8 (table 2). The positive predictive
value of a descended testicular size of 2 cm or greater for
monorchism was 88%. When compensatory hypertrophy of the
contralateral testis was defined as 1.8 cm or greater, the positive predictive value for monorchism remained at 88%.
initial scrotal incision without routine laparoscopy for unilateral nonpalpable testis and suggest an algorithm for managing
this condition based on our total experience.
Consecutive patients with unilateral nonpalpable testis diagnosed at the outpatient clinic and confirmed by examination under anesthesia underwent ipsilateral oblique upper
scrotal incision through a skin crease. No attempt was made
to palpate a nubbin. In every case the long axis of the
contralateral descended testis was measured using a ruler.
Following incision the upper aspect of the scrotum was inspected for a nubbin or for the leading edge of the processus
vaginalis. Dissection was continued to expose the external
ring if neither of these structures was immediately found.
Gubernacular attachments were released from apparent
nubbins and the spermatic cord was freed to the external
ring. These nubbins were then examined for visible hemosiderin and a patent processus vaginalis. When the processus was closed, no further exploration was performed and
the specimen was excised for subsequent histology. The finding of an apparent nubbin with a patent processus vaginalis
led to laparoscopy to rule out an intra-abdominal testis.
If a processus vaginalis was observed without an apparent nubbin, it was inspected for a looping vas. Those with a
vas were then dissected through the scrotal incision to the
internal ring, keeping the sac intact, to facilitate laparoscopic orchiopexy. When an extra-abdominal testis was
noted, orchiopexy was performed through the scrotal incision. If there was neither a testis nor a nubbin, laparoscopy
was performed.
RESULTS
There were a total of 43 patients with a median age of 23
months (range 6 to 154) with unilateral nonpalpable testis
(table 1). A viable testis was found in 13 patients (30%),
which was extra-abdominal and intra-abdominal in 7 and 6,
respectively. Monorchism was diagnosed in 30 patients
(70%), including 28 extra-abdominal nubbins and 2 intraabdominal vanished testes.
All extra-abdominal testes were discovered during scrotal
exploration with 6 undergoing orchiopexy through the scrotal incision and 1 requiring an inguinal incision. Obesity
may have contributed to failure to detect 2 of these 7 extraabdominal testes during preoperative and intraoperative
examinations. All 6 cases of an intra-abdominal testis had a
processus vaginalis without a nubbin found during scrotal
exploration, of which 4 appeared patent and 2 closed. A
looping vas was observed in 2 of those with a patent sac.
DISCUSSION
Scrotal exploration is more likely than laparoscopy to detect
and facilitate the management of the extra-abdominal testes
and testicular remnants found in the majority of patients
with unilateral nonpalpable testis. This is especially true
TABLE 1. Findings at scrotal exploration in 43 patients with unilateral nonpalpable testis

No. Scrotal Exploration
No. Testes
Viable testis:
Extra-abdominal
Intra-abdominal
Monorchism:
Scrotal nubbin
Inguinal nubbin
Intra-abdominal vanished
13 (30%)
7
6
30 (70%)
26
2
2
Nubbin
Testis
PV
N/A
4 Patent, 2 looping vas, 2 closed
26
1
24 Closed, 2 patent
1 Patent
1720

TABLE 2. Ipsilateral testicular presence or absence vs
contralateral descended testis length in 40 cases
No. Contralat Testis Length (cm)
Monorchism
Testis
Less Than 1.8
1.81.9
2 or Greater
2
8
9
1
17
3
when the contralateral descended testis is hypertrophied

since monorchism is found in almost 90% of cases. As previously noted by Elder2 and Belman and Rushton,3 most
testicular remnants are located in the upper scrotum and
they can be readily identified through a scrotal incision. We
have not found condensations of gubernacular tissue or scrotal fat that might resemble a nubbin, and so the differential
diagnosis when a potential nubbin is identified is limited to
a vaso-epididymal structure separated from a testis or a
looping vas. In such cases the processus vaginalis should be
patent, leading to further exploration. Accordingly laparoscopy can be directed to the minority of patients likely to
benefit from the procedure, specifically those with no scrotal
or inguinal nubbin, and/or those with a patent processus
vaginalis found during scrotal exploration.
Enlargement of the contralateral descended testis is the
most helpful clinical sign predicting findings at surgery.
Simple measurements using a ruler were sufficiently reliable to estimate the length of the contralateral testis, which
in most cases of monorchism was 1.8 cm or greater. In fact,
length in the 2 boys with testicular loss and a smaller
descended testis was 1.7 cm. Conversely of 7 patients found
to have a viable testis and a less than 1.8 cm contralateral
descended testis length in 2 was 1.7 cm and in the remaining
5 it was 1.5 cm or less. Koff originally noted the relationship
of compensatory hypertrophy to monorchism, similarly measuring the testis with a ruler or calipers.5 He reported that
the contralateral testis was 2 cm or greater, corresponding
to a volume of 2 cc or greater, when there was no ipsilateral
viable testis, and it was 1.6 cm or less when a viable testis
was found. An intermediate group with atrophic ipsilateral
testes had contralateral measurements of 1.6 to 2 cm. Huff
et al subsequently reviewed their experience with surgically
exposing the contralateral testis and measuring it in 3 dimensions using calipers, from which volume was calculated
by a formula.6 They confirmed that the descended testis was
larger in 1 to 4-year-old children with monorchism vs that in
those with a viable testis but they found a wide overlap
between the 2 groups. While compensatory hypertrophy is
not conclusive evidence for monorchism, obviating the need
for exploration, we found that measuring descended testicular length is useful for predicting findings, counseling the
family and estimating operative time.
An extra-abdominal nubbin presumably represents the
testicular remnant following intrauterine torsion. We previously reported gross findings from visual inspection of the
spermatic cord extending from the nubbin into the external
ring and correlated them with laparoscopic observations.4
The vas and vessels were often but not invariably visible
since in some cases only a fibrous band subtended the nubbin. Laparoscopy in this latter group usually demonstrated a
blind-ending vas and vessels above the internal ring, indicating that this finding most often results from extravaginal
torsion in the scrotum, which twists the cord proximally to
within the abdomen, as predicted by Belman and Rushton.3

However, in our 2 series no nubbin was identified in 2
patients on scrotal exploration but laparoscopy revealed a
blind-ending vas and vessels. While it may be that an extraabdominal nubbin was overlooked, an alternative explanation is that an insult other than extravaginal torsion might
account for some cases of intra-abdominal vanished testis.
Similarly another case most likely not resulting from extravaginal torsion had a vas extending down a patent processus
vaginalis to a scrotal nubbin with laparoscopy demonstrating atretic vessels ending above the internal ring. Finally,
our other case of a scrotal nubbin, distally patent processus
vaginalis, and laparoscopic findings of vas and vessels entering a closed internal ring may have indicated a vascular
insult without torsion or intrauterine intravaginal torsion.
Visible hemosiderin and/or calcification confirm the impression that a structure represents the testicular remnant.
Unfortunately although these indicators of prior ischemic
necrosis were found by microscopy in 68% of removed nubbins, we noted them intraoperatively in only 25% of cases.
Consequently inspection of a nubbin for hemosiderin or calcification is helpful only when these findings are present.
Elder suggested that decision making for nonpalpable
testes could be based on the laparoscopic finding of a patent
vs closed processus vaginalis when there is no intra-abdominal testis.2 Specifically a patent processus vaginalis was
associated with an extra-abdominal testis in 91% of cases,
while a closed internal ring predicted monorchism in 97%.
This observation led to the speculation that extra-abdominal
exploration might not be needed when the ring is closed,
although there is concern that a retained testicular remnant
is a potential cancer risk. While there is no reported case of
malignancy arising from a nubbin, Rozanski et al7 reported
extensive intratubular germ cell neoplasm in a nubbin removed from a 9-year-old.7 Renzulli et al found that 7% of
extra-abdominal nubbins with a closed internal ring had
viable germ cells.8 They recommended inguinal exploration
when there is no intra-abdominal testis but a vas and vessels exiting a closed internal ring to remove extra-abdominal
testicular remnants.
While the importance of removing nubbins remains unclear, our prior experience with laparoscopy in patients with
scrotal nubbins demonstrated that the finding of a blindending vas and vessels above a closed internal ring was not
definitive as laparoscopists have claimed, since in 6 of 7
cases there was a scrotal nubbin. Furthermore, in the current series there was 1 patient with an open internal ring
and patent processus vaginalis leading to a scrotal nubbin.
Consequently neither the laparoscopic demonstration of a
blind-ending vas and vessels nor the appearance of the internal ring accurately predicted extra-abdominal findings.
Therefore, the only definitive finding on laparoscopy is an
intra-abdominal testis (or an extra-abdominal testis that
can be milked into the abdomen), which only occurs in a
third of patients with a nonpalpable testis. While beginning
an operation for unilateral nonpalpable testis with laparoscopy provides clear delineation of the intra-abdominal anatomy, it does not provide a diagnosis or assist with management in the majority of cases in which a testis is not seen.
A potential concern is that an apparent extra-abdominal
nubbin is not a testicular remnant, but rather a vaso-epididymal structure separated from a testis. Such malformations have been anecdotally reported and in clearly de-
1721
Dx confirmed by
EUA
Initial scrotal
exploration
Scrotal nubbin
Closed PV
Patent PV
Remnant
orchiectomy
Remnant
orchiectomy
And
laparoscopy
Extraabdominal testis
PV, no nubbin
Laparoscopy
Orchiopexy
Empty scrotum
Laparoscopy
Algorithm for unilateral nonpalpable testis management. DX, diagnosis. EUA, examination using anesthesia. PV, processus vaginalis.
scribed cases they were usually associated with a patent

processus vaginalis.9-12 For this reason we performed laparoscopy despite a scrotal nubbin when the processus vaginalis was patent.
We believe that initial scrotal exploration rather than
laparoscopy can be used to guide the management of unilateral nonpalpable testis, especially when the contralateral
testis is enlarged, suggesting monorchism (see figure). Scrotal incision allows the removal of nubbins as well as orchiopexy for inguinal testes. Laparoscopy is performed when
neither a testis nor a nubbin is found or when a patent
process vaginalis is noted. When an intra-abdominal testis is
identified, a scrotal incision is usually used during orchiopexy. If the descended testis is not enlarged, the surgeon has
the option of an initial scrotal incision or laparoscopy, given
the greater likelihood of a viable testis. Use of a scrotal
incision in this situation is based on the possible finding of
an extra-abdominal testis that was not palpated with the
patient under anesthesia or the occasional looping vas from
an intra-abdominal testis that can be dissected in the sac to
the internal ring to facilitate laparoscopic orchiopexy.
tients most likely to benefit from its use, including those

with no scrotal findings and those with a processus vaginalis
without a nubbin or extra-abdominal testis.
REFERENCES
1.
2.
3.
4.
5.
6.
7.
8.
CONCLUSIONS
Initial scrotal exploration is more likely than initial laparoscopy to establish the diagnosis and facilitate the management of unilateral nonpalpable testis, especially when the
length of the contralateral descended testis is 1.8 cm or
greater. Scrotal exploration detects most nubbins and extraabdominal testes, and facilitates mobilization of a looping
vas to assist with laparoscopic orchiopexy. Furthermore, the
scrotal incision is used in most cases of intra-abdominal
testis to reposition the testis. Using an initial scrotal approach laparoscopy can be directed to the minority of pa-
9.
10.
11.
12.
Merguerian PA, Mevorach RA, Shortliffe LD and Cendron M:

Laparoscopy for the evaluation and management of the
nonpalpable testicle. Urology 1998; 51: 3.
Elder JS: Laparoscopy for impalpable testes: significance of the
patent processus vaginalis. J Urol 1994; 152: 776.
Belman AB and Rushton HG: Is the vanished testis always a
scrotal event? BJU Int 2001; 87: 480.
Snodgrass W, Chen K and Harrison C: Initial scrotal incision
for unilateral nonpalpable testis. J Urol 2004; 172: 1742.
Koff SA: Does compensatory testicular enlargement predict
monorchism? J Urol 1991; 146: 632.
Huff DS, Snyder HM 3rd, Hadziselimovic F, Blyth B and
Duckett JW: An absent testis is associated with contralateral testicular hypertrophy. J Urol 1992; 148: 627.
Rozanski TA, Wojno KJ and Bloom DA: The remnant orchiectomy. J Urol 1996; 155: 712.
Renzulli JF 2nd, Shetty R, Mangray S, Anderson KR, Weiss
RM and Caldamone AA: Clinical and histological significance of the testicular remnant found on inguinal exploration after diagnostic laparoscopy in the absence of a patent
processus vaginalis. J Urol 2005; 174: 1584.
Dean AL Jr, Major JW and Ottenheimer EJ: Failure of fusion
of the testis and epididymis. J Urol 1952; 68: 754.
Lythgoe JP: Failure of fusion of the testis and epididymis. Br J
Urol 1961; 33: 80.
Wolffenbuttel KP, Kok DJ, Den Hollander JC and Nijman JM:
Vanished testis: be aware of an abdominal testis. J Urol
2000; 163: 957.
Foley PT, Sparnon AL and Lipsett J: Urogenital nonunion
the case for laparoscopy for the impalpable testis. Pediatr
Surg Int 2005; 21: 655.
Incidence of Testicular Ascent in Boys With Retractile Testes

Andrew A. Stec, John C. Thomas, Romano T. DeMarco, John C. Pope, IV, John W. Brock, III
and Mark C. Adams*
From the Division of Pediatric Urology, Vanderbilt University Medical Center, Nashville, Tennessee
Purpose: The incidence of testicular ascent of a previously descended but retractile testis is reported to be between 2% and
45%. It has been our bias that only a small percent of retractile testes ascend and ultimately require orchiopexy.
Materials and Methods: We performed a retrospective review of all retractile testes diagnosed by any of 4 pediatric
urologists at our institution from 1996 to 2005. A retractile testis was defined as one that could be brought to at least the mid
scrotum without tension despite a history of maldescent or a retractile nature. Patients were excluded if followup was less
than 6 months or they had a history of ipsilateral inguinal surgery. A followup telephone survey was done in patients not
reexamined at resolution.
Results: A total of 172 patients met our inclusion criteria and were followed to resolution. A total of 274 retractile testes were
initially identified, including 26% unilateral and 74% bilateral testes. A total of 19 testes ascended and ultimately required
orchiopexy within a followup of 6 to 101 months. A patent processus vaginalis was identified in 13 cases (68%). Another 235
patients initially seen with a diagnosis of retractile testis who had no followup examination at our institution were contacted
via telephone. This set of boys initially had a total of 392 involved testes, of which 2 underwent orchiopexy elsewhere before
resolution. The incidence of secondary ascent was 3.2% in the combined set of patients.
Conclusions: Retractile testes can ascend and, therefore, they require followup until resolution. In our experience most
retractile testes do not require surgery.
Key Words: abnormalities, testis, cryptorchidism
he incidence of congenital cryptorchidism is reported

to be around 1.6% in males at age 3 months.1 Interestingly the rate of orchiopexy remains substantially
higher in some series at 2.6% to 3.4%.2,3 Congenital cryptorchidism carries an increased risk of testicular trauma and
torsion, infertility, testicular cancer and impaired testicular
function. These risks justify careful screening as well as
prompt surgical correction of true UDT.4,5 The difference
between the rates of cryptorchidism and orchiopexy have
been attributed to initial misdiagnosis of a cryptorchid testis
as descended, surgery for a testis that is actually retractile
or a descended testis that has secondarily ascended. Secondary testicular ascent is a process by which a testis that was
previously noted to be in a normal dependent position in the
scrotum by examination has ascended into an inguinal position and can no longer be manipulated into the dependent
position. It has also been termed acquired UDT. Acquired
cryptorchidism was noted in 1975 when Myers and Officer
reported a total of 7 cases in 2 families with well documented
medical examinations.6 This observation was reconfirmed in
1985 when Atwell documented 10 cases,7 and in 1987 when
Schiffer et al reported 3 cases of testicular ascent requiring
intervention in patients with well documented testicular
examinations earlier in life.8
* Correspondence: Division of Pediatric Urology, Department of

Urology, Vanderbilt University, Monroe Carrell Jr. Vanderbilt Childrens Hospital, 4102 Doctors Office Tower, 2200 Childrens Way,
Nashville, Tennessee 37232-9820 (telephone: 615-936-1060; FAX:
615-936-1061; e-mail: mark.adams@vanderbilt.edu).
0022-5347/07/1784-1722/0
Retractile testis, defined as a gonad that can be brought

down into the scrotum without tension despite a history of
being above the scrotum most of the time or a location there
on previous examination by a primary care physician, have
been thought to be a variant of a normal gonad affected by a
hyperactive cremasteric reflex. They are generally expected
to ultimately resolve during the pubertal development of the
boy, if not before.9 However, the rate of secondary testicular
ascent of retractile testes is reported to be as high as 50%.10
It has been our bias that the incidence is much lower. Therefore, we reviewed our experience with boys initially documented to have retractile testis on examination at our institution.
We retrospectively reviewed all cases of retractile testis
diagnosed by any of the pediatric urologists at our institution from 1996 to 2005. All patients were initially referred by
the primary physician of the child for evaluation of undescended or retractile testis. A retractile testis was defined by
us as one that by patient and parental history was not down
or was undescended or retractile on examination by a primary care physician but that could be brought down at least
into the mid scrotum without tension at our examination.
Four pediatric urologists contributed patients to the series.
Almost all patients were seen by the same urologist at any
followup examination. We excluded patients with less than 6
months of followup and those who underwent prior ipsilateral inguinal surgery. A retractile testis was considered to
have ascended if it could no longer be manipulated into the
1722
Vol. 178, 1722-1725, October 2007

Printed in U.S.A.
DOI:10.1016/j.juro.2007.05.091
TESTICULAR ASCENT AND RETRACTILE TESTES

mid scrotum and remain there without tension. At any orchiopexy done at our institution all patients first underwent
examination under anesthesia to confirm that the testis was
truly undescended and not still retractile.
A subsequent simple telephone survey was performed of
patients who were not reexamined at our institution. Parents of patients were asked if the retractile testes now remained down in the scrotum, at what age the testes were
consistently down and whether their son had ever required
orchiopexy. Information was included only if the patient and
family remembered the initial problem and were aware that
the retractile nature had resolved to the satisfaction of the
patient, parent and pediatrician.
RESULTS
A total of 944 patients were referred for the evaluation of
undescended or retractile testes and diagnosed with retractile
testis by 1 of 4 pediatric urologists at our institution. A review
of the physical examinations revealed that 23 patients in this
series actually had a completely normal examination when
seen despite their history. Of the remaining 921 patients
whose testes could be brought down without tension 749 did
not return for any followup visits. A total of 172 patients were
followed, typically on an annual basis by the same pediatric
urologist, with serial examinations to reevaluate the location,
mobility and size of the gonads. Mean age of the 172 patients at
presentation to us was 50 months (range 1 to 151). A total of
274 testes were documented as retractile, 70 were unilateral
(26%) and 204 were bilateral (76%). Mean followup was 26
months (range 6 to 102).
Of the testes a total of 19 (6.9%) ultimately required
orchiopexy in 16 patients. Indication was ascent in each
case. Unilateral orchiopexy was performed in 13 boys, while
3 required bilateral orchiopexy. Mean age of the patients
who required orchiopexy was 61 months at presentation and
70.8 months at surgery. All children undergoing orchiopexy
had postoperative followup documenting that the testes
were in a satisfactory dependent position in the scrotum. Of
note, a patent processus vaginalis was documented in 13 of
the 19 cases (68%) of acquired UDT. In none of the patients
was a definite note made of any tethering fibrous band in the
location of the obliterated tunica vaginalis.
Followup telephone survey of the 749 patients without
subsequent pediatric urological examination at our institution captured 235 patients (31.3%) whose parents were
aware of the initial issue and its resolution. These 235 patients had a total of 392 total retractile testes, of which 157
were bilateral (67%) and 78 were unilateral (33%). Only 2
children, of whom each had a unilateral retractile testis,
subsequently underwent orchiopexy elsewhere. Including
those with telephone survey followup 21 of 666 retractile
testes (3.2%) underwent secondary testicular ascent and
orchiopexy.
DISCUSSION
The incidence of acquired cryptorchidism or secondary ascent
of a retractile testis is reported in the literature to occur in 2%
to 45% of cases.11,12 In 1988 Belman noted 7 retractile testes in
children who had acquired cryptorchidism and needed surgical
correction.13 Four major theories have been proposed to explain this process of secondary ascent. Ascent may be iatro-
1723
genic from the formation of scar tissue following inguinal surgery, such as hernia repair. The formation of adhesions to the
testicle or spermatic cord induces ascent of a testicle out of its
normal dependent scrotal position.14 A second theory is that
the peritoneal portion of the processus vaginalis becomes absorbed and integrated into the parietal peritoneum. This foreshortening of the process vaginalis causes traction on the descended testis and may ultimately result in its subsequent
ascent.15 A further extension of this theory by Robertson et al
suggests that a fibrous remnant of the processus vaginalis
remains tethered to the testis and peritoneum, and can pull the
testis up into the inguinal canal as the child grows.16 A third
hypothesis is that the testicle may ascend for an anatomical
reason, such as failure of natural elongation of the spermatic
cord, or an abnormal change in the size or length of the inguinal canal during growth.7 Finally, the GFN may have some
role in ascent, as it does in inguinoscrotal descent of the testis.15 Shono et al transected the GFN in neonatal rats and
observed not only testicular maldescent, but also testicular
ascent during adulthood.17 Further investigation revealed that
dividing the proximal GFN in prepubertal rats induced relative testicular ascent in adulthood by shortening the processus
vaginalis.17
No matter what the mechanism, Barthold and Gonzalez
suggested in 2003 that patients with significant testicular
retractility appeared to be at highest risk for acquired cryptorchidism.10 The theories noted could explain the true ascent
of a fully descended but retractile gonad. Another explanation
of the phenomenon should include initial misdiagnosis of a
UDT as retractile. The difference between the 2 situations
may be subtle in some cases even for experienced pediatric
urologists. Several orchiopexies performed in boys in this
series with regular followup were done after the first reexamination. It is quite possible, if not likely, that no real
ascent occurred in that time. Rather, what we initially
termed a retractile testis might truly have been a gliding,
undescended gonad that was increasingly concerning or obvious on subsequent examination. We believe that some of
our patients requiring orchiopexy during followup had misdiagnosed UDTs initially and, therefore, the incidence of
ascent in the cohort of boys with true retractile testis was
lower than 6.9%. A recently published series by Agarwal
et al of orchiopexy performed for secondary testicular ascent
of a retractile testis showed a fibrous remnant in 87% of
cases, including all in which no hernia sac was present.18
Findings of a patent processus in 68% of our cases would
support the idea that some of these testes that we initially
called retractile were gliding UDTs. Because the distinction
was not made preoperatively and the diagnosis of retractile
testes is a clinical one, we kept these patients in our cohort.
There are clear limitations of this study, most specifically
its retrospective nature. When reviewing this experience,
many patients did not return for serial examinations until
resolution. This was largely a function of our initial practice
pattern. Patients seen in the early years of the study were
assured that the problem was likely to resolve without any
need for surgery and were asked to return as needed based
on parent or pediatrician concerns. During the latter half of the
series patients were asked to return for serial examinations
annually or biannually until resolution. We recognize that
longer followup of the group may result in more orchiopexies,
although the telephone survey suggests that the additional
1724
number might be quite low. With telephone followup the rate of

secondary testicular ascent in our population undergoing orchiopexy was as low as 3.2%. In our experience the rate of
operative intervention for acquired cryptorchidism of a retractile gonad is between 3.2% and 6.9%. While the telephone
survey was not validated, it consisted of only a few simple
questions with the only end point used in our study being
whether the child ever required surgery. The application of
orchiopexy was considerably lower in the group of boys followed only by the telephone survey compared to that in boys
who we examined repeatedly. It is quite possible that we introduced considerable bias to explain the difference by encouraging and scheduling the latter group to be seen again if we
were more concerned or less sure about the initial examination. It should also be noted that our practice has been to
perform orchiopexy after a single clinical examination showed
ascent, which was confirmed by reexamination using anesthesia. A smaller number of boys may have required orchiopexy if
they had been followed nonoperatively and demonstrated
spontaneous re-descent. To date we have treated acquired
UDTs just as we would treat primary ones.
The potential consequences associated with UDT can be
quite detrimental to male children. An increased risk of
testicular cancer, torsion, trauma and infertility make diagnosis of this entity an important part of routine health
maintenance in boys. Mayr et al reported that testes that
ascend secondarily show the same histopathological changes
as testis that are congenitally undescended.19 This was reconfirmed in 2002 by Rusnack et al, who found that the
ascending testis has a decreased germ cell count similar to
that of the congenital UDT.20 Therefore, all children with
undescended or retractile testes should be referred to a
specialist early in their care.
In our experience 3.2% to 6.9% of retractile testes ultimately required orchiopexy. This range is considerably
lower than that in some other series. It is unlikely that the
natural history of retractile gonads is different at our institution than elsewhere. The variance noted was likely due to
subtle differences in which patients were included in initial
cohorts of those with retractile or undescended testes as well
as different definitions of what was considered ascent and an
indication for orchiopexy. Most cases of testicular ascent
occur within the first 5 years of childhood but they can occur
as late as age 12 years. This pattern indicates that patients
diagnosed with a retractile testis should undergo routine
reevaluation until the problem has resolved.
REFERENCES
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
15.
16.
17.
18.
19.
20.
CONCLUSIONS
Secondary testicular ascent is an uncommon but real occurrence in male children. Boys with retractile testis can ultimately have acquired UDT and require surgical correction.
However, the majority of retractile testes resolve without
surgical intervention.

GFN genitofemoral nerve
UDT undescended testicle
Cryptorchidism: a prospective study of 7500 consecutive male

births, 1984 8. John Radcliffe Hospital Cryptorchidism
Study Group. Arch Dis Child 1992; 67: 892.
Chilvers C, Pike MC, Forman D, Fogelman K and Wadsworth
ME: Apparent doubling of frequency of undescended testis
in England and Wales in 1962 81. Lancet 1984; 2: 330.
Swerdlow AJ, Wood KH and Smith PG: A case-control study of
the aetiology of cryptorchidism. J Epidemiol Commun Health
1983; 37: 238.
Chilvers C, Dudley NE, Gough MH, Jackson MB and Pike MC:
Undescended testis: the effect of treatment on subsequent
risk of subfertility and malignancy. J Pediatr Surg 1986;
21: 691.
Hand JR: Undescended testes: report of 153 cases with evaluation of clinical findings, treatment, and results on followup
up to thirty-three years. Trans Am Assoc Genitourin Surg
1955; 47: 9.
Myers NA and Officer CB: Undescended testis: congenital or
acquired? Aust Paediatr J 1975; 11: 76.
Atwell JD: Ascent of the testis: fact or fiction. Br J Urol 1985;
57: 474.
Schiffer KA, Kogan SJ, Reda EF and Levitt SB: Acquired
undescended testes. Am J Dis Child 1987; 141: 106.
Wyllie GG: The retractile testis. Med J Aust 1984; 140: 403.
Barthold JS and Gonzalez R: The epidemiology of congenital
cryptorchidism, testicular ascent and orchiopexy. J Urol
2003; 170: 2396.
Villumsen AL and Zachau-Christiansen B: Spontaneous alterations in position of the testes. Arch Dis Child 1966; 41:
198.
Lamah M, McCaughey ES, Finlay FO and Burge DM: The
ascending testis: is late orchidopexy due to failure of screening or late ascent? Pediatr Surg Int 2001; 17: 421.
Belman AB: Acquired undescended (ascended) testis: effects of
human chorionic gonadotropin. J Urol 1988; 140: 1189.
Kaplan GW: Iatrogenic cryptorchidism resulting from hernia
repair. Surg Gynecol Obstet 1976; 142: 671.
Clarnette TD and Hutson JM: Is the ascending testis actually
stationary? Normal elongation of the spermatic cord is
prevented by a fibrous remnant of the processus vaginalis.
Pediatr Surg Int 1997; 12: 155.
Robertson JF, Azmy AF and Cochran W: Assent to ascent of
the testis. Br J Urol 1988; 61: 146.
Shono T, Zakaria O, Imajima T, Suita S and Morimoto K:
Pubertal genitofemoral nerve division induces testicular
ascent in adult rats. J Pediatr Surg 1999; 34: 346.
Agarwal PK, Diaz M and Elder JS: Retractile testis: is it really
a normal variant? J Urol 2006; 175: 1496.
Mayr J, Rune GM, Holas A, Schimpl G, Schmidt B and
Haberlik A: Ascent of the testis in children. Eur J Pediatr
1995; 154: 893.
Rusnack SL, Wu HY, Huff DS, Snyder HM 3rd, Zderic SA,
Carr MC et al: The ascending testis and the testis undescended since birth share the same histopathology. J Urol
2002; 168: 2590.
EDITORIAL COMMENT
Why do only a small number of retractile testes ascend? Are
these testes really retractile? I think not. I personally believe that these testes are congenitally undescended but
they are initially low or what I like to call retroscrotal, that
is so low that they seem to be in the scrotum during infancy.
In many cases a silk glove sign can be appreciated when the
testis is examined through the scrotum. This means to me

that the testis is 1 layer from being in the scrotum. This
sensation, similar to the tethering noted by these authors,
may be a warning that ascent will occur, in many cases
within 6 to 12 months.
Until we can more accurately identify the cause of delayed ascent we must be vigilant in followup and recommend
that pediatric urological evaluation must be done in all cases
1725
in which testicular position is questioned by primary care

physicians.
Mark F. Bellinger
Childrens Hospital of Pittsburgh
Pittsburgh, Pennsylvania
Further Evidence for Spontaneous

Descent of Acquired Undescended Testes
S. W. Eijsbouts,* S. M. P. F. de Muinck Keizer-Schrama and F. W. J. Hazebroek
From the Departments of Pediatric Surgery and Pediatrics and Endocrinology (SMPFdMKS), Erasmus Medical Center-Sophia Childrens
Hospital, Rotterdam, The Netherlands
Purpose: It is still controversial whether acquired undescended testis can best be managed by orchiopexy or by the wait and
see method. We prospectively evaluated spontaneous descent of acquired undescended testes and possible predictive factors
in prepubertal boys.
Materials and Methods: From 1982 to 2004 spontaneous descent was awaited until at least Tanner stage P2G2 in 109 boys
with a total of 83 unilateral and 52 bilateral acquired undescended testes. Annually we established testis position and size.
After Tanner stage P2G2 orchiopexy was done for all testes in an unstable scrotal position.
Results: Two boys (3 acquired undescended testes) were excluded from analysis. Of 132 acquired undescended testes 75
descended spontaneously (57%, 95% CI 48 65), including 40 of 75 (57%) in early puberty or before puberty and 32 of 75 (43%) in
mid puberty. Orchiopexy was performed in 57 of 132 acquired undescended testes (43%). Acquired undescended testes showed an
increasing chance of descending spontaneously with increasing age (p trend 0.002). In 63 of 82 unilateral undescended testes we
were able to compare testis volume at the onset of puberty with that of the healthy contralateral side. Of 17 testes that needed
orchiopexy 12 (71%) had a volume that was more than 1 ml smaller than the healthy testis. This was noted in only 18 of 46
spontaneously descended acquired undescended testes (39%, p 0.053). Other factors, such as the most caudal testicular position
at referral or the frequency of confirmed descended testicular position before referral, were not predictive of spontaneous descent.
Conclusions: A conservative wait and see approach to acquired undescended testis until puberty could prevent more than
half of the boys from undergoing orchiopexy and it does not seem detrimental in terms of testicular volume.
Key Words: testis, abnormalities, cryptorchidism, puberty
ryptorchidism or UDT is the most common genital

anomaly seen in boys and it can be congenital or
acquired. The incidence of congenital UDT is approximately 3% but about two-thirds of congenital UDTs
descend spontaneously in the first 6 months of life.1 Early
treatment intended to preserve fertility potential is recommended, so that postnatal germ cell development can
proceed normally.2
However, despite this recommendation many boys
present with UDT at an older age. This can be accounted for
by acquired UDTs, ie testes ascending to a nonscrotal position after previous normal descent. The etiology, natural
course and treatment of such acquired UDTs are debated for
several reasons. 1) The history of the testis position must be
assessed meticulously to distinguish between congenital and
acquired forms of UDT. 2) It can be difficult to differentiate
acquired UDTs from retractile testes. 3) Acquired UDTs
frequently descend spontaneously in the peripubertal period. 4) Different diagnostic criteria make it hard to compare
data from different studies, which are already sparse. To
investigate the rate of spontaneous descent and possible
prognostic factors in this prospective study we followed boys
with acquired UDT from referral until puberty.
* Correspondence: Department of Pediatric Surgery, Erasmus

Medica Center-Sophia Childrens Hospital, SP-2480, P. O. Box
2060, 3000 CB Rotterdam, The Netherlands (telephone: 31 62 380
9987; FAX 31 10 463 6802; email: s.eijsbouts@erasmusmc.nl).
0022-5347/07/1784-1726/0

From 1982 to 2004 boys with acquired UDT who were referred to our tertiary hospital were offered conservative
treatment and followed yearly. Those who had undergone
operation were reexamined within 4 weeks after surgery.
Families that did not attend for followup were re-invited by
mail or telephone. Those that could not be traced were
considered lost to followup.
For a boy with unilateral or bilateral UDT to be included
in this prospective study certain criteria had to be met. At
least 1 medical documentation of earlier normal descent of
the testis had to be confirmed by the patient Regional Youth
Health Care Institution, general practitioner or hospital file.
Furthermore, the testis had to be palpable, without ipsilateral, inguinal, clinically present pathological findings or previous ipsilateral inguinal surgery and the most caudal position after manipulation had to be nonscrotal or scrotally
unstable. An unstable testis was defined as a testis that
immediately acquired a nonscrotal position after release. We
excluded boys with retractile testes, ie testes that could be
manipulated at examination into a stable scrotal position
until the cremasteric reflex was induced.
The annual genital examination in the study was performed by the same senior authors (SdeMK-S and/or FH)
with the boy supine in the cross-legged position. A 2-handed
technique was used to assess the most caudal testis position.
These positions were categorized as inguinal, scrotal entrance, unstable high scrotal, unstable low scrotal and stable
1726
Vol. 178, 1726-1729, October 2007

Printed in U.S.A.
DOI:10.1016/j.juro.2007.03.184
SPONTANEOUS DESCENT OF ACQUIRED UNDESCENDED TESTES

scrotal. Testis volume was measured using a Prader orchidometer3 and Tanner puberty stages were determined.4
Spontaneous redescent of the testis was awaited until
at least Tanner puberty stage P2G2. Puberty stages were
categorized into early, mid and late puberty. Early puberty included stage G2testicular volume 4 to 8 ml, mid
puberty G39 to 11 ml and G4 12 to 15 ml, and late
puberty G5more than 15 ml. The largest testis was chosen
for categorization into puberty stages. Orchiopexy was performed on all testes that had not descended to a stable scrotal
position at puberty stage P2G2. The criteria for earlier intervention were a request for surgery by the parents, inguinal or
testicular complaints and testes that became unpalpable or
were completely immobile in the inguinal area.
The rate of spontaneous descent or nondescent was compared with testis position after manipulation at the start of
the study, and the testis volume at the beginning and end of
the study period. The duration of these testes being in an
ascended position was prolonged by awaiting spontaneous
descent. The influence of this delay on testis volume was
evaluated by comparing the volume of the acquired UDT at
puberty onset with testis volume on the healthy contralateral side. Therefore, testis volume differences were only
calculated in boys with unilateral acquired UDT and a normally descended contralateral testis. Boys in whom inguinal
surgery was performed on either side before puberty were
excluded from this analysis. We also measured other parameters that could be associated with spontaneous descent,
including patient age at referral for testicular ascent, the
frequency of previous scrotal positions noted before referral
and physician who made these observations, ie those at
Regional Youth Health Care Institutions, general practitioners, pediatricians or pediatric surgeons.
The t test was used to compare mean testis volume differences for spontaneously descended and operated testes.5
We also estimated the frequency of these testis volume differences being more than 1 ml.6 The chi-square test was
used to calculate statistical significances for bivariate tabular analysis. Results were analyzed using SPSS, release
13.0 and significance was considered at 0.05.
RESULTS
We followed 109 boys (135 acquired UDTs, including 42 on
the left, 41 on the right and 52 bilateral) until at least
puberty stage P2G2. Mean patient age at referral was 8.9
2.9 years (range 1 to 14.9). Mean followup was 4.5 years
(range 0.3 to 12.1). Two boys were excluded from analysis,
including 1 with unilateral UDT who was lost to followup,
while the parents of a boy with bilateral UDTs requested
orchiopexy before puberty. Of the remaining 132 testes (82
unilateral and 50 bilateral) 53 were documented at least
1727
TABLE 2. Puberty stage when testis descended spontaneously or

orchiopexy was done
Stage
No. pts
Before (G1)
Early (G2)
Mid:
G3
G4
Late (G5)
Testis Vol
(ml)
No.
Descent
No.
Orchiopexy
Less than 4
48
75
3
40
57
10
19
911
1215
Greater than 15
28
4
0
17
8
3
once as scrotal, 39 were documented twice and 40 were

documented 3 times or more. Table 1 lists testis positions
after manipulation at referral for acquired UDT.
Of the 132 testes 75 (57%) descended spontaneously and,
thus, the other 57 needed orchiopexy. Orchiopexy was performed after Tanner stage P2G2 in 47 testes and in 10 (8
unilateral and 1 bilateral) before puberty. Reasons for earlier operation were inguinal complaints in 3 boys, a clinically
present ipsilateral inguinal hernia in 3 and completely immobile testis in the inguinal area in 2, while 1 with bilateral
acquired UDT had testicular torsion and contralateral anchoring orchiopexy.
In 43 of 75 testes (57%) descent occurred in early puberty
or before puberty, in 32 of 75 (43%) it occurred in mid
puberty and it occurred in none in late puberty (table 2). In
28 of 57 UDTs (49%) orchiopexy was performed at mid or
late puberty. Pubertal development was within the normal
range.7 Testes divided into tertile groups according to patient age at referral (1 to 8.5, 8.6 to 10.6 and 10.7 to 14.9
years) showed an increasing chance of descending spontaneously with increasing age (p trend 0.002, see figure).
Of the 82 unilateral acquired UDTs 63 had a healthy
contralateral testis and for these 63 volume differences were
calculated. Excluded from this analysis were 19 boys (19
unilateral UDTs), including the mentioned 8 who underwent
orchiopexy before puberty and 11 with previous contralateral inguinal surgery for congenital UDT in 9 and inguinal
hernia in 2. Of 17 testes 12 (71%) that needed orchiopexy
had a volume more than 1 ml smaller than the healthy
testis. This finding was noted in only 18 of 46 spontaneously
descended acquired UDTs (39%) (chi-square p 0.053).
TABLE 1. Caudal position of acquired UDT at referral

after manipulation
Testis
No. Pts (%)
Inguinal
Scrotal entrance
High scrotal, unstable
Low scrotal, unstable
3 (2)
44 (33)
56 (42)
29 (22)
Total No.
132
Tertile groups of age in years at referral for acquired UDT by

outcome after wait and see approach until puberty onset.
1728
Mean SEM testicular volume differences were significantly smaller for spontaneously descended testes than for
operated testes (1.5 0.37 and 3.9 0.97 ml, respectively, t test p 0.030).
Other parameters evaluated were not associated with
spontaneous descent or nondescent (data not shown), including the frequency of previous scrotal positions noted before
referral (chi-square p 0.20), cases in which a physician
made these observations at least once (chi-square p 0.34)
and the caudality of testis position at referral (chi-square
p 0.85).
DISCUSSION
The necessity for operative treatment of acquired UDT was
questioned and, therefore, spontaneous descent was awaited
until the onset of puberty, at least until Tanner stage P2G2.
Of acquired UDTs 75 of 132 (57%) descended spontaneously
in this study, mostly in early puberty (40 of 75 or 53%) and
mid puberty (32 of 75 of 43%). In a study by Sijstermans et al
98 of 129 acquired UDTs (76%) descended spontaneously, of
which 70 (71%) descended in early puberty (G2) and 26 of 98
(27%) descended in mid puberty.8,9 They operated on boys
from Tanner stage G3 and thereafter, while in our study
surgery was done from G2. The study designs were otherwise largely comparable. The significant difference in the
rates of spontaneous descent cannot be explained by the
data provided (chi-square p 0.002). These studies are
the only 2 long-term followup studies in boys with acquired
UDTs in whom spontaneous descent was awaited.
Analysis of age at referral for acquired UDT in our study
showed that the chance of spontaneous descent increased
with age at referral. This was recently also shown to be true
for retractile testes10 but these testes were excluded in our
study. Other parameters available at the diagnosis of testicular ascent were not predictive for spontaneous descent.
Rusnack et al postulated that the germ cell count is the
same in ascending and primary UDTs.11 A higher testis
volume at puberty positively correlates with fertility potential.12 We compared the volume of unilateral acquired UDTs
after spontaneous descent or before orchiopexy with the
volume on the contralateral healthy side. Testes that needed
orchiopexy were smaller than testes on the contralateral
side compared with testes that descended spontaneously.
Other studies showed that unilateral congenital UDTs corrected at a young age are also frequently smaller than the
testis on the contralateral side at puberty onset or adolescence.13,14 Thus, it might be that the etiology of decreased
pubertal size of acquired UDT is intrinsic. Consequently
delayed orchiopexy does not seem to be responsible for the
smaller size of testes that need orchiopexy.
Previous studies from as early as the 1930s recognized
the high frequency of spontaneous peripubertal descent of
UDTs.8 The disproportionately high orchiopexy rate in The
Netherlands,15 greater than the much lower incidence of
congenital UDT, can be explained by acquired UDT. This
relatively new phenomenon has gained more recognition in
the literature.11,16 18 Others believe that acquired UDT is
not a separate entity and claim that difficulties in examining
children and the subtleties of anatomy are responsible for
late diagnosis of UDT and subsequent late orchiopexy.19 We
agree that the sensitivity and reproducibility of determining
testicular position may be low. Repeat examination and
careful history taking are crucial for influencing this positively. However, recognition of the concept of acquired UDT
accentuates that the testis position can change with time.
Normally descended testes can become retractile or attain
an undescended position. A theory of testicular ascent is
deficient elongation of the spermatic cord.20 Other studies
indicate that a fibrous remnant of the processus vaginalis
that has not completely vanished would inhibit elongation of
the vas deferens and spermatic vessels with age.16 Pubertal
surges in gonadotropins and gonadal hormones are hypothesized to be responsible for pubertal spontaneous descent.8
As observed in our study, spontaneous prepubertal descent
might well be caused by surges of adrenal androgens.
The natural course of congenital UDT differs from that of
acquired UDT. Therefore, the management of acquired
UDTs (testes that ascended from a descended scrotal position during childhood) cannot be thoughtlessly copied from
the approach to congenital UDTs. More than half of acquired
UDTs have the tendency to descend spontaneously in early
and mid puberty. Unfortunately at the start of followup we
were not able to identify acquired UDTs that did not descent
spontaneously. More long-term followup studies are necessary to determine the fertility potential of boys with a history of acquired UDT who awaited spontaneous descent
until the pubertal age.
CONCLUSIONS
A conservative wait and see approach to acquired UDT until
puberty could prevent more than half of the boys from undergoing orchiopexy and it does not seem detrimental in
terms of testicular volume. An acquired UDT that ascends at
an older age seems to have a higher chance of descending
spontaneously.
ACKNOWLEDGMENTS
Vincent Cappendijk collected data and Ko Hagoort provided
editorial assistance.

UDT undescended testis
REFERENCES
1. Scorer CG: The descent of the testis. Arch Dis Child 1964; 39: 605.
2. Hutson JM and Hasthorpe S: Abnormalities of testicular descent. Cell Tissue Res 2005; 322: 155.
3. Prader A: Testicular size: assessment and clinical importance.
Triangle 1966; 7: 240.
4. Marshall WA and Tanner JM: Variations in the pattern of
pubertal changes in boys. Arch Dis Child 1970; 45: 13.
5. Pinto KJ, Kroovand RL and Jarow JP: Varicocele related testicular atrophy and its predictive effect upon fertility.
J Urol 1994; 152: 788.
6. Costabile RA, Skoog S and Radowich M: Testicular volume assessment in the adolescent with a varicocele. J Urol 1992; 147:
1348.
7. Mul D, Fredriks AM, van BS, Oostdijk W, Verloove-Vanhorick
SP and Wit JM: Pubertal development in The Netherlands
1965-1997. Pediatr Res 2001; 50: 479.

8.
9.
10.
11.
12.
13.
Sijstermans K, Hack WW, van der Voort-Doedens LM, Meijer

RW and Haasnoot K: Puberty stage and spontaneous descent of acquired undescended testis: implications for therapy? Int J Androl 2006; 29: 597.
Hack WW, Meijer RW, van der Voort-Doedens LM, Bos SD and
Haasnoot K: Natural course of acquired undescended testis
in boys. Br J Surg 2003; 90: 728.
Agarwal PK, Diaz M and Elder JS: Retractile testisis it really
a normal variant? J Urol 2006; 175: 1496.
Rusnack SL, Wu HY, Huff DS, Snyder HM III, Zderic SA, Carr MC
et al: The ascending testis and the testis undescended since birth
share the same histopathology J Urol 2002; 168: 2590.
Takihara H, Cosentino MJ, Sakatoku J and Cockett AT: Significance of testicular size measurement in andrology: II.
Correlation of testicular size with testicular function.
J Urol 1987; 137: 416.
Ku JH, Kim ME, Lee NK and Park YH: Testicular volume and
masculine identity in men with unilateral cryptorchidism:
results of a community-based survey in Korea. Urol Res
2003; 31: 312.
14.
15.
16.
17.
18.
19.
20.
1729
Lee PA, Coughlin MT and Bellinger MF: No relationship of testicular size at orchiopexy with fertility in men who previously
had unilateral cryptorchidism. J Urol 2001; 166: 236.
Snick HK: Excessive incidence of orchidopexy; study and measures on the island of Walcheren. Ned Tijdschr Geneeskd
1984; 128: 2077.
Clarnette TD, Rowe D, Hasthorpe S and Hutson JM: Incomplete disappearance of the processus vaginalis as a cause of
ascending testis. J Urol 1997; 157: 1889.
Hack WW, Meijer RW, Bos SD and Haasnoot K: A new clinical
classification for undescended testis. Scand J Urol Nephrol
2003; 37: 43.
Robertson JF, Azmy AF and Cochran W: Assent to ascent of
the testis. Br J Urol 1988; 61: 146.
Redman JF: The ascending (acquired undescended) testis: a
phenomenon? BJU Int 2005; 95: 1165.
Donaldson KM, Tong SY and Hutson JM: Prevalence of late
orchidopexy is consistent with some undescended testes
being acquired. Indian J Pediatr 1996; 63: 725.
Ultrasonography Underestimates the Volume

of Normal Parenchyma in Benign Testicular Masses
Ashay S. Patel, Brian D. Coley* and Venkata R. Jayanthi
From the Sections of Pediatric Urology and Radiology (BDC), Columbus Childrens Hospital, The Ohio State University, Columbus, Ohio
Purpose: In children with testicular masses characteristic ultrasonographic features may occasionally allow the preoperative determination of a benign process and permit a testis sparing procedure to be performed. Often the amount of normal
parenchyma surrounding the mass may be seemingly negligible, which may prompt some investigators to wonder what a
testis sparing procedure might realistically spare.
Materials and Methods: We retrospectively reviewed 7 consecutive patients with intratesticular lesions who underwent
surgical exploration. Several of these children seemingly had complete replacement of the testis with the pathological process
with little to no discernible normal parenchyma. The surgical technique involved delivery of the testis and control of the
vessels. The capsule was incised and the normal parenchyma was bluntly separated to the level of the tumor. The mass was
then mobilized circumferentially and enucleated intact. The residual parenchyma was reapproximated, the capsule was
closed and the testicle was replaced in the scrotum. All patients were followed postoperatively by physical examination and
ultrasonography.
Results: The series consists of 7 children 10 to 188 months old (mean age 68). The final diagnosis was cystic teratoma in 4
cases, epidermoid tumor in 2 and a simple cyst in 1. Postoperative physical examinations revealed symmetrical testes.
Postoperative ultrasonography demonstrated normal echo texture and testicular volumes. Furthermore, no tumor recurrence
was detected.
Conclusions: Classic ultrasonographic patterns of testicular masses may permit preoperative identification of a benign
process, permitting a testis sparing procedure to be planned. However, ultrasonography underestimates the amount of
normal residual parenchyma because this tissue is compressed against the capsule into a thin rim. The amount of normal
residual parenchyma identified on preoperative ultrasound should not be used as a factor when deciding whether a testis
sparing procedure might be appropriate.
Key Words: testis, ultrasonography, teratoma, testicular neoplasms
esticular tumors in the pediatric population account

for 1% of all pediatric solid tumors.1 Children presenting with irregular scrotal contents, or a firm or hard
testicle warrant assessment for testicular tumor. There has
been recent debate regarding the most common prepubertal
testicular tumor. Ross et al reviewed the Prepubertal Testis
Tumor Registry and noted that yolk sac tumor was the most
common prepubertal testicular tumor (62%), followed by
teratoma (23%).2 However, Pohl et al surveyed 4 major
pediatric institutions and concluded that benign tumors
comprised 74% of cases with teratoma accounting for 48%
and yolk sac tumor accounting for 15%.3
Scrotal ultrasound provides detailed images of testicular
pathology. While sonography is sensitive for testicular tumors, it is generally not histologically specific. However,
some lesions (epidermoid cysts, teratomas and simple cysts)
have sufficiently characteristic imaging features to suggest
a specific diagnosis. Historically management for pediatric
testicular tumors was based on testicular tumor management in the adult population. Traditional surgical methods
* Financial interest and/or other relationship with Siemens Medical, Philips Medical and Zonare Ultrasound.
Correspondence: Section of Urology, Columbus Childrens Hospital, 700 Childrens Dr., Columbus, Ohio 43205 (e-mail:
jayanthr@chi.osu.edu).
0022-5347/07/1784-1730/0
involve radical inguinal orchiectomy for presumed malignancy. Management for prepubertal testicular tumors has
shifted in the last several years. The potential for testis
sparing surgery in those tumors with benign imaging and
clinical features has begun to be more accepted with growing
clinical experience with pediatric tumors. Recent reports
described the testis sparing approach as a viable management option for prepubertal benign testicular masses, including teratoma and epidermoid cysts.
A large testicular mass may appear to completely replace
the testicle, masking the amount of normal testicular parenchyma and, thereby, bringing into question the usefulness of
a testis sparing procedure. We report our experience with
testis sparing surgery for benign masses. We observed normal postoperative testicular volume despite preoperative
testicular ultrasound indicating little residual normal testicular parenchyma.
METHODS
We retrospectively reviewed the consecutive charts of males
younger than 18 years from 2000 to 2006 for testicular
masses for which testis sparing surgery was done at our
institution by a single surgeon (VRJ). All patients underwent evaluation, including scrotal ultrasound, physical ex-
1730
Vol. 178, 1730-1732, October 2007

Printed in U.S.A.
DOI:10.1016/j.juro.2007.05.096
ULTRASONOGRAPHY UNDERESTIMATES PARENCHYMA IN TESTICULAR MASSES

amination and measurement of the tumor markers AFP,
-human chorionic gonadotropin and lactate dehydrogenase.
Patients with sonographically benign appearing testicular masses and normal tumor markers were subsequently
treated with testis sparing surgery. The surgical technique
involved a standard inguinal incision with early control of
the spermatic cord using a tourniquet to assist with hemostasis, followed by delivery of the testicle. The tunica was
incised over the mass, followed by mobilization of the mass.
The mass was enucleated without violating the pseudocapsule. The tunica was then reapproximated and the tourniquet was released to confirm hemostasis, followed by return
of the testicle to the scrotum. Frozen sections were not
performed in any patient.
All patients were seen for followup for physical examination
and repeat ultrasound. Serial ultrasounds were done subsequently to assess tumor recurrence, testicular growth and
sonographic testicular appearance. Because sonographic measurement was shown to be more sensitive than an orchidometer,4 testicular mass and overall testicular size were calculated preoperatively and postoperatively, and recorded using
the formula for a prolate ellipsoid, L H W 0.52, where L
represents length and W indicates width.5
RESULTS
Seven consecutive patients with a benign cystic mass underwent testis sparing surgery (see table). Median patient age
was 20 months (range 10 to 188), including 1 postpubertal
male who underwent removal of an epidermoid cyst because
of increasing size. Mean followup was 14 months (range 3 to
56, median 7). All masses were located on the left side except
1. All patients had normal serum AFP and -human chorionic gonadotropin except a 10-month-old boy, in whom AFP
was increased to 43 ng/ml (normal less than 6).2
The final histological diagnosis was cystic teratoma in 4
cases, an epidermoid cyst in 2 and a simple cyst in 1. Teratomas were sonographically characterized as well circumscribed masses with heterogeneous solid and cystic areas,
and occasionally calcification. Epidermoid cysts appeared as
well circumscribed masses with typical lamellated internal
echoes. The 1 simple cyst appeared as an anechoic, fluid
filled mass without demonstrable solid tissue components.
All sonographically suspected benign masses were confirmed to be benign postoperatively.
Followup ultrasound was available for all 7 patients with
no residual or recurrent masses observed. Preoperative
affected testicular median volume was 3.55 cm3 (range 2.53
to 13.69) with a median volume of 1.18 cm3 (range 0.01 to
2.98). The median difference in preoperative volume between the affected testicle and the mass was 1.52 cm3 (range
1731
FIG. 1. Testicular size comparison postoperatively. Numbers above

bar graphs indicate individual percent volume difference between
affected and normal testes in each patient.
0.57 to 13.68). Median volume of the affected testicle postoperatively was 0.83 cm3 (range 0.19 to 15.28). Median volume of the contralateral testicle postoperatively was 0.86
cm3 (range 0.57 to 16.73). The median difference in volume
between the affected testicle and the normal unaffected testicle was 0 cm3 (range 0 to 1.45) (fig. 1). The median postoperative percent volume difference between the affected
testicle and the normal testicle was 0.26% (range 54% to
71.43%) using the formula, (1 affected/normal) 100.
Figures 2 and 3 show representative preoperative and postoperative images of a patient with cystic teratoma.
DISCUSSION
Although ultrasound is highly sensitive for detecting intratesticular masses, histological specificity is generally poor.
Exceptions to this include epidermoid cysts, simple testicular cysts and teratomas, which in prepubertal males are
always benign. Because of this, in the last decade surgeons
have become increasingly inclined to perform testis sparing
tumor removal when possible. The procedure should still be
approached as an oncological operation and careful assessment of the mass, including frozen section if necessary,
should be performed during surgery with orchiectomy done
if sufficient doubt arises about the benign nature of the
tumor. Although we controlled the vessels, strictly speaking
Patient characteristics
Pt No.
1
2
3
4
5
6
7
Median
Age (mos)
Followup (mos)
Side
Pathological Results
20
18
10
36
135
12
188
20
56
7
16
15
5
5
3
7
Rt
Lt
Lt
Lt
Lt
Lt
Lt
Cystic teratoma
Cystic teratoma
Cystic teratoma
Cystic teratoma
Epidermoid cyst
Simple cyst
Epidermoid cyst
FIG. 2. Longitudinal sonogram reveals left testicular teratoma.

A, complex intratesticular solid and cystic mass with thin rim of
testicular parenchyma (arrowheads) surrounding tumor. B, 6
months after testis sparing tumor resection note normal-appearing
testis that was equal in size to contralateral testis. Arrow indicates
Doppler signal suggesting normal intratesticular blood flow
postoperatively.
1732
ULTRASONOGRAPHY UNDERESTIMATES PARENCHYMA IN TESTICULAR MASSES
FIG. 3. Ultrasound shows testicular teratoma. A, complex mass

with cystic and solid elements. Arrowheads indicate thin rim of
normal parenchyma. B, postoperative image demonstrates normal
sonographic appearance.
this may not be necessary for benign lesions. However, it

helps with hemostasis during tumor removal and the tourniquet is typically released after the testicular capsule has
been closed. In our series frozen section was not performed
because preoperative laboratory and ultrasonography findings were so characteristic of benign lesions.
Rushton et al reported a series of 5 prepubertal patients
who underwent testis sparing surgery for teratoma with a
mean followup of 96 months.6 In their series no patient was
reported to have any recurrent or metastatic disease. In
regard to postoperative testicular appearance 2 patients had
a normal echogenic appearance and all except 1 testicle was
equal in size and consistency. Ross et al described 4 patients
with epidermoid cysts at a mean followup of 15.5 months
after testis sparing surgery with no evidence of recurrence or
atrophy but only 2 underwent serial sonography.7
Despite ultrasound findings of benign testicular masses
the ability to predict how much normal testicular volume is
present at preoperative ultrasound can be difficult to assess
accurately. This uncertainty may persuade one to consider
radical orchiectomy over more challenging testis sparing
surgery.
In our series 7 patients underwent testis sparing surgery
for sonographically benign masses. Preoperatively all patients in our study had testicular enlargement compared to
the normal testicle with a median difference of 2.83 cm3
(range 1.15 to 4.13). However, comparison at postoperative
followup between the operated testicle and the contralateral
testicle demonstrated a median difference of 0 cm3 (range 0
to 1.45). Additionally, the median postoperative percent volume difference was 0.26% (range 54% to 71.43%). The lack
of difference in testicular size postoperatively suggests that
the amount of normal parenchyma preoperatively was similar or there is some capacity for testicular parenchymal
growth or catch-up growth, as seen after varicocelectomy for
testicular atrophy. In patient 3 we observed a decrease in
postoperative testicular size and volume compared to the
contralateral testicle. It is possible this patient underwent
wider excision or experienced atrophy secondary to surgical
manipulation, although no surgical complications were
noted. Patient 5 showed increased volume of the affected
testicle compared to the contralateral testicle, although no
recurrent mass was present. These findings are similar to
those in the series by Shukla et al, in which 7 of 9 patients
had some testicular atrophy and 2 of 9 sustained enlargement of the affected testicle.8 In all of their patients ultrasound confirmed no evidence of recurrence.
All patients had normal preoperative tumor markers except patient 3, in whom AFP was increased to 43 ng/ml. This
patient was 10 months old at surgery and final pathological
study revealed cystic teratoma. In a patient with increased
AFP there is concern regarding performing a testis sparing
procedure. Ross et al recently reported that despite AFP
levels that were above normal no patient older that 6
months with teratoma on testicular resection had AFP more
than 100 ng/ml.2
CONCLUSIONS
Prepubertal, preoperatively benign testis tumors are increasingly being successfully managed by testis sparing surgery. Although preoperative ultrasound accurately shows
the character of these masses, the amount of normal parenchyma present can be underappreciated with large masses
that compress and thin the testicular parenchyma. Our experience demonstrates that the testicular parenchyma is
typically preserved and usually equal to contralateral testicular volume on followup physical examination and ultrasound. Because of this, surgeons should not hesitate to
perform testis sparing procedures even with large, benignappearing masses and seemingly minimal residual normal
parenchyma.

AFP -fetoprotein
REFERENCES
1.
2.
3.
4.
5.
6.
7.
8.
Kay R: Prepubertal Testicular Tumor Registry. Urol Clin North

Am 1993; 20: 1.
Ross JH, Rybicki L and Kay R: Clinical behavior and contemporary management algorithm for prepubertal testis tumor:
a summary of the Prepubertal Testis Tumor Registry. J Urol
2002; 168: 1675.
Pohl H, Shukla A, Metcalf PD, Cilento BG, Retik AB, Bagli DJ
et al: Prepubertal testis tumor: actual prevalence rate of
histological types. J Urol 2004; 172: 2370.
Paltiel HJ, Diamond DA, DiCanzio J, Zurakowski D, Borer JG
and Atala A: Testicular volume: comparison of orchidometer
and US measurements in dogs. Radiology 2002; 222: 114.
Taskinen S, Taavitsainen M and Wikstrom S: Measurement of
testicular volume: comparison of 3 different methods. J Urol
1996; 155: 930.
Rushton HG, Belman AB, Sesterhenn I, Patterson K and
Mostofi FK: Testicular sparing surgery for prepubertal teratoma of the testis: a clinical and pathologic study. J Urol
1990; 14: 726.
Ross JH, Kay R and Elder J: Testis sparing surgery for pediatric
epidermoid cysts of testis. J Urol 1993; 149: 353.
Shukla AR, Woodard C, Carr MC, Huff DS, Canning DA, Zderic
SA et al: Experience with testis sparing surgery for testicular teratoma. J Urol 2004; 171: 161.
Guest Lecture
Doping in Sports: The Pediatric Perspective
James M. Betts*
From the Childrens Hospital of Oakland, Oakland, California
he boys of fall are out there, and it is a real push for

the playoffs and the World Series. Football is in its
fourth or fifth week. There is so much happening but
that is not the only issue. When you stop and think about
performance-enhancing drugs, you will see that this is a
national and international issue. It is not an issue for just
the elite athletes, but it is affecting high school and elementary school athletes as well.
HISTORY OF DOPING
My interest in performance-enhancing drugs began when I
was a freshman at the University if Vermont. I was a discus
thrower on the track team and my coach said, Id like you to
be a hammer thrower as well. I did not know what the
hammer involved. It is a 16-pound weight, basically a shot
put that is on the end of a cable, which you swing around and
toss through the air. I worked on my technique and went to
my first meet in the Boston area. One competitor, who happened to be the NCAA Division 1 champ, came out and
threw. I looked at him in awe. He was about 6 foot 5 inches,
weighed about 270 pounds and could bench press a bulldozer. I was a little over 6 feet tall, weighed 190 pounds and
thought to myself, this is not going to work. I went back to
the trainer in Vermont and said, I need some performance
enhancement. He sent me to the dairy bar and said, Morning, evening and night youll need to get a milkshake or
sundae. After a month of my makeshift performance enhancement diet I did not gain a pound or throw any farther
and I began to greatly dislike ice cream. That was my introduction, at a Vermont level, to doping in sport.
Winston Churchill said during the Battle of Britain, summarizing it all with regard to sport and war, Victory at all
costs, victory in spite of all terror, victory, however long and
hard the road may be, for without victory, there is no survival. I am going to provide a perspective on what is going
on with doping; including methods, youth, and performance
enhancement, surveillance and testing.
We have been enthralled with performance-enhancing
substances since the early 1930s when the cartoon character
Popeye was first created. He could eat spinach, immediately
become this super strong individual and basically take care
of all the evils he encountered. When I was a kid you could
send away to Joe Weider and for 10 cents receive a basic
training schedule that would allow you to become Mr. America. Now we put a lot of pressure on ourselves, and especially
our children, to perform in sports. This is an issue of significant concern for pediatricians and all health care profes-
* Former Chair, United States Olympic Anti-Doping Committee.
0022-5347/07/1784-1733/0
sionals. When you go to a Little League game, a Junior

League basketball tournament, a swim meet or a rowing
competition, you can see that we are pushing our kids beyond their physical capabilities.
Historically, the ancient Greeks used mushrooms, as
some of the psychedelic and stimulant characteristics of
these substances were thought to help in performance. In
the 1900s strychnine was used and, although a potent stimulant, it can also kill you. Even today, with some of the poly
chemistry being used, we still see athletes who die of overdoses or of drug combinations. In 1936 testosterone was
isolated. In 1967 at the Montreal Games the first formal
steroid tests were given. In 1983 at the Pan Am Games
United States athletes tested positive for anabolic steroids,
and I was one of the U.S. team physicians at those events. In
1984 at the Los Angeles Olympics there was an illegal blood
doping lab set up for the U.S. cyclists. As you may know, in
1980 we did not go to the summer games as the U.S. government was protesting political issues with the Russians.
In 1984 most of the Eastern Bloc countries did not come to
the United States Games. In 1992 at the Barcelona Games
there were U.S. medalists who might have used performance-enhancing substances. And since then through the
most recent Olympic Games there have been athletes who
tested positive.
Banned substance groups are stimulants, narcotics, anabolic steroids and their derivatives, beta-blockers, diuretics
and peptide hormones. There are various other doping methods, such as blood doping, which is not seen much anymore
because athletes are taking EPO, as well as pharmacological, chemical and physical manipulation of the urine, all of
which are banned. Some people may be using manipulation
and substances to mask the illegal drugs during the test.
Other restricted and banned substances are alcohol and
marijuana. A number of athletes in the Winter Games tested
positive for marijuana. Local anesthetics have been used,
especially in sports like boxing, and although athletes are
able to compete while injured, competition after injection for
a joint or some other area of injury is restricted or banned.
Corticosteroids can be an issue not so much with regard to
performance enhancement, but the euphoria often seen with
these drugs is believed to possibly enhance performance,
and so it is banned.
In 1935 testosterone was isolated and used in starvation
studies by the Germans during Word War II in some concentration camps. They studied the effect of anabolic androgenic steroids after prisoners had experienced a long period
of forced starvation. In 1952 there was suspected anabolic
steroid abuse in the Olympic Games. In 1954 Ciba marketed
the first semisynthetic steroid, Dianabol, which still is used
today as a doping agent. Androgenic has the ability to pro-
1733
Vol. 178, 1733-1737, October 2007

Printed in U.S.A.
DOI:10.1016/j.juro.2007.07.071
1734
PEDIATRIC PERSPECTIVE ON DOPING IN SPORTS
duce male secondary sex characteristics, and anabolic is the

capacity to promote tissue growth and repair.
STEROIDS
Do the steroids work? They do but there are adverse sequelae. A woman requires a much smaller dose for a significant enhancing response than a man. Women can suffer
male pattern baldness, clitoral hypertrophy and a deepened
voice, which are permanent, as well as menstrual irregularities and varying degrees of hirsutism. Men can suffer from
testicular atrophy, decreased sperm count, increased libido,
steroid rage, increased prostate size (it is not uncommon for
weight lifters and bodybuilders to have to catheterize themselves because of prostatic hypertrophy) and gynecomastia.
Some of the adverse sequelae that affect males and females
include hypertension (including adolescents), arteriosclerosis, acne, muscle tendon tears (the muscle becomes too big,
too large and too powerful for its tendon interface), acromegalic changes and thrombogenic activity. The sequelae of
these drugs can be significant. In younger children premature closure of the epiphyses and growth retardation may
occur. Many professional bodybuilders undergo subcutaneous mastectomies because they have painful breast tissue
after using anabolic steroids. In addition, some of the anabolic steroids are injected. There have been multiple cases of
transmission of HIV and hepatitis, as well as local problems
with abscess formation with unsterile and shared needles.
If you can go on the Internet and search anabolic steroids
or performance enhancement, you will find a multitude of
sites describing which drug regimens to use or even where to
purchase these illegal drugs. Steroid stacking is a cycling of
various medications, multiple types and doses with the use
of growth hormone, or human chorionic gonadotropin, for
synergism. Use of these drugs can minimize some of the side
effects of the steroids but does not necessarily eliminate the
presence of the steroid in the urine. The Steroid Handbook,
an underground users manual published in California, can
be found on-line.
Some of the accepted medical indications for anabolic
steroids include hereditary angioedema, aplastic anemia,
metastatic breast cancer and hypogonadism. I am sure that
some of you have seen male patients in your practice who
have lost both gonads for some reason and need to receive
testosterone injections. Use of this drug in this setting, while
medically indicated, would require review by U.S. and world
anti-doping authorities before it might be allowed. In the
1990s we were asked to drug test what was touted as a
steroid-free bodybuilding contest. Bodybuilder drug testing has historically been lie detector tests, which is not a
reliable method to determine doping. However, when we
obtained the urine samples, under a strict chain of custody,
there was evidence for significant doping.
Florence Griffith-Joyner was a nationally known, internationally recognized, world-class athlete, who unfortunately died at a young age a few years ago. At issue with her
was that she had a meteoric increase in what she accomplished athletically. She had been tested repeatedly, and
never tested positive throughout her period of competition.
For me, the bottom line is either you do or do not test
positive. All of the rumors only tend to stir the pot and,
unfortunately, after her death it all came up again, as people
questioned her cause of death and whether she was taking

any kind of performance-enhancing drugs.
Ben Johnson was a Canadian runner. If you compare his
appearance 16 months before he competed in the 1988
Olympics to his appearance during the Olympics, there is a
significant difference in the deltoids, and thighs when he
beat Carl Lewis in the 100 meter dash. After he was
awarded the gold medal the urine test was positive for
stanozolol. He lost that record and was banned from the
sport for 2 years. He came back to compete but never was
able to duplicate his previous athletic accomplishments.
Stimulants are also banned, including amphetamines and
sympathomimetics. Caffeine is a banned substance in an
amount that is performance-enhancing. Stimulants decrease fatigue and increase energy but they can mask pain.
Some of the adverse sequelae are aggressiveness, anxiety,
tremors, increased heart rate, blood pressure and dehydration, which can lead to a myocardial infarction or a stroke.
I drug tested the U.S. 1992 Olympic basketball team,
which was the first time pros were on the team. As is required by our protocol, each urine specimen was obtained
under direct visualization. Most of them did not appreciate
what was going on with performance-enhancing drugs. They
knew street drugs but not those taken for performance enhancement. On the other hand, the next time around, in
1996 when I tested most of them again in Portland, Oregon,
they were all well informed of doping issues. And, of course,
all tests were negative. We want a level playing field but for
these professional athletes who are making millions of dollars a year, they did not appreciate what the implications
were to the nonprofessional athletes, as well as to the youth
of our nation.
Theres an official United States Olympic Committee list
of drugs that you can take for certain conditions, whether for
asthma or another medical disorder. Many of the unapproved over-the-counter drugs, including Contac and Drixoral, contain ephedrine and other stimulants, and are not
allowed. A number of high profile U.S. athletes have unwisely not followed the committees instruction and guidelines, and taken nonapproved drugs even while competing at
an Olympic games event. Some have tested positive, and a
few have lost their medals. The Olympic team is carefully
instructed in what drugs they can and cannot take. Even the
youngest of these athletes knows what drugs they are allowed to use. The use of narcotics has not been a significant
problem, but of course these are banned. The athletes need
to be careful, as someone with an injury might be on a
narcotic and would need to declare that drug at the time of
testing. We want to make certain that the U.S. team is
pretested before the games and all the athletes are clean.
BETA-BLOCKERS
How are beta-blockers related to performance enhancement? They decrease anxiety and can steady your hands.
Surgical and urology residents have been known to take
beta-blockers because they want to decrease hand tremor.
They also decrease your heart rate, and that is what is
important in some of these sports. If you look at these events
in terms of decreased heart rate, how would beta blockade
help? In the Winter Olympics biathlon is a sport that combines skiing and shooting. The athlete is skiing aggressively
and then stops and shoots at a target. If you can get your

heart rate down to the 40s and you are literally pulling the
trigger between a heart beat, you may perform better than
the next person. Bobsledding, figure skating, luge and ski
jumping as well as some of the summer sports, such as
archery, diving, synchronized swimming, gymnastics, modern pentathlon, sailing shooting and equestrian, have been
implicated because of improved eye-hand or body coordination with the use of beta-blockers.
DIURETICS
For equestrian events the horses can be doped as well. One
can dope your horse up to, but not during, the competition.
Horses are given diuretics and they may be given other
compounds. The horses and riders are drug tested at the
time of these events. We went to the equestrian associations
and asked them about banning the drugs throughout the
course of training as well as at the events and to do some
random testing. However, they rejected this suggestion and
still allow the horses to be doped before the competition
during their training. The riders are subjected to the drug
testing process at all times.
One can dilute the urine to mask analysis as when
people have been on diuretics. When we examined the
samples the specific gravity reflected almost that of tap
water. You have to obtain multiple samples until the urine
is a certain specific gravity and concentration. Now there are
tests for specific diuretics. Some of the adverse effects of
diuretics, as you well know, include dehydration, electrolyte
imbalance and cardiac arrhythmias.
1735
phy, can increase testosterone production and is difficult to

detect.
EPO is a renal glycoprotein that stimulates red blood cell
production and a major natural stimulus is hypoxia. Now
made by recombinant synthesis, EPO improves endurance
by increasing oxygen carrying capacity. Some of the adverse
effects of EPO include hypertension and increased blood
viscosity. On EPO hemoglobin can increase to 20 or more
and hematocrit can increase to greater than 60 with increased potential for stroke, seizures and myocardial infarction. The old-fashioned method of blood doping was by autologous donation, which is what we dealt with during the
Los Angeles Olympics in 1984. One would donate a unit of
their own blood, which was then auto-transfused after an
athlete trained at high altitude. When one trains at high
altitude there can be a natural increase in hemoglobin and
hematocrit. At the Los Angeles Olympics an illegal transfusion unit for the cycling team was set up in a motel. They
took some blood from nonrelated people, went to a local
hospital and said, We need to have this cross-matched.
Although the hospital staff did not do a complete crossmatch, they did the Rh and ABO tests but apparently did not
complete some of the minor antibodies. When some of the
athletes were transfused a few of them had a reaction. This
was all hearsay but the USOC found out about it. We were
especially concerned for the women who had reactions, as
they were believed to be illegally enhanced U.S. medalists.
The USOC knew that they had won medals but at this time
there was no test for this illegal procedure. In women especially, having a transfusion reaction can lead to lifelong
problems.
HORMONES AND ANALOGUES

Peptide hormones and analogues include human growth
hormone, human chorionic gonadotropin, corticotropin and
erythropoietin. When EPO was first formulated back in the
1980s, we went to the company and requested they tag it so
that it could be identified in the urine when ingested. They
said they could not do that because it had required years to
gain Food and Drug Administration approval and they
would have to go back and resubmit for additional approval.
HGH is now genetically engineered and difficult to detect.
In the past it was extracted from cadaver pituitaries. HGH
stimulates bone growth, increases protein synthesis and
improves collagen synthesis. Some of the adverse effects of
cadaver HGH include Crutchfield-Jacob disease, which is a
form of mad cow disease caused by a contaminating virus
within the human cadaveric pituitary that can be transmitted. Now that HGH is genetically engineered, the virus is no
longer a problem. However, acromegalic changes, thyroid
dysfunction, glucose intolerance, antibody formation, transient edema and slipped femoral capital epiphyses, especially in children, remain adverse consequences. Growth
hormone has been called the bodybuilders perfect drug. The
professional wrestler Andre the Giant was thought to have
suffered from hyperpituitarism as an acromegalic. He was 7
feet 4 inches. He died at the age of 41 of cardiac disease and
was implicated as having used HGH and testosterone.
Whether his medical condition and death were a natural
progression of the acromegaly or something related to performance-enhancing drugs we will never know. Human chorionic gonadotropin is found in the placenta. It counters
some of the side effects of steroids, such as testicular atro-
DETECTION AND REPORTING

How do you avoid detection? Specimen substitution, dilution
of specimen, tampering of the urine, substance substitution,
nondetectable substances and then rapidly metabolized substances, masking agents and timely discontinuance are
some methods used illegally. Timely discontinuance is beneficial to an athlete when testing only at the event, which is
why the USOC instituted a random, unannounced program.
For someone who is a registered athlete, and most of the
elite level athletes are, they are subjecting themselves to
random testing 24/7, 365 days a year. This is something that
we are pushing for professional athletes as well but we are
meeting resistance from professional sports and their athletes. Also there are cooperative efforts between various
organizations and other countries.
The athlete is informed when he/she tests positive and
the national governing body is notified. This is after an A
sample and then there is an appeal process. There should
not be any public announcement after a positive A sample.
The B sample is tested to confirm the A analysis. Marion
Jones, the runner, recently tested positive for EPO on the A
sample, which should have never been publicly announced.
The only people who should have known of the sample
results would have been the athlete and the national governing body. Someone announced the sample results and
when the B sample was tested, it was negative and no
sanction was imposed.
Urine testing is a witnessed void. The athlete voids with
a member of the testing team present. There is a strict chain
of custody and the urine is analyzed at an International
1736
Olympic Committee certified laboratory. Urinalysis is performed using a high resolution mass spectrometer. The normal distribution of testosterone-to-epitestosterone should be
1:1. Someone who is taking testosterone will have a much
higher testosterone-to-epitestosterone ratio, or someone who
is chronically taking an anabolic steroid may have a lower
than normal testosterone due to suppression of the normal
production by the testis. An illegal semisynthetic substance
can be detected in the urine. Urine is a great modality to
examine and test for a number of other substances as well.
For most of these athletes you really have a fingerprint for a
number of months or years of testing as to what the urine
looks like under mass spectrometry analysis. For the testosterone-to-epitestosterone ratio the previous upper limit was
greater than 6:1 but it has now been lowered to 4:1. The
chance of anyone normally having a 4:1 testosterone-toepitestosterone ratio is extremely unlikely. The testosterone-to-epitestosterone ratio for Floyd Landis, the winner of
the Tour de France, was 11:1, which is considered positive
but he claims to be innocent.
OVER-THE-COUNTER DOPING AGENTS

Some doping agents have been available over-the-counter.
You may be asked, as a clinician, to talk about this with
some of your patients. Creatine is a nitrogenous organic
compound that is readily available. The source is meat and
fish. It provides an integral role in energy metabolism and is
thought to increase strength, power and performance for
certain sports. Taking the substance increases the creatine
pool size and the phosphocreatine level. It acts as a hydrogen
ion buffer that transfers energy from adenosine diphosphate
to allow it to convert to adenosine triphosphate. Therefore, if
you are training, your muscles can recover more quickly and,
theoretically, allow them to perform better. Creatine stimulates synthesis and promotes muscle hypertrophy. However,
there is suspicion of long-term consequences of kidney or
cardiac development for some youngsters on a large amount
of a protein load. The manufacturer suggested daily dose
might be 5 or 10 grams. However, for some, if a little is good,
a lot is better, and we have seen young athletes on 40 to 50
grams of creatine a day, which is a significant and possibly
dangerous protein load.
Dehydroepiandrosterone has been available over-thecounter. It is questionably a testosterone precursor often
taken by women to increase muscle tone and alters the
testosterone-to-epitestosterone ratio. Androstenedione, or
andro, has also been easily available. About 15% is metabolized to testosterone, and so if enough is taken it can result
in a testosterone boost. Andro also alters the testosteroneto-epitestosterone ratio. It is banned by baseball now but
there has been a cloud surrounding this substance. In 1997
Mark McGuire broke the home run record, and he announced at a press conference that he had been taking
androstenedione. I tested Mark McGuire in 1984 when he
was a member of the U.S. Olympic baseball team. At that
time he was a 6 foot 5, lean appearing ball player from
University of Southern California. When he broke the record
he was 6 foot 5, weighed much more and had a much different appearance. Before Congress he allegedly admitted taking androstenedione, which was not banned at that time, but
the whole issue of his alleged use of other steroids lingers.
Currently dehydroepiandrosterone is banned, andro is

banned and creatine is allowed.
EDUCATION
What is the issue with kids now? We know that there is a
body image crisis among our young people, especially during
puberty and the early teenage years. Saying Dont do
drugs, and Dont go out after midnight, to our children is
not always effective, as negative reinforcement is not often
successful. There needs to be innovative education. We need
people within their age group and their sport group saying,
You shouldnt be taking these substances. Currently, it is
believed that 4% to 6%, and possibly more, of high school
students across the country are using anabolic steroids,
which is a significant number. For most of them appearance
enhancement is desired rather than performance enhancement. Among police, fire and other professions that require
increased strength, there have been abuses as well. It has
been found that some steroid users have shown signs of drug
dependence.
A high school athlete may be accomplished in a sport,
perhaps at the all-league or all-conference level, but he/she
might not be quite good enough to attend the college of
choice and compete at that level. An anabolic steroid might
allow that student to realize that goal. Since the 1996 Atlanta games there have been significant efforts to promote
drug-free sports. Janet Evans, the outstanding swimmer,
collaborated through the USOC with a number of public
speaking engagements and poster efforts with regard to
drug-free sports. In 2000 the anti-doping efforts were outsourced from the U.S. Olympic Committee to the United
States Anti-Doping Association. On an international level
the World Anti-Doping Association was formed. Both are
independent organizations that work closely with each
other. Their mission is to mandate a drug- free, level playing
field.
BARRY BONDS AND BALCO

In the Bay Area there have been doping allegations regarding Barry Bonds, the San Francisco Giant outfielder. A company called BALCO (Bay Area Lab Cooperative) has been
implicated in a doping scheme. Apparently BALCO was
initially investigated by Jeff Novitsky, an IRS auditor and a
federally sworn officer, from the San Francisco area. He
attended college locally and participated in intercollegiate
athletics. During his investigation of BALCO, he was
searching through the company garbage and discovered suspicious records. He found evidence of much more than just
an IRS issue. BALCO and its owner became implicated in a
doping scheme. It involved pro football players, pro baseball
players, and track and field athletes. The substances they
were using, the so-called clear and the cream, were designer drugs initially undetectable. Someone gave a syringe
with some of this substance to the UCLA Olympic laboratory. Through a laborious process this designer anabolic
steroid that had been previously undetectable was identified, and a grand jury investigation ensued. The complete
story of BALCO, Bonds and the others involved continues to
unfold.

CONCLUSIONS
So I and many others have a serious concern with doping in
sports, which is combined with the emphasis that we as
adults put on the athletic performance of our children. There
are moral, ethical and health issues for the children, not to
mention the billions of dollars that are involved with the
sports industry. However, there is a disturbing public tolerance with doping behavior. In the Bay Area an informal
survey was taken regarding the allegations that Barry
Bonds took steroids and many responded who cares. I
think that we as a nation have a moral obligation to our
patients, our kids and ourselves to address this issue. I think
that we as surgeons, pediatricians, primary caregivers and
1737
family practitioners have an obligation to become involved

in this effort. We need to emphasize that these drugs can
have adverse sequelae on all who use them. The consequences could possibly last a lifetime especially for a child
using these substances.

EPO
HGH
IRS
USCO
erythropoietin
human growth hormone
Internal Revenue Service
United States Olympic Committee
Voiding Dysfunction
Clinical Urological Outcomes Following Primary
Tethered Cord Release in Children Younger Than 3 Years
Amanda M. Macejko,* Earl Y. Cheng, Elizabeth B. Yerkes, Theresa Meyer, Robin M. Bowman
and William E. Kaplan
From the Divisions of Urology and Neurosurgery (RMB), Childrens Memorial Hospital, Chicago, Illinois
Purpose: Current practice at our institution is to recommend tethered cord release at diagnosis to prevent the onset or
worsening of symptoms. Tethered cord release is frequently performed in children younger than 3 years who often have no
urological manifestations. To our knowledge there are currently no long-term data on urological outcomes in this age group.
Materials and Methods: We completed a retrospective review of 475 cases of tethered cord release performed at a single
institution between 1995 and 2002. Of these surgeries 173 were performed in children younger than 3 years. Clinical
outcomes, and preoperative and postoperative urodynamic and radiographic studies were evaluated.
Results: A total of 79 patients met study criteria. Average age at surgery was 9.6 months and average followup was 5.2 years
(range 6 months to 11.2 years). At followup 49 patients (62.1%) had no urological complaints and 30 (38%) had urological
problems. A total of 20 children (25.3%) had minor problems (constipation, delayed toilet training or other) and 10 (12.7%)
had major problems (need for clean intermittent catheterization, febrile urinary tract infection or reflux). Of 66 patients 30
(45.5%) had abnormal preoperative urodynamics. One of 31 patients (3.2%) had hydronephrosis on preoperative ultrasound.
Statistical analysis revealed that abnormal preoperative urodynamics and ultrasound were not predictive of major urological
problems. Lipomatous dysraphism and preoperative musculoskeletal symptoms positively correlated with major urological
problems (p 0.0076 and 0.0484, respectively).
Conclusions: The majority of children did not experience urological problems following tethered cord release. Only a small
set of children had major urological problems. Children with lipomatous dysraphism and musculoskeletal symptoms were
more likely to experience poor urological outcomes.
Key Words: bladder, neural tube defects, urodynamics, urination disorders, spinal dysraphism
ethered cord is abnormal fixation of the spinal cord in

patients with spinal dysraphism. It is thought that
this fixation leads to stretching of the spinal cord, and
subsequent ischemia and hypoxia of the spinal cord and
sacral nerve roots.1 Tethered cord syndrome can be primary
or secondary. Primary tethered cord results from sacral dysgenesis and occult spinal dysraphism. Secondary tethered
cord primarily occurs following surgical repair of myelomeningocele or other dysraphic entities.2 In either case the
hypoxemic damage can result in upper motor neuron deficits, which may include motor or sensory symptoms in the
lower limbs and/or bladder dysfunction.
Currently the timing of TCR is controversial. This is
especially true for primary tethered cord when patients have
primarily cutaneous manifestations, such as a sacral dimple
and minimal to no symptoms. Some groups advocate early
untethering, reporting that it can slow progression or possibly prevent the onset of neurological deficits and bladder
dysfunction.37 Others advocate close monitoring with surgery only after upper motor neuron symptoms occur.8
Current practice at our institution is to recommend TCR
for patients with primary tethered cord at diagnosis in an
* Correspondence: Northwestern University Feinberg School of Medicine, 303 East Chicago Ave., Tarry 16-703, Chicago, Illinois 60611
(telephone: 312-908-0494; FAX: 312-908-7275; e-mail: a-macejko@
md.northwestern.edu).
0022-5347/07/1784-1738/0
effort to prevent the onset or worsening of symptoms. These

procedures are frequently performed in children younger
than 3 years. At this young age urinary continence has not
yet been attained and, therefore, clinical urological manifestations of tethered cord are rarely apparent. To our knowledge there are currently no long-term data on the urological
outcomes of TCR in patients with primary tethered cord in
this age group.
Patient Selection
We completed a retrospective review of 475 cases of TCR
performed at our institution between 1995 and 2002. Of
these patients 173 were younger than 3 years. Only patients
with primary tethered cord were included in the study.
Patient diagnoses included L, DSM, DST, FF and MMM.
Children with myelomeningocele, imperforate anus, cloacal
exstrophy, previous TCR or less than 6 months of followup
were excluded from study.
Urodynamic Studies
Preoperative and postoperative UDS were evaluated and
classified as normal or abnormal. Studies were classified as
abnormal if they demonstrated 1 of decreased compliance in
the filling phase and/or hypotonia, increased post-void residual urine, DSD or increased voiding pressures in the voiding
1738
Vol. 178, 1738-1743, October 2007

Printed in U.S.A.
DOI:10.1016/j.juro.2007.03.193
UROLOGICAL OUTCOMES AFTER PRIMARY TETHERED CORD RELEASE

phase. Since uninhibited detrusor contractions are difficult
to interpret and they are a frequent finding in patients with
a normal bladder in this age group, UDS with isolated uninhibited detrusor contractions were not considered abnormal for study purposes.
Radiographic Studies
Preoperative and postoperative results of renal US and
VCUG were reviewed. US was classified as abnormal if any
degree of hydronephrosis was present. Anatomical abnormalities, such as horseshoe kidney or a duplicated system,
were not considered abnormal in the absence of hydronephrosis. VCUG was defined as abnormal if any degree of
VUR was present. Comments on bladder changes were also
noted.
Clinical Outcomes
Clinic notes were reviewed to evaluate patient clinical status. Additionally, parents of patients were contacted and
information was obtained regarding history of UTI, age at
toilet training, constipation, need for CIC and other urinary
complaints. Minor poor urological outcomes following TCR
included a history of afebrile UTI, constipation and delayed
toilet training, defined as the absence of daytime urinary
continence by age 4 years. Major poor urological outcomes
included the requirement of long-term CIC, history of FUTI
and documented new onset or progression of VUR or hydronephrosis. Followup was calculated from the day of surgery
to the last date of contact with the patient.
Chi-square analysis was performed on variables, including
sex, preoperative diagnosis, preoperative motor symptoms,
and preoperative and postoperative urodynamic and radiographic studies, to determine factors predictive of poor urological outcomes.
RESULTS
Preoperative Parameters
A total of 79 patients met study criteria. The average age at
which TCR was performed was 9.6 months. Mean followup
was 5.2 years (median 4.8, range 6 months to 11.2 years).
Table 1 lists preoperative patient characteristics. The most
common diagnoses in the study group were FF in 26 of 79
patients (32.9%) and lipomatous dysraphism in 26 (32.9%).
A total of 69 patients (87.3%) initially presented with a
cutaneous sacral abnormality, including a dimple, mass or
hemangioma. Of the patients 42 (53.2%) and 37 (46.8%)
were female and male, respectively. A total of 14 patients
(17.7%) presented with musculoskeletal symptoms at diagnosis, including lower extremity weakness, club foot or delayed motor milestones. Only 2 patients (2.5%) had urological symptoms at diagnosis. One patient presented in
urinary retention following cardiac surgery and was found to
have a tethered cord on MRI. The other patient had a sacral
dimple with 2 FUTIs before TCR.
Preoperatively 36 of 66 patients (45.5%) had normal UDS
and 30 (54.5%) had abnormal UDS. Table 2 lists preoperative urodynamic abnormalities.
A total of 31 patients were evaluated with preoperative
US, of whom 30 (96.8%) were found to be normal, whereas 1
1739
TABLE 1. Preoperative patient characteristics

No. Pts (%)
Diagnosis:
L
DST
DSM
FF
MMM
Sex:
M
F
Presentation:
Cutaneous
Musculoskeletal
Urological
Preop UDS:
Normal
Abnormal
Preop US:
Normal
Abnormal
Preop VCUG:
No reflux
Reflux
26 (32.9)
18 (22.8)
5 (6.33)
26 (32.9)
4 (5.1)
37 (46.8)
42 (53.2)
69 (87.3)
14 (17.7)
2 (2.5)
36 (54.5)
30 (45.5)
30 (96.8)
1 (3.2)
11 (91.7)
1 (8.3)
(3.2%) demonstrated grade I hydronephrosis. Of the patients

12 underwent preoperative VCUG, of which 1 (8.3%) was
positive for bilateral grade II VUR and 1 (8.3%) revealed
mild bladder trabeculation.
Postoperative Outcomes
Table 3 lists postoperative urological outcomes. A total of 49
patients (62.1%) had no urological complaints following surgery and 30 (38%) complained of minor or major urological
problems. The 20 children (25.3%) with minor urological
complaints experienced constipation, delayed toilet training,
afebrile UTIs and/or dysfunctional voiding. Ten patients
(12.7%) had major urological problems, including the need
for CIC in 6, FUTIs in 8, new onset hydronephrosis in 5,
and/or new onset or progression of VUR in 5. Interestingly
the 2 children who presented with urological complaints
before TCR showed improvement and neither experienced
major urological problems.
Risk Factors for Poor Urological Outcomes
Table 4 lists clinical characteristics, and urodynamic and
imaging results for each of the 10 patients with major urological problems. Table 5 shows the statistical correlation of
clinical characteristics and preoperative studies with major
urological problems.
Eight children (80%) with major urological problems were
female, while only 2 (20%) were male. This compares with 35
male patients (50.7%) and 34 female patients (49.3%) without major urological problems. This difference approached
but did not attain statistical significance (p 0.0688).
Seven of the 10 children (70%) with major urological
problems were diagnosed with lipomatous dysraphism. The
other 3 children with major urological problems were diagnosed with DST (2) and FF (1). Therefore, 7 of 26 children
(26.9%) with lipomatous dysraphism had a major urological
problem in comparison to only 3 of 53 (5.6%) with any other
primary tethered cord diagnosis (p 0.0076). Children with
FF in our cohort were less likely to have a poor urological
outcome, although this did not achieve statistical significance (p 0.099).
1740

TABLE 2. Preoperative UDS
Diagnosis
No. Pts (%)
Hyperreflexia
Hypotonia
Noncompliant
DSD
High voiding pressure
1 (3.3)
14 (46.7)
2 (6.7)
7 (23.3)
6 (20)
The figure shows preoperative and postoperative urodynamic results in all patients. When examining this group as
a whole, it is clear that there was a significant amount of
crossover between normal and abnormal, preoperative and
postoperative study results. Statistical analysis showed no
correlation between preoperative and postoperative urodynamic results and clinical outcomes. All 10 children with
major urological problems underwent UDS before and following TCR. Four children (40%) had abnormal preoperative
UDS, including a hypotonic bladder in 3 and high voiding
pressure in 1. Statistical analysis revealed that abnormal
preoperative UDS did not predict poor urological outcomes
(p 0.71). Postoperatively 7 children (70%) had abnormal
UDS, including hypotonia in 5, hyperreflexia in 1 and high
voiding pressure in 1. Abnormal postoperative UDS also did
not correlate with major urological problems postoperatively
(p 0.14). Furthermore, no specific preoperative or postoperative urodynamic abnormality correlated with major urological problems postoperatively.
Evaluation of radiographic studies revealed that 6 of 10
children (60%) with poor urological outcomes underwent preoperative US, which was normal in all. Hydronephrosis in the
only child with abnormal preoperative US resolved postoperatively and this child did not experience any major urological
sequelae. One of 12 VCUGs (8.3%) performed preoperatively
demonstrated bilateral grade II reflux. This male patient experienced reflux progression postoperatively. Additionally, 4
other children showed new onset reflux following TCR. In
terms of bladder morphology 1 child had mild bladder trabeculations preoperatively, which normalized on VCUG following surgery. This child did not subsequently experience
major urological problems. Three of 5 children (60%) with
reflux also had evidence of abnormal bladder morphology on
postoperative VCUG. No other children were reported to
have abnormal bladder morphology postoperatively. Neither
preoperative hydronephrosis on US nor VCUG findings correlated with postoperative clinical outcomes.
The 4 children (40%) who showed major urological problems had preoperative musculoskeletal symptoms, including
lower extremity weakness in 2, delayed walking in 1 and
club foot in 1. On the other hand, only 10 of 69 children
(14.5%) without poor urological outcomes had preoperative
motor deficits. Thus, preoperative motor symptoms positively correlated with poor urological outcomes (p 0.048).
Additionally, 4 children (40%) with poor urological outcomes required repeat TCR, whereas only 4 of 69 (5.6%) without major poor urological outcomes required a second surgery
(p 0.0008). In these cases urological symptoms often precipitated the assessment leading to the diagnosis of retethering.
DISCUSSION
To date the timing of TCR in patients with primary tethered
cord has been controversial. Previous reports suggest that
urological problems are improved and/or prevented by TCR.

The practice at our institution is to perform TCR at diagnosis, rather than wait for upper motor neuron or bladder
symptoms to become evident. Because these children are not
yet continent, most do not have urological manifestations at
diagnosis. We reviewed the urological outcomes in children
undergoing TCR for primary tethered cord before the age of
toilet training.
A review of the literature demonstrated that few groups
have examined the urological outcomes of prophylactic TCR
in infants. In 2004 Noguiera et al reviewed the records of
patients following TCR.9 A subset of these children included
16 who presented with cutaneous lesions. This group of
patients is similar to our study group in that all except 1
child presented before toilet training. Noguiera et al found a
25% incidence of preoperative urodynamic abnormalities,
which improved following surgery. Based on their findings
they advocated urodynamic evaluation in all patients since
it is not possible to predict which patients have abnormalities. Additionally, they reported a 31% rate of reflux that
improved in 2 of 5 patients (40%) following TCR. In their
study no patient was reported to experience UTI.
Atala et al studied children undergoing TCR for lipomyelomeningocele in 1992.6 Of the children 29 were 15 months or
younger at surgery. Of the 29 patients 11 (38%) had abnormal
urodynamics at diagnosis. They reported that 82% of patients
had improved bladder function following surgery. While this
group also supports urodynamic assessment preoperatively,
they did not comment on clinical urological outcomes.
These 2 studies with a small number of patients represent the only previous reports of prophylactic TCR specifically in children before toilet training.6,9 They show that
25% to 38% of children in this age group have urodynamic
abnormalities preoperatively and the majority improved
urodynamically following surgery. However, these investigators did not comment on clinical urological outcomes. Reporting urodynamic results in the absence of clinical correlation could be potentially misleading because our findings
suggest that urodynamic findings do not always correlate
with clinical outcomes in this age group. Furthermore, these
previous groups did not comment on risk factors for poor
urological outcomes.
In our series of patients most children did not show
urological problems following TCR. However, a small minority of children experienced major urological problems postoperatively, including FUTIs, hydronephrosis, reflux and/or
the requirement of CIC. We found that the diagnosis of lipomatous dysraphism and the presence of preoperative musculoskeletal symptoms were the only factors that correlated with
TABLE 3. Urological outcomes after TCR

Postop Urological Outcome
No. Pts/Total No. (%)
No problems
Minor problems:
Constipation
Delayed toilet training
Other
Major urological problems:
CIC
FUTI
VUR
New onset hydronephrosis
49/79 (62.1)
20/79 (25.3)
13/79 (16.5)
6/43 (14)
7/79 (8.9)
10/79 (12.7)
6/79 (7.6)
8/79 (10.1)
5/79 (6.3)
5/79 (6.3)
1741
TABLE 4. Characteristics of children with major poor urological outcomes

Pt
No.Sex
Urological
Problem
UDS
Diagnosis
Preop
US
Postop
Preop
VCUG
Postop
Preop
Postop
Bilat grade III
VUR, bladder
wall
irregularity
No VUR, no
bladder
irregularity
No VUR, no
bladder
irregularity
None
1M
FUTI, VUR
Normal
Initially normal,
later DSD
Normal
Normal
2F
FUTI
DST
Hypotonic
Normal
Normal
Normal
Bilat grade II
VUR, no
bladder
irregularity
None
3F
CIC
High voiding
pressure
High voiding
pressure
None
Normal
None
4M
FUTI, CIC,
hydronephrosis
VUR
Normal
Hyperreflexia
None
None
FF
Normal
Hypotonic
Normal
Initially normal,
later abnormal
Normal
6F
FUTI, VUR, CIC,

hydronephrosis
Normal
Initially normal,
later
hypertonia
Normal
Initially normal,
later abnormal
No VUR, no
bladder
irregularity
7F
FUTI,
hydronephrosis
Hypotonic
Hypotonic
Normal
Abnormal
None
8F
FUTI, VUR, CIC,

hydronephrosis
Normal
Hypotonic
Normal
Abnormal
No VUR, no
bladder
irregularity
9F
FUTI, CIC
DST
Hypotonic
Hypotonic
None
Normal
None
FUTI, VUR, CIC,

hydronephrosis
Normal
Hypotonic
None
Normal, later
abnormal
No VUR, no
bladder
irregularity
5F
10F
major urological problems postoperatively. This information

may be helpful when counseling parents preoperatively.
It is routine for many of these children to undergo extensive urological evaluation before TCR, including radiographic studies (US and VCUG) and UDS. Interestingly
none of these routine preoperative studies correlated with
the clinical urological outcome following TCR. The finding
that preoperative and postoperative UDS were not predictive of urological problems was particularly surprising. A
possible explanation for this lack of correlation is that it is
well known that performing UDS in this age group can be
TABLE 5. Risk factors for major urological problems
None
Bilat grade II
VUR, no
bladder
irregularity
Unilat grade I
VUR,
trabeculated
bladder
No VUR, no
bladder
irregularity
Unilat grade
III VUR, no
bladder
irregularity
No VUR, no
bladder
irregularity
Rt grade III, lt
grade I VUR,
trabeculated
bladder
Preop Motor
Symptoms
No
No
Lower
extremity
weakness
Delayed
walking
Club foot
No
No
Lt leg
weakness
No
No
difficult and challenging. Thus, the results and interpretation of the study can be somewhat subjective. However, in
this study we evaluated urodynamic parameters that are
objective in nature and less open to interpretation. In addition, investigators at our UDS laboratory are experienced
with performing studies in young children and they make
every attempt to eliminate confounding factors related to
patient age. Thus, it appears from these results that one
cannot reliably use urodynamic studies for prognosticating
the urological outcome following TCR. However, it is not
clear from this data as to the importance of UDS for following patients as they age and how they may impact clinical
management. Future studies are needed to better determine
the usefulness and necessity of preoperative and postopera-
No. Problems (%)

Major
Overall
Diagnosis:
LM
DST
DSM
FF
MMM
Sex:
F
M
Preop motor symptoms:
Yes
No
Preop UDS:
Normal
Abnormal
Preop US:
Normal
Abnormal
10
No Major
p Value
69
7 (70)
2 (20)
0
1 (1)
0
19 (27.5)
16 (23.2)
5 (7.2)
25 (36.2)
4 (5.8)
0.0076
0.8222
0.3791
0.099
0.4345
8 (80)
2 (20)
34 (49.3)
35 (50.7)
0.0688
4 (40)
6 (60)
10 (14.5)
59 (85.5)
0.0484
6 (60)
4 (40)
30 (53.6)
26 (46.4)
0.71
6 (100)
0
24
1
0.618
(96)
(4)
Preoperative and postoperative UDS. Children without postoperative urodynamics had no major urological problems. First number in
each box represents total number of patients in that group. First
and second numbers in parentheses represent those with and without major urological problems, respectively.
1742
tive urodynamic studies in this age group of patients who

undergo primary TCR.
Collectively the findings in this study call into question
the necessity of preoperative radiographic and urodynamic
studies before TCR in infants and young children. One can
make the argument that, since none of these individual
studies are predictive of poor outcomes or alter the decision
of whether to perform TCR, routine radiographic studies
and UDS should not be part of the standard preoperative
assessment. In addition, since a minority of patients have
urological problems postoperatively, these studies appear
unnecessary in the majority of patients and they are not
cost-effective for management.
However, several unknowns remain with regard to the
long-term followup of these patients and the pertinence of
these studies. For example, it is extremely important to
have baseline objective urological data when one is attempting to determine if a patient has deterioration secondary to retethering. In this small subset of patients
baseline and followup radiographic and urodynamic data
may prove to be valuable. Thus, at our institution we
continue to perform routine radiographic and urodynamic
studies preoperatively and postoperatively in children
younger than 3 years who undergo TCR until we can
better define their appropriateness. Hopefully further
long-term followup and future studies would provide more
definitive answers regarding the usefulness and costeffectiveness of the routine use of these studies and
whether there may be a certain subset of patients in
whom they are most appropriate, such as those at greatest
risk for urological problems and/or retethering.
CONCLUSIONS
Most children younger than 3 years did not experience urological problems following TCR. Only a small set of children
had major urological problems postoperatively. The results
of preoperative urodynamic and radiological studies did not
predict the development of poor urological problems. Children with preoperative musculoskeletal symptoms and lipomatous dysraphism may be more likely to experience poor
urological outcomes.
ACKNOWLEDGMENTS
Bryan Voelzke initiated the subject research and contributed to the beginning of this article.

CIC
DSD
DSM
DST
FF
FUTI
L
MMM
TCR
UDS
US
VCUG
VUR
clean intermittent catheterization

detrusor-sphincter dyssynergia
diastematomyelia
dermal sinus tract
fatty filum
febrile UTI
lipoma, lipomeningocele and
lipomyelomenigocele evaluated together
myelomeningocele manqu
tethered cord release
urodynamics
ultrasound
REFERENCES
1.
Cornette L, Verpoorten C, Lagae L, Van Calenbergh F, Plets C,

Vereecken R et al: Tethered cord syndrome in occult spinal
dysraphism: timing and outcome of surgical release. Neurology 1998; 50: 1761.
2. Boemers TM, van Gool JD and de Jong TP: Tethered spinal
cord: the effect of neurosurgery on the lower urinary tract
and male sexual function. Br J Urol 1995; 76: 747.
3. Yoneyama T, Fukui J, Ohtsuka K, Komatsu H and Ogawa A:
Urinary tract dysfunctions in tethered spinal cord syndrome:
improvement after surgical untethering. J Urol 1985; 133:
999.
4. Hellstrom WJ, Edwards MS and Kogan BA: Urological aspects
of the tethered cord syndrome. J Urol 1986; 135: 317.
5. Kaplan WE, McLone DG and Richards I: The urological manifestations of the tethered spinal cord. J Urol 1988; 140: 1285.
6. Atala A, Bauer SB, Dyro FM, Shefner J, Shillito J, Sathi S et al:
Bladder functional changes resulting from lipomyelomeningocele repair. J Urol 1992; 148: 592.
7. De Gennaro M, Lais A, Fariello G, Caldarelli M, Capozza N,
Talamo M et al: Early diagnosis and treatment of spinal
dysraphism to prevent urinary incontinence. Eur Urol 1991;
20: 140.
8. Lagae L, Verpoorten C, Casaer P, Vereecken R, Fabry G and
Plets C: Conservative versus neurosurgical treatment of
tethered cord patients. Z Kinderchir, suppl., 1990; 45: 16.
9. Nogueira M, Greenfield SP, Wan J, Santana A and Li V: Tethered cord in children: a clinical classification with urodynamic correlation. J Urol 2004; 172: 1677.
EDITORIAL COMMENTS
These authors attempt to provide clarity in an area fraught
with ambiguity. Not everyone agrees on who should be categorized as having tethered cord, how they should be evaluated and who should be treated. However, infants who
present with cutaneous lesions appear to be a subset of
patients who have a predictable clinical outcome. As reported by these authors and others, the majority do not have
major neurourological problems in later childhood if they
undergo early cord untethering (reference 9 in article).1 Almost 90% of the patients in this study had no or minor
urinary symptoms after 5 years of followup. Older children,
who are found to have tethered cord after years of voiding
dysfunction and/or recurrent UTI, fare less well and cord
untethering often does not improve symptoms or urodynamic abnormalities (reference 9 in article). Seven of the 10
patients in this series who had major urological difficulties
also had associated lipomatous dysraphism and, therefore,
they may represent a subset of this subset of infants. These
children presumably also comprised the majority of the 4
with poor urological outcomes who required repeat untethering when older.
In Chicago UDS were not predictive of outcome in these
infants. It is difficult to perform UDS in otherwise normal
and sensate infants, which may be part of the explanation. It
is also possible that, if left untreated, a large number would
have manifested urodynamic abnormalities at a later date.
Since videourodynamics were not routinely done, it is also
possible that some children with reflux or bladder wall abnormalities were not identified before cord untethering. In
our experience the majority of preoperative urodynamic
studies were normal in this group, although a few demonstrated hyperreflexia, which resolved postoperatively in all

cases (reference 9 in article). UDS still have a role in longterm management, although many studies are normal. They
may be most useful for documenting the need, admittedly
not common, for secondary cord surgery. Given that 9 of the
patients (90%) with severe urological problems had reflux or
bladder wall changes on postoperative studies, videourodynamics should be done at followup.
Average followup was 5 years. Growth continues until
adolescence and it would be interesting to learn whether
these children continue to do well as teenagers. Perhaps the
authors would be able to provide an update in several years.
Saul P. Greenfield
Womens and Childrens Hospital
Buffalo, New York
1.
Keating MS, Rink RC, Bauer SB, Krarup C, Dyro FM, Winston
KR et al: Neurourologic implications of the changing approach in management of occult spinal lesions. J Urol 140:
1299, 1988.
1743
Tethered cords have interested urologists because they are

found in patients with spina bifida and other neurourological disorders. They can also occur incidentally. However,
who benefits most from an untethering operation? Little
data exist on this point. This study adds to the literature but
no clear answer is present. It is a retrospective review of a
mixed group of children younger than 3 years who underwent tethered cord release surgery. Within the limitations of
a retrospective study more information is needed. Why was
the surgery done and when did postoperative concerns develop? In the discussion there are assertions about the lack
of usefulness of imaging and UDS without clarifying followup and findings. UDS are predictive of future concerns,
eg high storage pressure. Clinical manifestations, such as
hydronephrosis, calculi, pyelonephritis, scarring and renal
damage, may take months and years to develop. The
6-month followup in this study may not be sufficient.
Julian Wan
University of Michigan
Ann Arbor, Michigan
Treatment of Nocturnal Enuresis in Children

With Attention Deficit Hyperactivity Disorder
Boris Chertin,* Dmitry Koulikov, Wael Abu-Arafeh, Yoram Mor, Ofer Z. Shenfeld
and Amicur Farkas
From the Departments of Urology, Shaare Zedek Medical Center, Jerusalem, Faculty for the Health Science, Ben-Gurion University of the
Negev, Beer Sheva and Chaim Sheba Medical Center, Faculty of Medicine, Sackler School of Medicine, Tel Aviv University (YM),
Tel Aviv, Israel
Purpose: Children with attention deficit hyperactivity disorder disproportionately experience voiding dysfunction and
persistent nocturnal enuresis due to a combination of sphincter and detrusor overactivity and nocturnal polyuria. The
different treatment approaches to nocturnal enuresis often fail in these patients. Therefore, we performed a prospective study
to compare the efficacy of combination therapy with desmopressin and oxybutynin vs the tricyclic antidepressant imipramine
in patients with attention deficit hyperactivity disorder who have nocturnal enuresis.
Materials and Methods: A total of 54 patients with attention deficit hyperactivity disorder and nocturnal enuresis were
randomly stratified into 2 groups. Demographic data on patient age and gender were identical in the 2 groups. Functional
bladder symptoms were judged using the dysfunctional voiding symptoms survey. The initial dysfunctional voiding symptoms
survey score was similar in the 2 groups. The total survey score was compared between the 2 groups in aggregate as well as
specifically regarding the incidence of nocturnal enuresis following treatment.
Results: The first group consisted of 27 patients who received desmopressin and oxybutynin, and the second group of 27 was
treated with imipramine. Of the 27 children in each group 23 (85%) received methylphenidate for attention deficit hyperactivity disorder. The mean SD initial dysfunctional voiding symptoms survey score in groups 1 and 2 was 20.5 3.3 and
20.9 4.1, respectively. Following treatment the mean survey score decreased significantly in groups 1 and 2 (6.5 2.5 and
9.4 2.1, respectively, p 0.001). However, between groups analysis showed that the dysfunctional voiding symptoms survey
score was significantly lower in group 1 than in group 2 (mean 6.5 0.5 vs 9.6 0.4, p 0.001). There was also a statistically
significant decrease in the incidence of nocturnal enuresis in group 1 (survey question 2 score 0.9 0.2 vs 2.9 0.2).
Conclusions: Our data show that there is a high incidence of voiding dysfunction in children with attention deficit
hyperactivity disorder. Combination therapy with desmopressin and oxybutynin is a feasible, safe and effective treatment for
nocturnal enuresis in these children.
Key Words: bladder, oxybutynin, nocturnal enuresis, attention deficit disorder with hyperactivity,
deamino arginine vasopressin
ttention deficit hyperactivity disorder is relatively

common, affecting approximately 3% to 5% of children in the Western world. ADHD manifests as impulsive, hyperactive, inattentive behavior.1,2 Symptoms
present before age 6 years and they are noted in 2 distinct
settings (home and school).
Bedwetting or NE is a common clinical disorder in Western countries.35 It is well known that NE may have a
profound psychological and social impact on the affected
child. Furthermore, almost all studies of the comorbidity
between primary NE and psychopathology in children suggest an increased prevalence of behavioral disorders in general and of ADHD in particular.6 9 Some investigators
pointed out that the combination of nocturnal polyuria with
bladder and sphincter overactivity may be responsible for
almost 25% of NE episodes in children with ADHD.7 Different treatment modalities have been suggested for NE in
* Correspondence: Department of Urology, Shaare Zedek Medical

Center, Jerusalem, 91031, P. O. 3235, Israel (telephone: 972-26555560; FAX: 972-2-6555062; e-mail: bchertin@yahoo.com).
0022-5347/07/1784-1744/0
children with ADHD, including urotherapy, DDAVP and the

tricyclic antidepressant imipramine. However, none of these
therapies have shown significant effectiveness for ADHD
with voiding dysfunction.7,8 Therefore, we performed a prospective study to compare the effectiveness of combined
treatment with DDAVP and oxybutynin vs the tricyclic antidepressant imipramine.

A total of 54 patients with ADHD who were referred for NE
were randomly stratified into 2 groups. Group 1 of 27 patients was treated with DDAVP and oxybutynin, while 27 in
group 2 received imipramine. We excluded children with
mental retardation, anatomical abnormalities of the urinary
system on imaging and psychosis. The table shows clinical
and demographic data on the 2 groups.
Functional bladder symptoms were judged using DVSS.
A DVSS was designed at Hospital for Sick Children in Toronto for quantification voiding symptoms in children.10 We
used an established, modified form of that DVSS, consisting
1744
Vol. 178, 1744-1747, October 2007

Printed in U.S.A.
DOI:10.1016/j.juro.2007.03.171
NOCTURNAL ENURESIS AND ATTENTION DEFICIT HYPERACTIVITY DISORDER
1745
Demographic and clinical data on 54 studied children
No. boys
No. girls
Mean SD age (yrs)
Mean SD symptom duration
Mean SD initial DVSS score
Group 1
Group 2
17
10
8.07 2.13
2.2 0.87
20.9 4.2
17
10
8.22 2.19
2.2 0.98
20.5 3.3
of 10 questions, omitting the question on stressful events in

the life of the child, as previously suggested (see Appendix).6
Each question in the questionnaire was scored on a 0 to 3
scale, including 0 never, 1less than half of the time,
2about half of the time and 3almost always.
The initial DVSS score was similar in groups 1 and 2
(mean SD 20.8 4.18 and 20.5 3.31, respectively,
p 0.05). The total DVSS score was compared between the
2 groups in aggregate as well as specifically regarding the
incidence of NE following treatment (DVSS question 2). The
child and parents were asked to complete the questionnaire
together.
A bladder training program designed to decrease bladder
overactivity and engage the child in the treatment process
commenced at the beginning of the treatment protocol. It
included an increase in daytime fluid intake, the need to
respond to any sense of urgency in urination, timed voiding
every 2 to 3 hours or during school breaks to establish
cognitive control over voiding and treatment of encopresis, if
it existed. During the bladder training program child compliance was evaluated and DVSS was reevaluated. Only
children with compliance participated in the study. Since
none of the children demonstrated significant improvement
on DVSS, we proceeded with medical therapy.
Group 1 patients received DDAVP at an initial dose of 0.2
mg with an increase to 0.4 mg if the episodes of nighttime
wetting did not decrease at least 50%. Also, oxybutynin was
given at 0.2 mg/kg 3 times during the day. In group 2 imipramine was given at a dose of 25 mg. Patients were followed at
least 1 year. Success was defined as symptom resolution for 6
months. We attempted to wean the children from the medications 1 year after treatment began, while continuing timed
voiding, constipation treatment when indicated and dietary
FIG. 2. DVSS before and after imipramine treatment
suggestions. GraphPad Prism, version 4.0 for Windows

was used for statistical evaluation with the paired and unpaired t test, and p 0.05 considered significant.
RESULTS
Demographic data on patient age and gender were identical
in groups 1 and 2. There were 17 boys and 10 girls in each
group. Mean age in groups 1 and 2 was 8.1 2.1 and 8.2
2.2 years, respectively. Of the 27 children in each group 23
(85%) received methylphenidate for ADHD. Of the 54 children in the 2 groups 39 (72%) had inattentive ADHD. Following treatment the DVSS score decreased significantly in
group 1 from 20.8 4.18 to 6.5 2.5 and in group 2 from
20.5 3.31 to 9.4 2.1 (figs. 1 and 2, p 0.001). However,
between groups analysis showed that the posttreatment
DVSS score was significantly lower in group 1 than in group
2 (mean 6.5 0.5 vs 9.6 0.4, p 0.001, fig. 3). There was
also a statistically significant decrease in the incidence of
NE in group 1 vs group 2 (DVSS question 2 score 0.9 0.2
vs 2.9 0.2, p 0.0001, fig. 4). None of the children in group
2 were completely dry during the study period.
DISCUSSION
NE can be a distressing, humiliating and perplexing experience for children and young people.35 It may lead to
FIG. 1. DVSS before and after DDAVP and oxybutynin treatment
FIG. 3. DVSS in patients in 2 groups
1746
FIG. 4. DVSS question 2 in patients in 2 groups
feelings of guilt and shame, avoidance of social activities, a

sense of difference from others, victimization and the loss of
self-esteem.4 ADHD is another well-known condition with a
3% to 5% prevalence in 6 to 12-year-old children.1,2
Previous studies strongly suggested a high incidence of
NE in children with ADHD.6 9 On the other hand, most
mental health practitioners have pointed out that children
with ADHD disproportionately experience various form of
urinary incontinence, such as NE, dysfunctional voiding and
diurnal incontinence. To our knowledge there is little in the
literature addressing the treatment of children with ADHD
who experience NE.
Crimmins et al retrospectively evaluated their experience
with the treatment of children with ADHD and NE.7 Different treatment modalities were used in these children, such
as an enuretic alarm, DDAVP and imipramine. The investigators noted a 68% rate of diurnal continence with timed
voiding and anticholinergics. Moreover, they found that children responded efficaciously to imipramine and DDAVP,
while showing a low cure rate with an enuretic alarm. They
concluded that the current modalities for NE in ADHD children are inadequate.
It was suggested that voiding disturbances, particularly
NE, in children with ADHD are due to nocturnal polyuria
and detrusor overactivity.7 It is well known that urine output is regulated by circadian secretion of AVP. Moreover, it
was reported that DDAVP, a pharmacological analogue of
AVP, regulates urinary output, is also involved in the regulation of dopamine pathways and enhances the brain response to dopamine.11 The tricyclic antidepressant imipramine has been used for years to treat NE, particularly in
children with ADHD. The advocators of imipramine treatment in children with ADHD and NE pointed out that the
tricyclic antidepressant may treat the 2 disorders at the
same time through the dopamine pathways. However, because of its cardiac toxicity and possibility of death by overdose in children, imipramine has largely fallen out of favor
in the pediatric population.3 Recently the selective norepinephrine reuptake inhibitor reboxetine was effective for primary NE in 12 to 15-year-old children with ADHD.12
However, low response rates in terms of daily bladder overactivity with these pharmacological formulations justify the
search for new treatment modalities. Data from the current
study support this. Use of a validated voiding survey such as
DVSS demonstrated a high incidence of voiding dysfunction in
the study population. Also, voiding problems were severe with

a mean DVSS score of 20 in children in each group. The high
DVSS score leaves no doubt about the need for appropriate
treatment of bladder overactivity. It was reported that bladder
overactivity is responsible for 44% of treatment failures in
children with monosymptomatic NE. Yeung et al reported that
even in children with normal daytime urodynamic studies unstable bladder contractions may be found following sleep, resulting in enuresis.4,5 Therefore, it is extremely important to
take care of bladder instability to achieve patient dryness not
only during the day, but also at night.
Antimuscarinic drugs are the most common agents used
to treat overactive bladder.13,14 Oxybutynin has proved to be
highly effective and safe for children with dysfunctional
voiding. Moreover, in contrast to the adult population, there
is no fear about disrupting cognitive function with nonselective antimuscarinic drugs in the pediatric population.14 In
the current study combined therapy with DDAVP and oxybutynin resulted in significant improvement in DVSS compared to that in the imipramine group. Moreover, when the
incidence of NE was compared between the 2 groups, children in group 1 did significantly better.
It is well known that NE relapse often occurs following
the cessation of DDAVP treatment. We adopted the more
than 8-week withdrawal DDAVP program, as suggested by
Butler et al,15 while keeping the children on timing voiding
and oxybutynin. It focuses on engaging the child in the
process of internalizing success and highlighting the effective process by which this is accomplished. We believe that
engaging the child using the slogan, I am dry, not because
I have been treated, but because I am dry, with target
therapy, leading to a decrease in nocturnal urine production
and bladder instability, enables the achievement of a reasonable success rate in this difficult patient population.
Another important issue is the influence of successful NE
treatment on psychological status in this cohort of children.
Changes in the Test of Variables of Attention score following
treatment was not evaluated in the current study. However,
we intend to address that in our further studies, which are
ongoing at our institution.
CONCLUSIONS
Our data show that there is a high incidence of voiding
dysfunction in children with ADHD. Combination therapy
with desmopressin and oxybutynin is a feasible, safe and
more effective treatment for NE than imipramine in these
children. The impact of this treatment on the psychological
status of the child will be addressed in further studies.
APPENDIX
DVSS in the Last Month
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
I have had wet clothes or wet underwear during the day

I wet the bed at night
When I wet myself, my underwear is soaked
I miss having a bowel movement every day
I have to push for my bowel movement to come out
I only go to the bathroom 1 or 2 times each day
I can hold onto my pee by crossing my legs
When I have a pee, I cannot wait
I have to push to pee
When I pee, it hurts

8.

ADHD
AVP
DDAVP
DVSS
NE
attention deficit hyperactivity disorder

arginine vasopressin
desmopressin
dysfunctional voiding symptoms survey
nocturnal enuresis
REFERENCES
1.
2.
3.
4.
5.
6.
7.
Clinical practice guideline: diagnosis and management of the

child with attention deficit/hyperactivity disorder. Committee for Quality of Improvement, American Academy of Pediatrics. Pediatrics 2000; 105: 5.
Klassen AE, Miller A and Fine S: Health-related quality of life in
children and adolescents who have a diagnosis of attentiondeficit/hyperactivity disorder. Pediatrics 2004; 114: e541.
Hjalmas K: Nocturnal enuresis. In: Pediatric Urology. Edited
by JP Gearhart, RC Rink and P Mouriquand. Philadelphia:
WB Saunders Co 2001; pp 497510.
Bulter RJ and Holland P: The three systems: a conceptual way
of understanding nocturnal enuresis. Scan J Urol Nephrol
2000; 34: 270.
Yeung CK: Nocturnal enuresis (bedwetting). Curr Opin Urol
2003; 13: 337.
Duel BP, Steinberg-Epstein R, Hill M and Lerner M: A survey
of voiding dysfunction in children with attention deficithyperactivity disorder. J Urol 2003; 170: 1521.
Crimmins CR, Rathbun SR and Husmann DA: Management of
urinary incontinence and nocturnal enuresis in attentiondeficit hyperactivity disorder. J Urol 2003; 170: 1347.
9.
10.
11.
12.
13.
14.
15.
1747
Baeyens D, Roeyers H, Hoebke P, Verte S, Van Hoecke E and

Vande Walle J: Attention deficit/hyperactivity disorder in
children with nocturnal enuresis. J Urol 2004; 171: 2576.
Baeyens D, Roeyers H, Demeyere I, Verte S, Hoebke P and
Vande Walle J: Attention deficit/hyperactivity disorder
(ADHD) as a risk factor for persistent nocturnal enuresis in
children: a two year follow up study. Acta Paediatr 2005;
94: 1619.
Farhat W, Bagli DJ, Capolicchio G, OReilly S, Merguerian P,
Khoury A et al: The dysfunctional voiding system: quantitative standardization of dysfunctional voiding symptoms
in children. J Urol 2000; 164: 1011.
Di Michele S, Sillen U, Engel JA, Hjalmas EK, Rubenson A
and Soderpalm B: Desmopressin and vasopressin increase
locomotor activity in the rat via a central mechanism: implications for nocturnal enuresis. J Urol 1996; 156: 1164.
Toren P, Ratner S, Laor N, Lerer-Amisar D and Weizman A: A
possible effect of reboxetine in children and adolescent with
attention deficit/hyperactivity disorder: case series. Neuropsychobiology 2005; 51: 239.
Caione P, Arena F, Biraghi M, Cigna RM, Chendi RM and
Chiozza ML: Nocturnal enuresis and daytime wetting: a
multicentre trial with oxybutynin and desmopressin. Eur
Urol 1997; 31: 459.
Sommer BR, OHara R, Askari N, Kraemer HC and Kennedy
WA 2nd: The effect of oxybutynin treatment on cognition
in children with diurnal incontinence. J Urol 2005; 173:
2125.
Butler RJ, Holland P, Hilley E and Redfern E: Preventing
relapse following medication for the treatment of childhood
nocturnal enuresis. Paediatr Today 1998; 6: 30.
Spinal Cord Magnetic Resonance Imaging

for Investigation of Nonneurogenic Lower
Urinary Tract DysfunctionCan the Yield be Improved?
K. Afshar, T. Blake, S. Jaffari, A. E. MacNeily, K. Poskitt and M. Sargent
From the Departments of Urological Sciences and Radiology (KP, MS), University of British Columbia, Vancouver, British Columbia,
Canada
Purpose: Magnetic resonance imaging has been used to detect occult neuropathy in patients with nonneurogenic lower
urinary tract dysfunction. There is substantial controversy surrounding the role of this test for lower urinary tract
dysfunction. We identified factors associated with positive magnetic resonance imaging to improve patient selection.
Materials and Methods: A case-control study was done in all pediatric patients referred to our radiology department for
spinal magnetic resonance imaging primarily because of lower urinary tract symptoms between 1995 and 2004. Patients with
known neurological disorders or anomalies associated with neurogenic bladder (overt spinal dysraphism, imperforate anus,
etc) were excluded. A total of 80 patients with a median age of 6.5 years (range 4 to 17) were identified, of whom 47 (59%) were
female. Bivariate analysis was used to evaluate the association of certain variables with positive magnetic resonance imaging
findings, including patient age, gender, type of urinary symptoms, fecal soiling, abnormal neuro-orthopedic examination,
lumbar cutaneous findings, resistance to medical management and urodynamic findings.
Results: Magnetic resonance imaging revealed spinal abnormalities in 6 cases (7.5%), including intradural arachnoid cyst in
1, sacral dysgenesis in 3, syrinx/hydromyelia in 1 and tethered cord in 1. An abnormal lumbar cutaneous finding was the only
variable associated with positive magnetic resonance imaging (Fishers exact test p 0.002).
Conclusions: Spinal magnetic resonance imaging has a low impact in the management of lower urinary tract dysfunction. With
proper patient selection the pretest probability of positive magnetic resonance imaging may be increased and, therefore, many
unnecessary studies may be avoided. Abnormal cutaneous findings are associated with abnormal magnetic resonance imaging.
Key Words: bladder, magnetic resonance imaging, spinal cord, urination, abnormalities
onneurogenic LUTD is one of the most common disorders observed by pediatric urologists. It represents
up to 20% of referrals at some centers1 and it can be
associated with urinary tract infections and renal damage.2
The clinical presentation of LUTD may be similar to that of
neurogenic bladder. Differentiating these 2 entities has high
clinical importance. Various clinical clues, such as abnormal
physical examination or urodynamic findings, can be used to
make the correct diagnosis. Nevertheless, clinicians are regularly faced with cases that do not fit either category.
MRI of the spine has been used to diagnose occult spinal
pathology as the cause of lower urinary tract dysfunction.
However, its application is problematic. MRI in young children may require general anesthesia or intravenous sedation, it is expensive and the usefulness of MRI in this setting
remains controversial. Like any other diagnostic test the
predictive value of MRI depends on pretest probability,
which in turn is a function of the population to which the
test is applied. Therefore, meticulous case selection can improve MRI performance. We identified factors associated
with positive MRI to ameliorate patient selection.
ily because of lower urinary tract symptoms, between 1999

and 2005. Patients with known neurological or urological
diagnosis (spina bifida, spinal tumors, myelitis, imperforate
anus, etc) were excluded. We defined cases as children with
any spinal pathology depicted by MRI and controls as subjects who underwent MRI for the same indications with no
abnormalities identified. MRI was evaluated by 1 pediatric
neuroradiologist (MS).
Based on our literature review we selected 9 variables a
priori that might be associated with abnormal MRI. They
included patient age, gender, type of symptoms (primarily
irritative vs obstructive lower urinary tract symptoms), abnormal neuro-orthopedic physical examination, abnormal
back examination (cutaneous findings), failure to respond to
first line management, fecal soiling and abnormal urodynamic studies.
Univariate analysis was done to evaluate the association
between MRI findings and these variables. The Fisher exact
and the nonparametric Mann-Whitney U tests were used
for categorical and continuous variables, respectively, with
S-PLUS software.

We performed a retrospective case-control study in all children referred to the Department of Radiology at British
Columbia Childrens Hospital for MRI of the spine, primar-
0022-5347/07/1784-1748/0
RESULTS
A total of 80 cases were identified with a median age of 6.5
years (range 4 to 17), of whom 47 (59%) were female. The
1748
Vol. 178, 1748-1751, October 2007

Printed in U.S.A.
DOI:10.1016/j.juro.2007.03.179
SPINAL CORD MAGNETIC RESONANCE IMAGING FOR URINARY TRACT DYSFUNCTION

Univariate analysis of case vs control characteristics
No. MRI/Total No.
Normal Abnormal p Value
% Female
Irritative lower urinary tract symptoms
Obstructive lower urinary tract symptoms
Abnormal physical examination
Abnormal cutaneous findings
Failed management
Fecal soiling
Abnormal urodynamics
47/74
22/74
31/74
42/74
7/74
32/67
22/74
23/52
3/6
3/6
4/6
4/6
4/6
3/6
1/6
4/6
0.27
0.20
0.39
0.99
0.002*
0.43
0.66
0.40
* OR 19 (95% CI 3 to 123).
table shows the distribution of variables in cases and controls.

Abnormal MRI findings included sacral dysgenesis in 3
cases, tethered cord in 1, syrinx in 1 and arachnoid cyst in 1
(see figure). Abnormal lumbosacral cutaneous findings in
this group included sacral lipoma in 2 cases and sacral
dimples high in the gluteal cleft in 2. In children with normal MRI these abnormal cutaneous findings included sacral
dimples in 4 and an asymmetrical gluteal cleft in 3. In all
except 1 case the dimples were superficial. An abnormal cutaneous finding was the only factor associated with positive MRI
(p 0.002). This effect remained unchanged after adjusting for
patient age or gender. Information about conservative management was available in 72 patients, including 67 with normal
MRI. Conservative treatment included bladder retraining with
or without anticholinergics and treatment for constipation. No
other patient characteristics that were selected a priori attained statistical significance.
A total of 31 subjects, including all 6 cases and 25
controls, underwent lumbosacral plain x-ray. Four of the 6
cases showed abnormalities, including sacral bony dysgenesis in 3 and nonfusion of the posterior element in 1.
All sacral dysgenesis cases had an abnormal back examination. In addition, 7 of the 25 controls had open posterior elements.
Of the 80 patients 58 (73%) underwent at least 1 filling
cystometrogram, including 52 of the 74 controls (70%). All
studies were reviewed by 1 of the 2 pediatric urologists
(KA or AEM). Of these 52 controls 23 (44%) had abnormal
cystometrograms compared to 4 of 6 cases with abnormal
MRI findings. Abnormalities in cases included hyperreflexic contractions in 2 and hypocompliance with decreased capacity in 2. In controls the urodynamic abnormalities included atonicity in 4, detrusor overactivity in 4,
decreased capacity and compliance in 6, sensory delay in
1, and mixed findings of overactivity and decreased compliance in 8.
Statistical analysis using Students t test for age and
Fishers exact test for categorical variables revealed that the
only variable associated with abnormal MRI was an abnormal cutaneous finding of the lower back (p 0.002). Due to
our low number of abnormal cases multivariate analysis was
not feasible (see table).
function of the lower urinary tract. Clinical clues along

with diagnostic tests are used to achieve the diagnosis.
MRI of the spine has been used in this regard with variable success. Results of several available studies are
somehow conflicting. This is most likely due to inconsistent patient referral patterns and case selection, affecting
pretest probability. For example, some studies included
only patients with no response to conservative management3 and others were more inclusive.4 In our study the
only pretest characteristic associated with increased odds
of positive MRI (19 times) was abnormal cutaneous findings on the lower back.
Pippi Salle et al performed a study in 23 patients with
normal neuro-orthopedic examination and a history of failed
conservative treatment for dysfunctional elimination syndrome.3 They found 2 positive MRI studies (9%). They concluded that MRI should be used in patients with abnormal
physical findings or complex abnormalities on spinal x-ray.
This was a descriptive study with no statistical analysis.
One of the 2 patients with abnormal MRI had an abnormal
plain x-ray but also had central nervous system abnormalities. A post hoc Fishers exact test of their findings did not
support their conclusion regarding the usefulness of plain
x-ray for selecting patients for MRI (p 0.47). Although our
findings were similar in terms of the positive MRI rate, they
did not show any difference in physical findings between
cases and controls. Since only a minority of our patients
underwent plain spinal x-ray, we could not analyze this
variable, but this tool may be useful in cases of severe bony
abnormalities.
Arikan et al reviewed the records of 44 patients with
normal physical examination who underwent spinal
DISCUSSION
LUTD is a common cause of referral to pediatric urologists. Not uncommonly clinicians are faced with the problem of differentiating between LUTD and neurogenic dys-
1749
MRI of spine shows thoracic arachnoid cyst
1750
MRI.4 Of these patients 17 (39%) had abnormal MRI, of

whom only 5 showed long-term improvement after neurosurgical intervention. Ten of the 17 cases were refractory
to medical treatment. The group concluded that failure of
conservative management should alert the physician to
the diagnosis of neurological disorders. These conclusions
were not based on statistical comparison among patients
with different MRI results.
Wraige and Borzyskowski reported on 543 patients
referred from a neurology clinic, of whom 48 underwent
MRI.5 Five children (10%) had positive MRI findings. The
investigators concluded that abnormal urodynamics are
the most important predictor of abnormal MRI. Their
criteria used to obtain MRI are not clear and no statistical analysis was provided to support their conclusions. In
our study abnormal urodynamics were not associated
with positive MRI. This may have been due to our low
number of cases and we believe that urodynamics are a
valuable test for diagnosis and followup. The usefulness of
MRI for adult LUTD is also limited with a yield of less
than 5%.6
Even when MRI abnormalities are detected, it is controversial whether intervention is beneficial. The outcome
of surgical treatment for spinal pathologies diagnosed
based on urinary tract dysfunction is variable.710 Hellstrom et al reported outcomes in a group of patients with
tethered spinal cord. Improvement in urodynamic and
clinical findings occurred in 10 of 18 patients (55%) after
detethering.7 On the other hand, Arikan et al found that
only 5 of 17 patients (29%) had long-standing improvement in urodynamic and/or clinical parameters.4 This
may be due to dissimilar populations, different definitions
of success and variable followup. Two of our 6 patients
with positive MRI underwent neurosurgical interventions. The girl with tethered cord underwent untethering,
resulting in significant improvement in incontinence. The
girl with thoracic arachnoid cyst showed deterioration
following the neurosurgical operation and upper tract involvement developed. She underwent bladder augmentation and appendicovesicostomy after the failure of medical
management.
Abnormal cutaneous findings in the lumbosacral region
are a potential marker of occult spinal dysraphism. In a
review of 54 asymptomatic cases Guggisberg et al concluded that the presence of 2 or more cutaneous findings is
a strong marker of an underlying abnormal spine.11 It is
possible that in symptomatic patients even 1 skin abnormality points to occult dysraphism, as supported by our
findings.
The current study is not without pitfalls, of which the
most important is its retrospective nature. In addition, we
did not use standard criteria to obtain MRI and patients
were selected according to the clinical judgment of the 3
referring pediatric urologists. This may have resulted in
selection bias. For the same reason not all patients underwent urodynamics. In most patients a combination of 2
factors was the reason for imaging. In addition, the low
number of positive MRIs decreased our power to detect some
significant associations and precluded multivariate analysis. Combining similar data from multiple institutions
would be useful in this regard.
CONCLUSIONS
MRI has a limited role in the diagnosis of spinal abnormalities in children with nonneurogenic LUTD. Clinical clues
may be used to increase the yield of this test. Positive MRI
is associated with lumbosacral cutaneous anomalies.

LUTD nonneurogenic lower urinary tract
dysfunction
MRI magnetic resonance imaging
REFERENCES
1.
Rushton HG: Wetting and functional voiding disorders. Urol

Clin North Am 1995; 22: 75.
2. Koff SA, Wagner TT and Jayanthi VR: The relationship among
dysfunctional elimination syndromes, primary vesicoureteral reflux and urinary tract infections in children. J Urol
1998; 160: 1019.
3. Pippi Salle JL, Capolicchio G, Houle AM, Vernet O, Jednak R,
OGorman AM et al: Magnetic resonance imaging in children with voiding dysfunction: is it indicated? J Urol 1998;
160: 1080.
4. Arikan N, Soygur T, Selcuki M, Erden I and Gogus O: Role of
magnetic resonance imaging in children with voiding dysfunction: retrospective analysis of 81 patients. Urology
1999; 54: 157.
5. Wraige E and Borzyskowski M: Investigation of daytime wetting: when is spinal cord imaging indicated? Arch Dis Child
2002; 87: 151.
6. Mak V and Radomski SB: Value of magnetic resonance imaging of the lumbosacral spinal cord in patients with voiding
dysfunction. J Urol 1996; 156: 1421.
7. Hellstrom WJ, Edwards MS and Kogan BA: Urological aspects
of the tethered cord syndrome. J Urol 1986; 135: 317.
8. Yoneyama T, Fukui J, Ohtsuka K, Komatsu H and Ogawa A:
Urinary tract dysfunctions in tethered spinal cord syndrome: improvement after surgical untethering. J Urol
1985; 133: 999.
9. Khoury AE, Hendrick EB, McLorie GA, Kulkarni A and
Churchill BM: Occult spinal dysraphism: clinical and urodynamic outcome after division of the filum terminale.
J Urol 1990; 144: 426.
10. Zoller G, Schoner W and Ringert RH: Pre- and postoperative
urodynamic findings in children with tethered spinal cord
syndrome. Eur Urol 1991; 19: 139.
11. Guggisberg D, Hadj-Rabia S, Viney C, Bodemer C, Brunelle F,
Zerah M et al: Skin markers of occult spinal dysraphism in
children: a review of 54 cases. Arch Dermatol 2004; 140:
1109.
EDITORIAL COMMENT
This study addresses an important issue in the evaluation
of voiding dysfunction, namely the need for selectivity
when recommending spinal MRI. These authors conclude
that of the factors evaluated only abnormal lumbosacral
cutaneous findings were predictive of MRI abnormality,
reminding us of the value of a carefully performed physical examination in this population. The authors appeared
to observe a significant predictive value of plain spinal
x-rays, which were positive in 67% of patients with abnor-

mal MRI vs 28% with a normal study. Thorough urodynamics, which we would perform before recommending
MRI, was done in a fraction of these patients.
The conclusions must be accepted in the context of the
study limitations, as acknowledged by the authors. This is a
retrospective review of a small number of patients subject to
selection bias by the authors and, therefore, without a pure
1751
control group. The author recommendation of a prospective

multicenter study should be pursued.
David A. Diamond
Department of Urology (Surgery)
Harvard Medical School
Development of an Objective Score

to Quantify the Pediatric Cystometrogram
A. L. MacNeily,* M. P. Leonard, P. D. Metcalfe, A. Casale, L. Guerra, P. Steinbok and H. Garton
From the Divisions of Pediatric Urology (ALM) and Pediatric Neurosurgery (PS), University of British Columbia, Vancouver, British
Columbia, and Division of Pediatric Urology, University of Ottawa (MPL, LG), Ottawa, Ontario, Canada, Division of Pediatric Urology,
Indiana University (PDM, AC), Bloomington, Indiana, and Division of Pediatric Neurosurgery, University of Michigan (HG), Ann Arbor,
Michigan
Purpose: We developed a reliable tool for quantitative assessment of the pediatric cystometrogram.
Materials and Methods: Scores for expected capacity (EV), compliance (EV20), activity and sensation were developed
according to established formulas for norms. Ordinal scores were derived by calculating observed over expected findings.
Based on the derived percents scores of 1 to 5 were assigned. For EV 0 to 5 the formula used was EV (age 2) 30 and
for EV20 0 to 5 the formula used was EV20 17 age 55. Activity was determined as the volume of the first, total number
and magnitude of involuntary contractions, each scored 0 to 5 and divided by 3. Sensation was scored as 0 to 3 according to
volume at first sensation. A total of 87 blinded cystometrograms in 49 patients were independently scored twice by 3 pediatric
urologists. The resultant 522 total and 2,088 component scores were assessed for reliability.
Results: Intrarater reliability was strong with 80% of total scores (208 of 261) within 1 point from initial to subsequent retest.
Reliability component scores were stronger with 94% (983 of 1,044) within 1 point from test to retest. Spearmans rank
correlations for total score was 0.82 to 0.98, indicating a strong relationship between test and retest. Interrater reliability of
components was strong with 89% of scores (1,851 of 2,088) between urologists within 1 point. Correlation coefficients for
total scores were 0.80 to 0.96, indicating a high degree of consistency between urologist assessments (p 0.05).
Conclusions: This score appears to reliably quantify the pediatric cystometrogram. Its application may be useful for the
objective assessment of detrusor behavior before and after intervention. Further applications should allow refinement of this
tool.
Key Words: bladder, urodynamics, reproducibility of results
lthough there are guidelines pertaining to what represents normal bladder capacity, compliance, contractility and sensation, there does not exist 1 unifying
and practical tool that allows an objective assessment of
overall bladder behavior. Several investigators have quantified individual components of the bladder filling/storage
and voiding phases but none have developed a tool that uses
multiple parameter evaluation with proven reliability.1 4 As
such, UDS remains an interpretive art. We built on this
previous study by applying ordinal scales to 4 components of
the pediatric CMG, including capacity, compliance, the timing/number/magnitude of involuntary contractions and sensation. Summing these individual scores would create a total
UDS score to provide a global assessment of the storage/
filling behavior of the bladder that could be evaluated for
reliability.

Based primarily on the previous study of Meyrat et al4 we
agreed that the 4 detrusor parameters of capacity, compli-
Study received institutional ethics review board approval.

* Correspondence: Department of Pediatric Urology, K0-134 BC
Childrens Hospital, 4480 Oak St., Vancouver, British Columbia,
Canada V6H 3V4 (e-mail: amacneily@cw.bc.ca).
0022-5347/07/1784-1752/0
ance, activity and sensation were important domains to

quantify (see Appendix).
EV
We adopted the most widely used formula for expected age adjusted bladder capacity in cc, that is EV (age in years 2)
30.5,6 Urodynamic findings were scored as 0 to 5 according to
the percent of EV achieved.
EV20
We applied the pressure-volume relationship previously described by Houle et al.3 Briefly, they reported that 95% of
normal children 1.25 to 16.6 years old can store 95% or greater
of bladder capacity at below 20 cm H2O pressure.3 When normalized for age, the minimal acceptable volume stored below
20 cm H2O (EV20) was characterized by the linear equation,
EV20 17 (age in years) 55 cc. CMGs were scored as 0 to
5 according to the percent of EV20 achieved.
Detrusor Activity
We believed that it was important to measure 3 subcomponents of involuntary detrusor overactivity during filling,
including the timing (volume) of the first involuntary contraction, and the total number and magnitude of the largest
contraction. These 3 aspects of detrusor overactivity have
been shown to be worthy of measurement.7,8 An involuntary
1752
Vol. 178, 1752-1757, October 2007

Printed in U.S.A.
DOI:10.1016/j.juro.2007.03.157
OBJECTIVE SCORE TO QUANTIFY PEDIATRIC CYSTOMETROGRAM

contraction was defined as an increase of more than 5 cm H2O
above baseline detrusor pressure.8 Scores of 0 to 5 were assigned for each subcomponent, summed and then divided by 3.
Sensation
Scoring from 0 to 3 was based on volume infused at the time
that patients recorded the first sensation of bladder filling as
a percent of EV. If first sensation was not indicated on the
tracing, the first evidence of detrusor activity was used as a
surrogate.
Total Score
The total score was a summation of component subscores. It
could theoretically be a minimum of 0 in patients with a
small noncompliant bladder demonstrating early, frequent
and high amplitude involuntary detrusor contractions and
early or absent bladder sensation. A maximum score of 18
could be achieved by patients with normal parameters in all
of these domains. Noninteger scores were possible because
detrusor activity was scored by averaging its 3 related subcomponents (see Appendix).
Study Material
Institutional ethics review board approval was obtained.
Nurse clinicians working with 3 pediatric urologists (AEM,
MPL and PDM), selected CMG tracings for scoring from
their institutional urodynamic databases. These nonvideo
studies had been performed at each institution by respective
urology nurse clinicians. The urologists did not actually
participate in the technical aspect of the studies. All studies
were done using institutional equipment with the patient
supine using a fill rate 10% or less of expected capacity for
age per minute. Electronic commentaries were routinely
made by the nurses regarding events such as first sensation
and leakage but not for involuntary detrusor activity. A
unique alphanumeric code was assigned to each tracing.
Diagnoses and all other patient identifiers except age in
years were removed before scoring.
A total of 87 CMGs in 49 patients were analyzed. Median
subject age was 9 years (range 3 to 18). We noted underlying
patient diagnoses. A total of 30 patients underwent UDS
before and following section of the filum terminale for refractory dysfunctional voiding and suspected tethered cord
syndrome (17) or radiographically and clinically proven tethered cord syndrome (13), representing 63 of the 87 UDS
tracings studied (72%). The remaining 19 subjects underwent a total of 24 studies for certain indications, including
spina bifida in 7, suspected tethered cord in 6, other neurological conditions in 3, lipomeningocele in 1, traumatic spinal cord injury in 1 and refractory dysfunctional voiding
with no discernible anatomical or neurogenic cause in 1.
CMGs were scored independently by all 3 urologists on 2
occasions a minimum of 2 to 3 weeks apart. The resultant
522 total scores and 2,088 component scores were entered
into a database for subsequent statistical analysis.
For purposes of analysis UDSs and their scores were divided
into 3 subgroups, including pre-interventiona total of 30
studies in 30 patients before filum section, post-interventiona total of 33 studies in the same group of 30 patients
3 to 12 months after filum section and regularthe total of
1753
24 remaining studies in 19 patients who did not undergo

intervention.
Intrarater Reliability
Intrarater reliability measures the consistency of a rating
system across multiple replications by 1 individual. In the
ideal case scores on an initial test and a subsequent retest
would be identical. This was first assessed by determining
the percent of identical scores and percent of scores within
1 point from initial test to subsequent retest. Intrarater
reliability was assessed across each combination of component and total scores, patient group and physician. Overall
this resulted in 45 subgroups being tested, that is 5 tool
components 3 patient groups 3 urologists, or 9 subgroups for each component. Spearmans rank correlations
and tests of significance were used as additional measures of
intrarater reliability. Spearmans rank correlations allows
each variable to be measured on an ordinal scale and it does
not assume that variables will be distributed normally. It is
measured from 1 to 1 with 1 indicating perfect correlation,
1 indicating perfect negative correlation and 0 indicating
no relationship between 2 variables. The t test was performed on correlation coefficients to determine significance
at a confidence level of 95%.
Interrater Reliability
Interrater reliability measures the consistency of a rating system by comparing assessment scores from numerous sources.
In the ideal case scores derived by different urologists would be
identical. Interrater reliability was assessed by looking at the
percent of identical scores and the percent of responses within
1 point from the initial test to the subsequent retest. It was
assessed across each combination of component and total score,
patient group, urologist pair and replication. This resulted in a
total of 90 subgroups being tested, that is 5 tool components
3 patient groups 3 urologist pairs 2 replications, or 18
subgroups per component. Spearmans rank correlations and
tests of significance were used as additional measures of interrater reliability.
Comparison of UDS
Score to Clinical Outcomes
We tested model usefulness as a clinical indicator by comparing the change in total UDS score to the clinical impression by a treating urologists of overall urological improvement following neurosurgical section of the filum terminale
for dysfunctional voiding, ie groups 1 and 2 of 30 patients
and a total of 63 scores. Clinical improvement was graded
0 no improvement, 1slightly improved, 2mostly improved or 3 completely cured based on verbal questioning
of the patients and families. The treating urologists providing a clinical impression (AC and LG) were not the same
individuals scoring the UDS tracings and they were blinded
to the results of UDS scoring.
RESULTS
Intrarater Reliability
Overall intrarater reliability appears quite strong with 80%
of total scores (208 of 261) within 1 point from initial test
to subsequent retest. The reliability of individual component
scores was even stronger with 94% (983 of 1,044) within 1
1754
point from test to retest. All 4 tool components showed

strong intrarater reliability with a percent of responses
within 1 point for capacity of 99% (259 of 261 scores),
compliance of 95% (248 of 261), detrusor activity of 89% (232
of 261) and sensation of 94% (244 of 261). Spearmans rank
correlations for total UDS score was 0.82 to 0.98, indicating
a strong relationship from initial test to subsequent retest.
Significance testing on correlation values revealed that only
1 of the 45 tests was nonsignificant. More detailed results at
the component, physician and patient group levels were also
calculated (tables 1 and 2).
Interrater Reliability
Comparisons at the component level were strong with 89% of
individual component scores (1,851 of 2,088) between different
urologists within 1 point. Each individual tool component
showed strong interrater reliability. The capacity component
appeared most reliable with 98% of physician pair comparisons
(514 of 522) within 1 point. Sensation and compliance followed closely with 91% (476 of 522) and 90% of pair comparisons (470 of 522), respectively, within 1 point. Detrusor
activity had the lowest reliability with 75% of comparisons (391
of 522) within 1 point. Overall the tool showed fairly strong
interrater reliability with 60% of aggregate score comparisons
(315 of 522) between urologists within 1 point. Spearmans
rank correlation coefficients for total UDS score was 0.80 to
0.96, again indicating a high degree of consistency between
urologist assessments. Significance testing supported these
positive results. The correlation coefficients of all 90 subgroups
tested were significant. More detailed results at the component, physician pair, patient group and replication levels were
also calculated (see figure and table 3).
Comparison of UDS Score
to Clinical Outcomes Following Filum Section
Median change in total UDS score for the group of 30 patients was 2.75 (95% CI 5.359.3) with an average clinical
impression of improvement of 1.4 (range 0 to 3). Analysis did
not identify significant correlations between changes in UDS
scores and clinical assessments in any of the 6 subgroups,
that is 3 urologists scoring on 2 occasion each.
TABLE 2. Interrater reliability

Spearmans Correlation Coefficients
Test Type
Urologist 1:
Regular
Pre-intervention
Post-intervention
Urologist 2:
Regular
Pre-intervention
Post-intervention
Urologist 3:
Regular
Pre-intervention
Post-intervention
EV
EV20
Activity
Sensation
Total
Score
0.99
0.99
0.92
0.87
0.94
0.99
0.92
0.81
0.97
0.76
0.95
0.88
0.90
0.92
0.96
1.00
1.00
0.98
0.96
0.98
0.67
0.90
0.88
0.97
0.82
0.64
0.91
0.98
0.96
0.93
0.99
0.91
1.00
0.79
0.85
0.84
0.68
0.91
0.91
0.32
0.87
0.97
0.82
0.92
0.95
Urologist 3 regular sensation p 0.12 and all other comparisons p 0.05.
made similar attempts but without proven reliability. For

example, in 1991 Galloway et al described their attempts to
quantify videourodynamics in 171 patients with myelodysplasia.2 They applied a score of 0 to 2 for the 5 variables
outlet resistance, sphincter behavior, bladder compliance,
detrusor contractility and vesicoureteral reflux. The resulting hostility score was 0 favorable to 10 unfavorable. No
assessments of bladder volume or sensation were included in
their score. Only the 3 components reflux, leak pressure and
compliance correlated significantly with upper tract changes
on pyelography or ultrasound. Patients with a hostility score
of 5 or greater were at risk for upper tract deterioration. A
subsequent retrospective review of 14 patients suggested
that this score was responsive to therapeutic bladder interventions.9 Unfortunately neither the reliability nor the validity of the score was assessed. In addition, there was no
heterogeneity in the population of patients to whom the
score was applied. All patients had spina bifida. Our score
was designed to apply to a broad variety of patients regardless of underlying diagnosis.
DISCUSSION
We applied ordinal scores to the pediatric filling CMG. The
tool that we propose is simple, objective and reliable. Others
TABLE 1. Interrater reliability

% Test/Retest Comparisons Within 1 Point
Test Type
Urologist 1:
Regular
Pre-intervention
Post-intervention
Urologist 2:
Regular
Pre-intervention
Post-intervention
Urologist 3:
Regular
Pre-intervention
Post-intervention
EV
EV20
Activity
Sensation
Total
Score
100
97
100
92
93
100
92
90
94
83
100
97
75
80
79
100
100
100
100
97
94
88
100
97
92
90
97
83
87
88
100
97
100
88
93
97
75
77
85
79
97
100
75
67
82
Interrater reliability of test components showing percent of test/

retest comparisons within 1 point.

TABLE 3. Test 2 interrater reliability
Spearmans Correlation Coefficients
(urologists)
Component (type)
Expected bladder vol:
Pre-intervention
Post-intervention
Regular
Bladder compliance:
Pre-intervention
Post-intervention
Regular
Detrusor activity:
Pre-intervention
Post-intervention
Regular
Bladder sensation:
Pre-intervention
Post-intervention
Regular
Totals:
Pre-intervention
Post-intervention
Regular
1 vs 2
2 vs 3
1 vs 3
1.00
0.98
1.00
0.95
0.95
0.99
0.95
0.92
0.99
0.98
0.71
0.77
0.64
0.70
0.84
0.64
0.72
0.74
0.86
0.95
0.83
0.89
0.69
0.74
0.83
0.76
0.79
0.65
0.97
0.62
0.57
0.82
0.41
0.66
0.78
0.70
0.96
0.94
0.91
0.89
0.86
0.82
0.90
0.92
0.80
p 0.05.
More recently Meyrat et al reported a preliminary assessment of a urodynamic scoring system in 15 patients undergoing spinal cord untethering for closed spinal dysraphism.4
They combined data on 3 components of the filling CMG
(volume, compliance and detrusor activity) and 1 from voiding studies (vesicosphincteric synergy) to derive a total score
of 0 to 17. They noted a significant difference between the
preoperative scores of their subjects in comparison to those
of a control group with primary and secondary enuresis,
suggesting that the parameters measured had discriminative value. They also noted improved UDS scores for up to 60
months following untethering. Interestingly their assessments at 12 months following untethering demonstrated an
average improvement in score of 2 of 17, similar to our
findings of a median improvement of 2.75 of 18 between 3
and 12 months after untethering.
Our proposed tool differs from that of Meyrat et al4 in
several ways. We chose not to include a measure of voiding
synergy in our score. In our experience it is difficult to assess
this in children. Many children voluntarily contract the perineal musculature when voiding with a catheter in place or
under stressful situations, such as during UDS. In comparison to our proposed tool their assessment of detrusor activity was somewhat qualitative and no measure of bladder
sensation was included. To our knowledge no formal assessment of the validity or reliability of their instrument has
been published to date.
There are potential pitfalls with the application of our
tool. As it currently exists, the tool does not detect abnormalities manifesting as exaggerated capacity. An abnormally large bladder, even with impaired sensation, could
potentially score a total of 15 of 18. A possible solution to this
would be to assign a decreasing score for EV after an individual exceeded expected capacity for age, eg a score of 4 for
120% to 140% of EV, a score of 3 for 140% to 160% of EV, etc.
This would not change the maximum attainable score of 18
but it would require more study to assess reliability and
validity subsequent to modifying our proposed tool. Con-
1755
versely patients with a small capacity bladder but normal

filling pressure could potentially score poorly for compliance
because they fail to meet the volume below 20 cm H2O
pressure expected for age, although storage pressures never
increase to dangerous levels. Although all components were
reliable, one must view individual component scores to judge
how they contribute to the overall score to assess which
detrusor parameter is most influential for deriving the score
in a given patient. Finally, the formulas for normal values
that we used are based on the assumption of a normal body
habitus, which clearly is not always the case in some of our
patient population, eg those with spina bifida.
We were unable to observe any significant correlation
between clinical impressions of urological change after section of the filum terminale and changes in UDS scores.
There are several possible interpretations for this finding.
For example, it is possible that the ability of the treating
urologists to objectively assess clinical change was flawed.
The fact that voiding function was not assessed clinically
with a validated tool before and following detethering casts
doubt on the accuracy of their clinical assessments. In addition, a form of reporting bias, in which patients (parents)
exaggerate the improvement achieved by an intervention to
please the physician, is a well described phenomenon.10 The
latter may have influenced the clinical assessment of outcomes in comparison to our objective measures. It is also
possible that our tool as it currently exists is not sensitive
enough to detect subtle clinical changes that patients, parents and urologists may perceive. Alternatively we may require more experience and a greater number of patients to
adequately assess the responsiveness of this tool to interventions.
CONCLUSIONS
This study is an initial attempt to quantify pediatric CMG
by combining 4 commonly measured components of bladder
behavior into a total score. The tool and its components
appear reliable when applied to various patient conditions.
Future prospective applications may allow further refinement as well as assessment of its validity and responsiveness to interventions.
ACKNOWLEDGMENTS
Statistical analysis was performed at AnalysisWorks, Vancouver, British Columbia, Canada. Institutional equipment
was obtained from Laborie Medical Technologies.
APPENDIX
UDS Score Components
Component
Definitions
Scores Attainable
Bladder capacity
(EV)
EV 30 (years old
2) ml
0 0% to less than 20% EV

1 20% to 39% EV
2 40% to 59% EV
3 60% to 79% EV
4 80% to 99% EV
5 100% EV or greater
(appendix continued)
1756
APPENDIX continued
UDS Score Components
Component
Definitions
Scores Attainable
Bladder compliance
(EV20)
Expected volume less than

20 cm H2O EV20 17 (years
old) 55 ml
0
1
2
3
4
5
Detrusor Activity (A)
Volume, Number and Magnitude of

Involuntary Detrusor Contractions
Vol at First Contraction as

% of EV
0
1
2
3
4
5
Bladder sensation (S)
Bladder volume at first sensation
0
1
2
3
3
2
1
0
0% to Less than 20% EV20

20% to 39% EV20
40% to 59% EV20
60% to 79% EV20
80% to 99% EV20
100% EV20 or greater
Total Number of
Contractions
Magnitude of Greatest
Contraction
Less than 25% EV

0 5 Contractions or greater
0 60 cm H2O or greater
25% to 49% EV
14
1 45% to 60 cm H2O
50% to 74% EV
23
2 30% to 44 cm H2O
75% to 100% EV
32
3 15% to 29 cm H2O
End fill
41
4 Less than 15 cm H2O
None
5 None
5 None
Detrusor activity score sum of 3 categories divided by 3
Less than 10% EV
10% to 19% EV
20% to 29% EV
30% to 49% EV
50% to 75% EV
76% to 90% EV
91% to 100% EV
100% EV or greater
Total score is sum of each of the 4 component scores

CMG
EV
EV20
UDS
cystometrogram
expected volume
compliance
urodynamic study
REFERENCES
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
Norgaard JP, Van Gool JD, Hjalmas Djurhuus JC and

Hellstrom AL: Standardization and definitions in lower
urinary tract dysfunction in children. Br J Urol, suppl.,
1998; 81: 1.
Galloway NT, Mekras JA, Helms M and Webster GD: An
objective score to predict upper tract deterioration in myelodysplasia. J Urol 1991; 145: 535.
Houle AM, Gilmour RF, Churchill BM, Gaumond M and
Bissonnette B: What volume can a child normally store in
the bladder at a safe pressure? J Urol 1993; 149: 561.
Meyrat BJ, Tercier S, Lutz N, Rilliet B, Forcada-Guex M and
Vernet O: Introduction of a urodynamic score to detect pre
and postoperative neurological deficits in children with a
primary tethered cord. Childs Nerv Syst 2003; 19: 716.
Koff SA: Estimating bladder capacity in children. Urology
1983; 21: 248.
Berger RM, Maizels M, Moran GC, Conway JJ and Firlit CF:
Bladder Capacity (Ounces) equals age (years) plus 2 predicts normal bladder capacity and aids in diagnosis of abnormal voiding habits. J Urol 1983; 129: 347.
Brading AF and Turner WH: The unstable bladder: towards a
common mechanism. Br J Urol 1994; 73: 3.
Abrams P, Cardozo L, Fall M, Griffiths D, Rosier P, Ulmsten U
et al: The standardization of terminology in lower urinary
tract function. Neurourol Urodyn 2002; 21: 167.
Perez LM, Khoury J and Webster GD: The value of urodynamic
studies in infants less than 1 year old with congenital
spinal dysraphism. J Urol 1992; 148: 584.
Last JM: A Dictionary of Epidemiology, 4th ed. Oxford: Oxford
University Press 2001; p 157.
EDITORIAL COMMENTS
The pediatric CMG has a central role in diagnostic testing
for neurogenic bladder dysfunction. Interpreting UDS
findings often results in proactive pharmacological intervention and major reconstructive surgery. There is no
other pediatric diagnostic study that carries as much therapeutic responsibility while lacking its own objective assessment. These authors recognize the limitations of interpreting the pediatric CMD and they propose an
objective method to quantify the data. While this may not
be the final version, it shows promise in intrarater and
interrater reliability, which is a requirement for any
multi-institutional protocol. The clinical significance of
this methodology is seen in their evaluation of outcome
following transection of the filum, comparing objective
CMG data to the subjective interpretation obtained from
patient and family interviews. The lack of correlation
between the 2 methods may indicate limitations with this
methodology but I suspect that it supports the faults of
measuring success of intervention based on weak subjective interview data.
Objective assessment of CMG is an important step
forward but we must recognize that any analysis is only as
good as the data obtained. Pediatric CMG data are influenced and manipulated by several technical factors encountered when performing the study, limiting the ability
to compare institutional results. The International Childrens Continence Society recognizes the importance of
standardizing the terminology of pediatric lower urinary
tract function, allowing us to communicate on an international level.1 The same standardization of pediatric UDS
must occur with the development of an international consensus regarding the technique used to generate and interpret a reliable, reproducible, objective study. I look
forward to further refinements in the methodology of
these authors. Embracing an objective score to quantify

the pediatric CMG can help us achieve sound, evidence
based outcome data.
David B. Joseph
Pediatric Urology
University of Alabama at
Birmingham Childrens Hospital
Birmingham, Alabama
1.
of paper and has computer programs that offer assistance

with interpretation. Despite early promise computer assisted electrocardiogram interpretation has not yet attained
the level at which it can be a substitute for expert interpretation.2 4 The method presented likewise requires further
revision in a larger and less heterogeneous group of patients
before it can replace the current method of interpretation
based on experience and expertise.
Nevus T, von Gontard A, Hoebeke P, Hjlms K, Bauer S,

Bower W et al: The standardization of terminology of lower
urinary tract function in children and adolescents: report
from the standardization committee of the International
Childrens Continence Society. J Urol 2006; 176: 314.
EDITORIAL COMMENTS
A reproducible system for interpreting UDS would be useful.
The reproducibility of this system is not surprising, given its
methodology, which centers on EV. A major concern is that
all scoring criteria depend on EV. Patients with neurogenic
bladder often have an abnormal body. Formulas that assume a normal body habitus may yield a misleading estimate of capacity (references 3 and 5 in article).1 Many urologists would like UDS to be like the electrocardiogram,
which seems to offer objective information on a single sheet
1757
Julian Wan
University of Michigan
Ann Arbor, Michigan
1.
2.
3.
4.
Fairhurst JJ, Rubin CM, Hyde I, Freeman NV and Williams JD:

Bladder capacity in infants. J Ped Surg 1991; 26: 55.
Hurst JW: What do good doctors try to do? Arch Intern Med
2003; 163: 2681.
Salerno SM, Alguire PC and Waxman HS: Competency in interpretation of 12-lead electrocardiograms: a summary and
appraisal of published evidence. Ann Intern Med 2003; 138:
751, 2003.
Heden B, Ohlin H, Rittner R and Edenbrandt L: Acute myocardial infarction detected in the 12-lead ECG by artificial neural networks. Circulation 1997; 96: 1798.
Office Management of Pediatric Primary

Nocturnal Enuresis: A Comparison of Physician
Advised and Parent Chosen Alternative Treatment Outcomes
Dawn Diaz Saldano,* Antonio H. Chaviano, Max Maizels, Elizabeth B. Yerkes, Earl Y. Cheng,
Jennifer Losavio, Sima P. Porten, Christine Sullivan, Kerry F. Zebold, Jennifer Hagerty
and William E. Kaplan
From the Childrens Memorial Hospital, Chicago, Illinois
Purpose: We compared the remission of pediatric primary nocturnal enuresis in groups of children who used a physician
advised practice plan vs a parent chosen alternative.
Materials and Methods: Between January 2004 and January 2006 there were 119 patients with primary nocturnal
enuresis enrolled in this prospective, nonrandomized study. For this study primary nocturnal enuresis was defined as wetting
at night during sleep during any 6-month interval without any known causative problem. A total of 76 children received the
physician advised treatment plan and used an alarm, oxybutynin, desmopressin, an elimination diet and a bowel program,
as indicated. A total of 43 children received a parent chosen alternative treatment plan, which consisted of any single or
combination of treatments involving an alarm, oxybutynin, desmopressin and an elimination diet or bowel program. Parents
from each group completed an intake survey that measured functional bladder capacity using a 3-day home diary and they
identified demographic variables. Followup occurred at 2 weeks and then monthly for 12 weeks to study end.
Results: We found that the probability of remission by the end of the study for the physician advised treatment group was
significantly higher than that of the parent choice group (88% vs 29%, Kaplan-Meier curve p 0.0001).
Conclusions: The group of children who followed physician advised treatment for primary nocturnal enuresis showed
significantly earlier remission of primary nocturnal enuresis than children who followed the parent choice treatment (25th
percentile 2 vs 10 weeks).
Key Words: bladder, nocturnal enuresis, deamino arginine vasopressin, parents, questionnaires
hile PNE is a common pediatric urological diagnosis, surprisingly its treatment has not yet become
standardized. Currently therapy usually involves
single modal treatment, such as an enuresis alarm for sleep
arousal failure, desmopressin for nocturnal polyuria or oxybutynin for small bladder size. Such single treatment plans
have low remission rates, and so they are fraught with
dissatisfaction, leading to treatment abandonment. While
the moisture alarm treatment yields the highest remission
rate of 70%, the associated high 40% rate of dropout and
noncompliance1 restricts wide use of this treatment method.
Medication such as desmopressin may provide remission
sooner than alarm treatment but relapse is frequent and
immediate after the medication is discontinued.2 Treatment
combining the enuresis alarm and pharmacotherapy shows
a higher number of dry nights at the end of a treatment
period compared to the alarm alone.3 This combined approach may produce earlier positive results, which may increase compliance with treatment.4
Our treatment plan advises primary use of an enuresis
alarm along with supplemental treatments, such as oxybu-
Study received Childrens Memorial Hospital institutional review

board approval.
* Correspondence: Childrens Memorial Hospital, 2300 Childrens
Plaza, Chicago, Illinois 60614 (telephone: 773-880-4959; FAX: 773880-3339; e-mail: dsaldano@childrensmemorial.org).
0022-5347/07/1784-1758/0
tynin for small functional bladder capacity or a bowel program for constipation. The personal attitudes, home affairs
and/or social circumstances of parents may cause them to
resist accepting the physician advised treatment plan. Thus,
parents may make alternative treatment choices to meet
their needs. We believe that office management of PNE
could be improved if we could inform parents of the outcomes
of the 2 treatment plans. In this way parents could choose
their treatment plan and yet be forewarned regarding differences in the probability of dryness acquisition. We compared the remission results of PNE treatment in groups of
children who used a physician advised practice plan vs a
parent chosen alternative.
Study Design
In this prospective, nonrandomized study we compared the
effectiveness of treatment for PNE using a physician advised
treatment plan based on medical evaluation vs a parent
chosen alternative treatment plan based on parent needs.
All families using the enuresis alarm received personalized
instruction on its use by our nurse practitioner. Parents
completed an intake survey that identified the status of
certain demographic variables, including parent education
level, family history of bedwetting, with whom the child
lives, the number of siblings, motivation of the child, mother
1758
Vol. 178, 1758-1762, October 2007

Printed in U.S.A.
DOI:10.1016/j.juro.2007.03.195
OFFICE MANAGEMENT OF PEDIATRIC PRIMARY NOCTURNAL ENURESIS
1759
and father, deep sleep, snoring, nightmares, teeth grinding,

sleepwalking, attention deficit disorder with/without hyperactivity, allergies, physical and/or emotional developmental
delays, and asthma. All families completed a 3-day bladder
diary measuring urine output. The largest single output
during this interval was considered functional bladder capacity.
Inclusion Criteria
From January 2004 to January 2006 all children referred to
our division of urology for PNE management were prospectively enrolled in the study. In this study PNE was defined
as wetting at night during sleep during any 6-month interval without any known causative problem. Bedwetting was
defined as greater than 2 nights wet per week. Remission
was defined as dry for 14 consecutive nights. Relapse was
defined as bedwetting occurring twice weekly after being dry
for 6 months. Cure was defined as dry for 1 year or greater.
Exclusion Criteria
Children with nighttime wetting and coexisting anatomical
urological problems (vesicoureteral reflux or posterior urethral valves), dysfunctional elimination syndrome or urinary tract infection within a year before evaluation were
excluded. Children with daytime wetting were also excluded.
Medical Evaluation
At the initial visit a history was obtained and physical
examination was performed. Radiological studies were not
done in this group of children other than the occasional plain
x-ray of the kidneys, ureters and bladder to assess for occult
constipation. Flow rates were not determined routinely.
Treatment for PNE
Physician treatment plan(s) were advised based on clinical
evaluation of the child. Treatments were offered according to
the plan developed by our program, Try for Dry.5 This is a
multimodal plan which, as an algorithm, includes a moisture alarm to address sleep arousal failure, age appropriate
inducements to reward dryness, an elimination diet to address possible underlying food sensitivities (avoiding beverages that are carbonated or contain artificial colors, or citric,
heavily sugared or dairy products for 2 weeks), oxybutynin
to address small functional bladder capacity using a 3 times
daily dose when functional bladder capacity is decreased
according to the home diary, oxybutynin at a nightly dose
based on our empirical clinical experience (although functional bladder capacity is normal), desmopressin prescribed
at a low dose of 0.1 mg at bedtime for children 8 to 13 years
old and at 0.2 mg at bedtime for children older than 13 years,
and lastly a bowel program consisting of prescribed
Senekot or MiraLax if there was constipation (fig. 1).
Parent Chosen Alternative Treatment Plan(s)
Parents who rejected the physician advised treatment plan,
eg did not want pharmacotherapy and did not want to use
the moisture alarm, were permitted to choose a treatment
plan that best fit their needs. These parent chosen plans
included the personalized choice of single or combined use of
a moisture alarm with age appropriate inducements, oxybu-
FIG. 1. PNE treatment algorithm. AM, 8:00 a.m. PM, 4:00 p.m. HS,
before bedtime.
tynin/desmopressin according to the presented dose scheme

(fig. 1), an elimination diet and/or a bowel program.
Followup of Treatment
Followup of PNE treatment for all patients was done at
2-week intervals for month 1 and then monthly until the
child acquired 14 consecutive dry nights or up to a total of 12
weeks of treatment. PNE treatment was gradually tapered
and then discontinued after the child achieved 14 consecutive dry nights or at the end of 12 weeks of treatment (study
end). First alarm use was tapered by 1 night weekly for 7
weeks. After alarm use was discontinued medication use
was similarly tapered.
This study was approved by the Childrens Memorial
Hospital institutional review board. Statistical analysis was
performed using SAS, version 9.1.
RESULTS
There were 119 children, including 85 males and 34 females,
enrolled consecutively in the study. The age range was 5
years to 19 years (mean SD 10 3, median 9). The
frequency of dry nights for 2 weeks before treatment was 0
to 12 (mean 1.9 3.4, median 0). PNE was treated using
physician advised therapy in 76 children and parental chosen therapy in 43. The table shows treatment components in
these groups.
The 2 groups did not show significant differences in demographic composition for any variables examined. Demographic information was available on 30% of the patients.
There was a trend toward a difference in the maternal level
of education, in that the incidence of some college education
in the physician advised group was higher than in the parental choice group (88% vs 60%, trend toward significance
p 0.07).
Figure 2 shows the probability of dryness acquisition by
treatment group. Time to PNE remission using physician
advised treatment was significantly sooner than by parent
chosen therapy (25th percentile 2 vs 10 weeks). At the end of
12 weeks the probability of remission for the physician advised treatment group was significantly higher than for the
parent chosen alternative treatment group (88% vs 29%,
Kaplan-Meier curve p 0.0.0001).
DISCUSSION
Treatment for PNE in children currently lacks an established protocol for physicians and families to follow. There
are many therapeutic options for treatment (an alarm, an
elimination diet, pharmaceuticals, etc) and many times par-
1760

Component treatments chosen for PNE
Treatment Chosen
No. Physician
Advised
No. Parent
Chosen*
Overall
Alarm with inducements
Oxybutynin
Elimination diet
DDAVP
Bowel program
76
76
76
76
41
13
43
40
26
36
25
4
* Any combination of alarm, oxybutynin, diet, DDAVP and/or bowel

program.
ents choose to create a plan instead of following the regimen

outlined by the physician of the child. To date it has not been
proved that this choice of care was any less successful than
specifically adhering to the physician advised plan. However, our results clearly show that children have significantly lower remission success rates when following parent
chosen alternative treatments. This should encourage parents to adhere to the treatments that their physicians create
for them since that is the best chance for remission for the
child. Physicians can use these data to support their treatment plans with the family.
We also noted that families who chose a physician advised treatment plan tended to have at least a college level
of education more often than parents who chose an alternative plan (88% vs 60%). While this difference was not statistically significant, it may be that families with a higher level
of education may comply better with treatments suggested
by physicians than families with a lower level of education.
It is widely accepted that the enuresis alarm alone is the
mainstay treatment for PNE. However, we found that the
physician advised plan of alarm use and supplemental treatments presented provided dryness in 88% of cases by 12
weeks, which is more effective than a single use of the
enuresis alarm, which provides dryness in 63% of cases at 24
weeks.5 We believe that the physician advised plan works
because in combination with supplemental treatments there

is early success with alarm use. This early success in achieving dryness using the physician advised plan is the likely
motivating factor that stimulates parents/families to continue with the treatment program, thereby increasing compliance in the physician treatment group.
We speculated about the mechanism(s) of supplemental
treatments on PNE remission.
Oxybutynin
It is recognized that oxybutynin alone is not effective for PNE.6
It is our impression that oxybutynin, as an anticholinergic and
antispasmodic, decreases the number of times that the alarm
sounds because of a wetting episode. While it was not directly
related to nighttime wetting, oxybutynin also decreased associated daytime frequency. Furthermore, we speculate that oxybutynin extends the interval between the alarm first sounding
to initial, partial bladder emptying as a moist episode and
subsequent, more complete bladder emptying as a soaking
episode. This time extension permits families to take the child
to the toilet before a soaking episode.
Desmopressin
While desmopressin alone is recognized as effective treatment for PNE, the physician advised plan structures a subtherapeutic dose of the drug. We speculate that a subtherapeutic dose of desmopressin decreases polyuria without
causing nighttime dryness, thereby leading to fewer alarm
soundings. Relapse rates of PNE after desmopressin use
alone is about 90%.7 Our ongoing research is addressing the
relapse rates of PNE in our case load.
Elimination Diet
The use of diet to treat PNE is empirical. Our empirical
observations are that diet modification may benefit PNE
treatment. This experience led us to structure an elimina-
FIG. 2. Kaplan-Meier curve comparing probabilities of PNE remission in treatment groups. At end of 12-week treatment physician advised
(solid line) and parent choice (dashed line) treatment groups showed 88% and 29% PNE remission, respectively (p 0.0001).

tion diet, which involved avoiding beverages that are carbonated or contain artificial colors, and citric, heavily sugared or dairy products for 2 weeks as a supplemental
treatment option. We currently do not have a scientific basis
for this bias but we consider the results of controlled studies
on migraine headache triggered by diet and enuresis triggered by diet8 as a scientific basis for this treatment.
Bowel Program
Because it was reported that constipation can cause abnormal
urodynamic patterns, including bladder spasticity,9 we aggressively treat constipation recognized in children with PNE.
We believe that this study is important for establishing a
standardized protocol for pediatric PNE. Although bedwetting spontaneously resolves, it does so at a low rate of about
12% per year.10 Waiting until such time that PNE remits on
its own may cause anxiety, frustration and lower self-esteem
in the child. Therefore, it is important to develop a practical
and effective method to treat PNE. Currently office management for pediatric PNE mostly involves a single modal approach to treat this common disorder. The results of our
study show that using a physician advised, multimodal
treatment plan remits PNE in 88% of children by 12 weeks
of treatment. This experience is important for addressing
the needs of many children whom we see who have been
treated with conventional single modal therapy which, because it did not remit wetting, led to frustration and,
thereby, delayed further management. The results of our
study support the establishment of a physician advised multimodal protocol for children with PNE (fig. 1), such that
bedwetting shows prompt remission. This study provides
guidance to pediatric caregivers, such as pediatricians and
family practitioners, as well as physician extenders, such as
nurse practitioners and physician assistants, for the routine
care of PNE. We are currently examining extended followup
in these groups to compare the long-term remission and cure
of PNE using the 2 treatment plans.
Avoidance of PNE treatment should be considered in
certain circumstances. Because the physician advised plan
is successful when families are compliant with treatment,
one should avoid treating families who are not ready to
commit to the treatments because of transient social impediments. In other words PNE treatment should be delayed until
compliance with treatment is likely. Also, children should be
willing to accept the loud sound of the alarm during sleep.
Children who are fearful of the alarm noise should not be
treated until fearfulness is resolved. Fear is common before age
5 years and it is uncommon after age 7 years. Therefore,
between ages 5 and 7 years the clinician should judge the
child/family acceptance of alarm use. Additionally, bedwetting
that is associated with other wetting/elimination issues, such
as daytime wetting or dysfunctional elimination syndrome,
should undergo pediatric urological evaluation before treatment for wetting. Lastly children with profound constipation
who also show PNE are treated with a bowel program before
other PNE treatments are begun.
The study design was prospective but not randomized.
Since the study was designed to address the clinical circumstance that patients may be resistant to follow physician
directed treatments of alarm use or medication administration, we did not use randomized enrollment. Rather, we
simply allowed families to segregate into the 2 groups based
on their personal views.
1761
From this study we are not able to assign a value to the

effectiveness of individual treatments for remitting PNE.
Future studies may address this need. The results of this
study are confounded by our inability to track oppositional
behavior in children, for example those who claimed to adhere to diet modifications or ingest medications but did not.
This occurrence is likely segregated evenly between the 2
groups. Sleep disorders and airway obstruction were also not
considered in our study. Episodic wetting, ie night dryness
more than 12 weeks in duration, may confound studies on
PNE remission. In this study there were no children with
PNE who were dry more than 1 week. Children with daytime wetting who also had bedwetting were not included in
this series. These children require urological evaluation. In
our experience after evaluation shows normal results these
children attain daytime and nighttime dryness using a comparable physician advised treatment plan.
CONCLUSIONS
The group of children who followed physician advised treatment for PNE showed significantly earlier PNE remission
than children who followed parent choice treatment (25th
percentile at 2 vs 10 weeks). At the end of 12 weeks the
probability of remission for the physician advised group was
88% compared to 29% for the parent choice group.

DDAVP desmopressin
PNE primary nocturnal enuresis
REFERENCES
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
Monda JM and Husmann DA: Primary nocturnal enuresis: a

comparison among observation, imipramine, desmopressin
and bed-wetting alarm systems. J Urol 1995; 154: 745.
Evans JHC and Meadow SR: Desmopressin for bed wetting:
length of treatment, vasopressin secretion and responses.
Arch Dis Child 1992; 67: 184.
Bradbury MG and Meadow SR: Combined treatment with enuresis alarm and desmopressin for nocturnal enuresis. Acta
Paediatr 1995; 84: 1014.
Van Kampen M, Bogaert G, Feys H, Baert L, De Raeymaeker
I and De Weerdt W: High initial efficacy of full-spectrum
therapy for nocturnal enuresis in children and adolescents.
BJU Int 2002; 90: 84.
Maizels M, Rosenbaum D and Keating B: Beginning Treatments and Helper Treatments: Getting to Dry. Boston:
Harvard Common Press 1999; chapts 5 and 6, pp 94 158.
Lovering JS, Tallett SE and McKendry JB: Oxybutynin efficacy in the treatment of primary enuresis. Pediatrics 1988;
82: 104.
Monda J and Husmann DA: Primary nocturnal enuresis: a
comparison among observation, imipramine, desmopressin
acetate and bed-wetting alarm systems. J Urol 1995; 154:
745.
Egger J, Carter CH, Soothill JF and Wilson J: Effect of diet
treatment on enuresis in children with migraine or hyperkinetic behavior. Clin Pediatr 1992; 31: 302.
Neveus T: The evaluation and treatment of therapy-resistant
enuresis: a review. Uppsala J Med Sci 2006; 111: 61.
Forsythe W and Redmond A: Enuresis and spontaneous cure
rate. Arch Dis Child 1974; 49: 259.
1762
EDITORIAL COMMENT
This report is an attempt to compare a physician advised
regimen vs an alternative regimen chosen by parents. The
physician advised regimen is largely based on a combination
of therapies that rely on the empirical clinical experience of
the authors. The multifactorial etiology of enuresis would
justify such a multimodal approach. However, the doses of
DDAVP and oxybutynin were determined based on author
bias and speculation as to their mode of action, rather than
on scientific fact. The 2 medications were prescribed at doses
that are significantly lower than the usual recommended
doses, so that their effect becomes highly speculative. There
are so many regimens advocated for enuresis that it is no
surprise that some parents elected a regimen that reflected
their biases. Nevertheless, the authors achieved an enviable
success rate of 88% in 12 weeks with the physician advised
multimodal plan compared to 29% in the parent choice

group. The incidence of relapse is not stated. Patients were
closely followed for the 12-week treatment period but information on followup after 12 weeks is not available. Briefly,
the authors have put together a therapeutic plan that was
effective in 88% of patients with enuresis in 12 weeks, although the specific reasons for this high success rate are not
quite clear. They now must work out the specifics and tell us
why this plan is effective. Each aspect of their regimen
should be justified based on scientific fact rather than on
empirical observation and clinical bias.
Yves Homsy
Childrens Urology Group
University of South Florida
Tampa/St. Petersburg, Florida
Incontinence Symptom Index-Pediatric:

Development and Initial Validation of a Urinary
Incontinence Instrument for the Older Pediatric Population
Caleb P. Nelson,* John M. Park, David A. Bloom, Julian Wan, Rodney L. Dunn and John T. Wei
From the Departments of Urology, University of Michigan, Ann Arbor, Michigan, and Childrens Hospital Boston (CPN),
Purpose: Although urinary incontinence is common in children, no validated pediatric instruments exist for measuring
urinary incontinence symptoms and bother. We developed and validated a patient reported pediatric survey for urinary
incontinence measurement.
Materials and Methods: The Incontinence Symptom Index-Pediatric is an 11-item instrument comprising 2 domains, that
is 1) impairment and 2) symptom severity, including subdomains for stress urinary incontinence, urge urinary incontinence,
insensate urinary incontinence, nocturnal urinary incontinence and pad use. The survey was self-administered twice, 2 weeks
apart, to boys and girls ages 11 to 17 years. Children completed the survey independently. Cases consisted of patients
presenting to pediatric urology clinic with the chief complaint of urinary incontinence. Controls consisted of healthy children
presenting for evaluation up at a general pediatric practice. Formal validation analysis was performed.
Results: A total of 19 subjects per arm completed at least 1 survey. Internal consistency was good with a Cronbachs of 0.84
for the complete instrument. Item-scale correlations were greater than 0.60 for all except 1 item. Test-retest reliability was
also good (r 0.97, p 0.0001). Discriminative validity was good with a total severity scale score of 9.3 in wet children and
0.7 in controls (p 0.0001). Impairment scale scores differed by 2.2 points (p 0.0001). Mean scores differed significantly
between subscales for all domains except pad use. The most dramatic difference was in the urge urinary incontinence domain,
which differed by a mean of 3.6 points (p 0.0002).
Conclusions: This pilot study provides initial validation of a survey instrument for urinary incontinence in children and
adolescents. This instrument can be used in children ages 11 to 17 years to objectively and reproducibly measure patient
reported urinary incontinence.
Key Words: bladder, urinary incontinence, questionnaires, pediatrics, quality of life
rinary incontinence in the pediatric population is a

significant problem, particularly in patients seen by
pediatric urologists. UI may be relatively common
even among otherwise normal children1,2 and in most children incontinence has a substantial social and psychological
impact.3 Clearly in children with congenital conditions associated with UI the impact can be devastating.
Unfortunately our ability to quantify UI symptoms and
bother in children is limited at baseline and in response to
treatment. Historically outcomes assessments have consisted of physician reported results, physician mediated
patient reports or brief patient reported summaries. Quantification of incontinence has commonly been vaguely descriptive, eg mild or severe leak, or via approximate numerical scores, such as dry intervals. Such scales are simple to
use and administer but they have limited validity (severe
leak may mean different things to different patients) and
they do not adequately assess for bother with similar rankings from different patients having widely disparate impacts
Supported by NIH-1-T-32 DK07782 (University of Michigan).

* Correspondence: Division of Surgery, Harvard Medical School,
Department of Urology, Childrens Hospital Boston, 300 Longwood
Ave., Boston, Massachusetts 02155 (e-mail: caleb.nelson@childrens.
harvard.edu).
0022-5347/07/1784-1763/0
on their respective quality of life. We developed and performed initial validation of a patient reported survey instrument to assess symptoms and bother associated with UI in
children ages 11 years or older.
METHODS
To develop an instrument for UI measurement in children
we started with an existing instrument developed by 1 of us
(JTW) that is already in use for clinical and research purposes at University of Michigan. ISI was developed primarily to assess symptoms and bother associated with stress
and urge related UI in women.4 The ISI is a 10-item Likert
scaled instrument consisting of 2 biometrically robust domains, that is symptom severity and impairment (adaptation/bother). The symptom severity domain has 3 subdomains, including stress UI severity, urge UI severity and
pad use. The ISI 1) is self-administered, 2) is concise, 3) is
able to capture relevant symptoms within a defined time
frame, 4) is able to assess the psychological impairment of
incontinence and 5) has demonstrated reliability and validity. This instrument served as the basis for ISIP.
To modify ISI to be applicable to the pediatric population
we first consulted a panel of pediatric urology experts regarding the relevant clinical conditions found in the pediat-
1763
Vol. 178, 1763-1767, October 2007

Printed in U.S.A.
DOI:10.1016/j.juro.2007.03.180
1764
INCONTINENCE SYMPTOM INDEX-PEDIATRIC
ric population and coverage of those conditions by the original ISI. Based on these assessments we elected to add 2
domains to ISI on nighttime symptoms associated with nocturnal enuresis and insensate wetting symptoms associated
with neurogenic bladder related UI. Factor analysis was
performed to determine appropriate domains for the instrument. One item was dropped because it did not contribute
distinct information. After adjustment the final ISIP instrument included 11 items covering the domains of stress UI
symptoms, urge UI symptoms, insensate UI symptoms, nocturnal UI symptoms, pad use and adaptation/bother (see
figure).
Face validity of the preliminary ISIP was assessed by a
panel of pediatric urologists. A certified literacy expert was
consulted and the language of the instrument was modified
to bring it to a fifth grade (11-year-old) reading level.
ISIP validity testing was performed at the pediatric urology clinics at University of Michigan and at an academic
general pediatric practice associated with the university.
Two arms of subjects completed the survey. Cases consisted
of boys and girls ages 11 to 17 years presenting to the

pediatric urology clinic with a chief complaint of UI or some
reasonable synonym, eg wetting. Controls consisted of similarly aged healthy boys and girls presenting to the pediatric
clinic for well child evaluation. Children were asked to complete the surveys alone, although they were permitted to ask
for assistance from a parent, if needed.
Subjects were asked to complete the survey on 2 occasions, that is at the initial visit and again 2 weeks later.
Controls included some children presenting to the pediatrician with minor symptoms of URI. They were mailed the
initial questionnaire 2 weeks later to allow time for the
acute illness to resolve and the second questionnaire was
mailed 2 weeks after this.
Survey results were compiled in a database and statistical analysis was performed using SAS software. Item-scale
correlations were assessed with r 0.60 considered evidence
of internal consistency. Correlations were assessed between
each item and the total severity score as well as between
individual items, and the impairment score and the individ-
Text of ISI-Pediatric

TABLE 1. Study participant characteristics
No.
Mean SD age (yrs)
% Male (No.)
% Completed survey 2 (No.)
Cases
Controls
Overall
19
13.0 2.1
57.9 (11)
53 (10)
19
15.6 1.7
31.6 (6)
73.7 (14)
38
14.3 2.3
44.7 (17)
63.2 (24)
ual severity subscales. Reliability was measured with Cronbachs . Convergent validity was established by correlating
symptom severity with impairment. Discriminative validity
was assessed by comparing total and domain specific scores
between cases and controls.
RESULTS
A total of 38 pediatric subjects participated in this initial
phase study, that is 19 per arm completed at least 1 survey
(table 1). Mean age of the study participants was 14.4 years.
Case children were younger than control children and there
were more boys among cases (p 0.0002 and 0.10, respectively). Of the 22 children screened as controls 15 presented
for well child visits, while 7 presented for acute URI. Four of
the URI subjects who agreed to participate by mail returned
at least 1 survey. With respect to diagnoses in cases 4 had
neurogenic bladder, 6 had primary nocturnal enuresis, 6 had
nonneurogenic daytime and nighttime voiding dysfunction,
and 3 had detrusor instability.
Evidence of internal consistency was provided by itemscale correlations (table 2). Correlations for 8 of the 9 symptom severity items exceed 0.60, indicating internal consistency. The correlation of item 8 was 0.49 but it was left in the
survey because it is a lead-in item for the following item 9.
Likewise correlations of severity items with severity subscales were high, as were those of impairment items with
the impairment scale (table 2). Convergent validity was good
with excellent correlation of the symptom severity scale and
the impairment scale (r 0.818, p 0.0001).
As measured by Cronbachs , reliability for the symptom
severity measures was high (table 3). The score for the
impairment scale was lower.
Discriminative validity was good with a total score that
was significantly different between wet children and controls. The total severity scale score was 9.3 in wet children
and 0.7 in controls (p 0.0001, table 4). Impairment scale
scores differed by 2.2 points (p 0.0001). Subscale scores
differed significantly between the groups for all subscales
1765
TABLE 3. Cronbachs for each domain, subdomain and

full instrument
Cronbachs
Full instrument
Symptom severity scale:
Stress incontinence
UI
Insensate incontinence
Nighttime incontinence
Pad use
Impairment scale
0.84
0.88
0.86
0.90
0.71
0.48
There was acceptable reliability for all domains except impairment.
except pad use, although even pad use trended toward a

difference. Urge domain scores differed by a mean of 3.6
points (p 0.0002). Nighttime wetting domain scores differed by 2.1 points, which was a dramatic difference, given
that this domain only has a single query item.
A total of 24 subjects completed 2 surveys for measuring
test-retest reliability. Reliability was excellent. Individual
response measures varied little between the first and second
instrument administrations. The symptom severity scale had
an r value of 0.95 (p 0.0001). The impairment scale had an r
value of 0.90 (p 0.0001). The total survey score had an r value
of 0.97 (p 0.0001). The symptom severity score differed by 2
or fewer points in 92% of subjects, while impairment scale
scores differed by 1 or fewer points in 92%.
DISCUSSION
In this pilot study we developed and performed initial validation studies of an instrument designed to assess symptoms and bother associated with UI in a pediatric population
ages 11 years or older. The instrument had acceptable internal consistency and reliability on pilot testing.
We found that the instrument was simple to administer
and it could be completed in a short time by the pediatric
patient, usually without significant parental assistance, as
measured by informal observations of survey administration
in the clinic. We believe that this is an important feature of
this instrument because it allows patient self-reporting in
the pediatric group without the proxy lens of parental opinion. Patient reported assessment is considered to be the
most valid and accurate form of UI assessment. It is now
recognized that patient reported outcomes frequently differ
from physician reported outcomes and patient reported outcomes are considered more reliable and accurate.5 There-
TABLE 2. Item-scale correlations of each item with total severity, impairment and each of 5 severity subscales
Severity Subscales
ISIP Item No.
Full Instrument
Total Severity
Impairment
Stress
Urge
Insensate
Nocturnal
Pad Use
1
2
3
4
5
6
7
8
9
10
11
0.90
0.61
0.83
0.84
0.74
0.73
0.59
0.45
0.59
0.60
0.80
0.88
0.61
0.82
0.82
0.72
0.72
0.63
0.49
0.60
0.56
0.75
0.82
0.54
0.74
0.79
0.70
0.64
0.35
0.22
0.48
0.71
0.89
0.97
0.89
0.71
0.57
0.61
0.72
0.38
0.20
0.62
0.53
0.70
0.75
0.46
0.94
0.93
0.88
0.65
0.32
0.21
0.35
0.52
0.77
0.69
0.65
0.61
0.52
0.67
1.00
0.30
0.08
0.34
0.34
0.65
0.41
0.25
0.34
0.35
0.15
0.30
1.00
0.43
0.17
0.36
0.24
0.51
0.24
0.23
0.42
0.21
0.23
0.35
0.90
0.86
0.23
0.38
Correlations greater than 0.60 represent evidence of interval consistency.
1766
TABLE 4. Total, domain and subscale scores of cases with chief

complaint of wetting compared to those of normal controls
without incontinence complaint
Full instrument
Symptom severity scale:
Stress incontinence
UI
Insensate incontinence
Nighttime incontinence
Pad use
Impairment scale
Cases
Controls
p Value
11.6
9.3
1.6
4.0
0.6
2.2
1.0
2.3
0.8
0.7
0.1
0.4
0.1
0.1
0.1
0.1
0.0001
0.0001
0.0008
0.0002
0.0315
0.0003
0.0860
0.0001
fore, patient administered questionnaires should be used

when possible. For infants and young children the parent is
obviously the only plausible source of information. However,
for older children parent and patient assessments of health
status often diverge as the child becomes more independent,
particularly in areas concerned with emotional and social
well-being.6,7 For this reason it is preferable to use the
patient assessment even if that patient is a minor.8,9 Parents are useful for complementing the information provided
by the child but they should not be the primary source of
assessment, if possible.10,11
No validated instruments for incontinence assessment in
children are available. This reflects the broader lack of measures of health related quality of life of pediatric patients,
which is largely due to complex methodological issues. The
cognitive, emotional and behavioral development of children
is age dependent, so that instruments must be specific to the
particular developmental stage of the target population.12
Moreover, the rate and degree of cognitive development varies widely and intellectual ability may be affected by disease,
eg hydrocephalus. Although concerns have been raised regarding biases that may be present when children respond
to questionnaires, it is possible to develop useful instruments for disease specific and general pediatric health related quality of life.1315
Recently some efforts were made regarding UI measurement in children but none of these instruments adequately
assess child reported symptoms and bother associated with
UI specifically. Farhat et al developed a superb parent reported questionnaire that addresses dysfunctional voiding
in children and includes items addressing incontinence.16
However, this instrument is specifically aimed at the dysfunctional voider and it is not a global assessment of continence status or bother. A similar tool was developed by a
group from Turkey17 but this likewise uses parent reported
responses. Sureshkumar et al dispensed with the attempt at
patient reporting and developed a parent reported instrument for school aged children, which also focuses on urinary
tract infection risk factors.18 Finally, Bower et al recently
described a child reported instrument evaluating quality of
life associated with bladder dysfunction.19 While it is comprehensive, this instrument focuses exclusively on bother
and does not seek to measure symptom severity. Therefore,
none of the existing instruments encompass in a single practical tool the ability to measure the symptoms and bother
associated with incontinence and measure this through
child reported responses without parental proxy. For this
reason we believe that the current tool advances the field.
Although patient reporting is ideal for a survey instrument, not every age group would be appropriate for a self-
reported UI questionnaire. Cognitive development is a progressive process in children, requiring survey instruments
to be developmentally level appropriate. Children 11 years
or older have advanced abilities. They engage in abstract
thinking and self-conceptualization, and interact with other
individuals at a high level. They have a well developed
ability to judge the thoughts, emotions and behaviors of
other people. They have a solid understanding of time, including the frequency of events.14 They also have a clear
understanding of higher concepts, such as jealousy, shame
and worry.20 Taken together this suggests that children 11
years or older should be able to provide accurate responses
to basic questions about their health status, incontinence
severity and bother. Although we asked the children to
complete the survey themselves in this study, we did not
supervise the administration and we permitted parental
assistance if needed. We believe that this compromise approach is superior to enforcing a rigid child-only rule, resulting in incomplete forms. Of the children who completed the
initial survey 53% returned the followup survey. While this
is not ideal, this proportion was adequate for analysis and
there was no difference between the arms in the rate of
return of followup surveys (p 0.31).
We note that instruments such as this are designed to
assess the magnitude of symptoms and bother associated
with a given condition. The validation process demonstrates
that such instruments can reliably and reproducibly measure a given symptom complex. However, they are not intended to diagnose specific conditions or determine the underlying etiology. For these purposes a thorough history is
always essential.
CONCLUSIONS
This pilot study provides initial validation of a survey instrument for UI in children and adolescents. This instrument can be used in children 11 to 17 years old to objectively
and reproducibly measure patient reported UI. Further
study of this instrument will evaluate its usefulness for
assessing the treatment response in a variety of clinical
settings and patient populations. We hope that eventually
this instrument or ones like it will be routinely used in
pediatric urology to assess the severity of illness and the
effectiveness of our interventions.

ISI
ISIP
UI
URI
Incontinence Symptom Index

ISI-Pediatric
urinary incontinence
upper respiratory infection
REFERENCES
1.
Bloom DA, Seeley WW, Ritchey ML and McGuire EJ: Toilet

habits and continence in children: an opportunity sampling
in search of normal parameters. J Urol 1993; 149: 1087.
2. Bakker E, van Sprundel M, van der Auwera JC, van Gool JD
and Wyndaele JJ: Voiding habits and wetting in a population of 4,332 Belgian schoolchildren aged between 10 and 14
years. Scand J Urol Nephrol 2002; 36: 354.
3. Hagglof B, Andren O, Bergstrom E, Marklund L and Wendelius M: Self-esteem before and after treatment in children
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
15.
16.
with nocturnal enuresis and urinary incontinence. Scand

J Urol Nephrol, suppl., 1997; 183: 79.
Wei JT, Dunn RL, Hoag L, Faerber GJ, Dorr R and McGuire
EJ: The Incontinence Symptom Index (ISI): a novel and
practical symptom score for the evaluation of urinary incontinence severity. J Urol, suppl., 2003; 169: 33, abstract
128.
Wei JT and Montie JE: Comparison of patients and physicians
rating of urinary incontinence following radical prostatectomy. Semin Urol Oncol 2000; 18: 76.
Jokovic A, Locker D and Guyatt G: How well do parents know
their children? Implications for proxy reporting of child
health-related quality of life. Qual Life Res 2004; 13: 1297.
Theunissen NC, Vogels TG, Koopman HM, Verrips GH, Zwinderman KA, Verloove-Vanhorick SP et al: The proxy problem: child report versus parent report in health-related
quality of life research. Qual Life Res 1998; 7: 387.
Levi RB and Drotar D: Health-related quality of life in childhood cancer: discrepancy in parent-child reports. Int J Cancer, suppl., 1999; 12: 58.
Sawyer M, Antoniou G, Toogood I and Rice M: A comparison of
parent and adolescent reports describing the health-related
quality of life of adolescents treated for cancer. Int J Cancer, suppl., 1999; 12: 39.
Parsons SK, Barlow SE, Levy SL, Supran SE and Kaplan SH:
Health-related quality of life in pediatric bone marrow
transplant survivors: according to whom? Int J Cancer,
suppl., 1999; 12: 46.
le Coq EM, Boeke AJ, Bezemer PD, Colland VT and van Eijk
JT: Which source should we use to measure quality of life in
children with asthma: the children themselves or their
parents? Qual Life Res 2000; 9: 625.
Pal DK: Quality of life assessment in children: a review of
conceptual and methodological issues in multidimensional
health status measures. J Epidemiol Commun Health
1996; 50: 391.
Landgraf JM, Abetz L and Ware JR: Child Health Questionnaire (CHQ): A Users Manual. Boston: The Health Institute, New England Medical Center 1996.
French D and Christie M: Developing outcome measures for
children: quality of life assessment for pediatric asthma.
In: Health Outcome Measures in Primary and Outpatient
Care. Edited by A Hutchinson, E McColl and C Riccalton.
Amsterdam: Hardwood Academic Publishers 1996; pp 45
63.
Jokovic A, Locker D, Stephens M, Kenny D, Tompson B and
Guyatt G: Validity and reliability of a questionnaire for
measuring child oral-health-related quality of life. J Dental
Res 2002; 81: 459.
Farhat W, Bagli DJ, Capolicchio G, OReilly S, Merguerian PA,
Khoury A et al: The dysfunctional voiding scoring system:
quantitative standardization of dysfunctional voiding
symptoms in children. J Urol 2000; 164: 1011.
1767
17.
Akbal C, Genc Y, Burgu B, Ozden E and Tekgul S: Dysfunctional voiding and incontinence scoring system: quantitative evaluation of incontinence symptoms in pediatric population. J Urol 2005; 173: 969.
18. Sureshkumar P, Cumming RG and Craig JC: Validity and
reliability of parental report of frequency, severity and risk
factors of urinary tract infection and urinary incontinence
in children. J Urol 2006; 175: 2254.
19. Bower WF, Wong EM and Yeung CK: Development of a validated quality of life tool specific to children with bladder
dysfunction. Neurourol Urodyn 2006; 25: 221.
20. Bee H: Lifespan Development. New York: Addison Wesley
Longman 1998.
EDITORIAL COMMENT
These types of articles are long overdue in pediatric urology.
The authors followed the recommended basic steps in the
development of a questionnaire, including item generation,
item reduction, reliability assessment and different types of
validity.1 The main aim of this instrument is to quantify UI
severity in children who are 11 to 17 years old. It would be
of great value if this questionnaire were validated in a larger
population and for other purposes, such as responsiveness.
Many published questionnaires dealing with pediatric
lower urinary tract dysfunction are deficient in their psychometric methodology. For example, the questionnaires designed by Farhat et al (reference 16 in article) and Akbal
et al (reference 17 in article) lack an assessment of testretest reliability. An instrument is not valid unless it is
reliable.
Application of an instrument to individuals should be
done with caution. Common mistakes include application to
a different population and use for a nonvalidated purpose.
For example, the instrument developed by Akbal et al (reference 17 in article) cannot be used in North America unless
it is validated in an English speaking population. Similarly
the current questionnaire cannot be used to diagnose the
cause of incontinence. To ensure the quality of an instrument the involvement of experts with training in clinical
epidemiology and psychometrics is a crucial step. The authors used robust methods to validate this questionnaire.
Kourosh Afshar
Department of Urologic Sciences
University of British Columbia
Vancouver, British Columbia
Canada
1.
Streiner D and Norman G: Health Measurement Scales: A Practical Guide to Their Development and Use, 3rd ed. Oxford,
United Kingdom: Oxford University Press 2003.
Bladder II
The Durability of Intravesical Oxybutynin Solutions Over Time
Julian Wan*, and Cody Rickman
From the Division of Pediatric Urology, Department of Urology, University of Michigan Medical Center, Ann Arbor, Michigan
Purpose: Intravesical oxybutynin is prescribed for patients who do not tolerate the side effects of oral intake. Standard oral
tablets are broken up, dissolved and instilled. For convenience some families create a large volume of solutions to use for
weeks. We investigated the stability of oxybutynin concentration in these solutions and the effect of varying solvents and
solutes.
Materials and Methods: Solutions of oxybutynin were created with a starting concentration of 125 g/ml. We varied the
method of pill preparation (crushed vs passive dissolution), type of solvent (tap water vs normal saline) and the presence of
the commonly instilled antibiotic, gentamicin sulfate (0.48 mg/ml). Solutions were aged for 1 to 4 weeks. Concentrations of
oxybutynin were measured by gas chromatography.
Results: Tap water as a solvent had the greatest decrease in concentration, declining below 50% by week 2 and reaching 25%
by week 4. The mean concentration of the other solutions decreased less, maintaining 80% of the starting concentration at
week 4. The use of normal saline as a solvent and the addition of gentamicin sulfate helped slow the decrease. Crushing vs
passive dissolution showed no effect.
Conclusions: Homemade oxybutynin solutions degrade with time. Tap water solvent alone should be avoided. Solutions and
additives that lower pH and are more acidic help oxybutynin solubility and stability. Storing oxybutynin solutions for later
intravesical use appears feasible up to 4 weeks. The clinical effect of the 20% degradation remains to be demonstrated.
Key Words: oxybutynin; urinary bladder, neurogenic; cholinergic antagonists; pharmacokinetics;
administration, intravesical
nticholinergic medications used in conjunction with

CIC have been effective for treating patients with
neurogenic bladder, detrusor hyperreflexia and high
bladder storage pressures due to poor compliance.1 Oxybutynin has been the primary anticholinergic medication for
several decades and it remains the primary choice for most
patients.2 Unfortunately not all patients respond clinically
to oral oxybutynin therapy despite being given the maximal
dose. Others cannot tolerate oral oxybutynin because of side
effects, such as dry mouth, flushing, constipation and gastrointestinal upset. Oral oxybutynin is metabolized in the
liver and the N-desethyl-oxybutynin metabolites have an
important role in the development of side effects.3
For patients on CIC the intravesical administration of
oxybutynin (an off label, non-Food and Drug Administration
approved use) was introduced as a response to these 2 concerns. It allows the administration of oxybutynin to patients
who have reached their maximal tolerated dose of oral oxybutynin or who cannot tolerate oral therapy due to side
effects. Laboratory and clinical reports demonstrate that
intravesical oxybutynin therapy achieve these aims, while
avoiding the side effects generated by first pass metabolites
from the liver.4 8
* Correspondence: Division of Pediatric Urology, Department of

Urology, University of Michigan Medical Center, 3875 Taubman
Center 0330, 1500 East Medical Center Dr., Ann Arbor, Michigan
48109-0330 (telephone: 734-615-7996; FAX: 734-615-3520; e-mail:
juliwan@umich.edu).
Financial interest and/or other relationship with Surgical
Devices.
0022-5347/07/1784-1768/0
There is no commercial preparation of oxybutynin for

intravesical use, requiring patients and families to create
their own solutions for instillation. The necessity to make up
a solution each time can be a major practical inconvenience.
Palmer et al evaluated their patients who were prescribed to
instill solutions of crushed oxybutynin 3 times per day and
found that 27% discontinued therapy primarily because of
the inconvenience of having to make up the solutions.9 To
work around this concern some of our families have reported
that they create large quantities of oxybutynin solutions to
be stored and instilled over a period. We were not sure
whether this method would result in a loss of efficacy due to
oxybutynin breakdown. We could find no prior research on
the degradation of oxybutynin chloride with time, nor could
we find any information on what solutions may be better
suited for stable storage. To help answer these questions we
performed a study of the durability of intravesical oxybutynin solutions.
We investigated 4 factors that may affect the durability of
oxybutynin solutions, including time, solvent type, intravesical antibiotics and pill preparation. Oxybutynin chloride
tablets (5 mg size) were dissolved in 40 ml solvent, creating
a starting concentration of 125 g/ml. The pills were crushed
between 2 common household spoons or placed intact and
allowed to passively dissolve. The solvents used were potable tap water drawn from the laboratory sink and 0.9%
normal saline. We also included as a variable the presence of
the intravesical antibiotic gentamicin sulfate (480 g/ml),
1768
Vol. 178, 1768-1770, October 2007

Printed in U.S.A.
DOI:10.1016/j.juro.2007.03.181
DURABILITY OF INTRAVESICAL OXYBUTYNIN SOLUTIONS OVER TIME

Oxybutynin chloride concentration and starting value percent
Oxybutynin Concentration (mcg/ml)
Solution
Tap water &
gentamicin
sulfate:
Passive
Crushed
Normal saline
gentamicin
sulfate:
Passive
Crushed
Normal saline:
Crushed
Passive
Tap water:
Crushed
Passive
Wk 1
Wk 2
Wk 3
Wk 4 (% starting value)
116.3
116.3
111.3
100.0
96.3
107.5
98.8 (79)
106.3 (85)
120.0
112.5
112.5
112.5
123.8
106.8
102.5 (82)
101.3 (81)
111.3
115.0
127.5
115.0
97.5
95.0
107.5 (86)
106.3 (85)
53.8
65.0
31.3
41.3
31.3
31.3
31.3 (25)
31.3 (25)
which many of our patients commonly use as a form of

antibiotic prophylaxis.10 Solutions were created and aged for
4 weeks at room temperature. Samples were taken and
analyzed at 1, 2, 3 and 4 weeks. A total of 8 test solutions
with oxybutynin were created along with blank controls (see
table).
Oxybutynin is not a drug that is normally measured in
the blood, urine or other human fluid. Therefore, all samples
were measured using gas chromatography at a commercial
laboratory using a standard technique. Duplicate calibrators
were prepared at concentrations that were 120%, 100%,
75%, 50% and 25% of the expected concentration of 125
g/ml. Duplicated controls were prepared at 100% and 50%
of the target. Controls were dispersed throughout the analytical batch. An internal standard (8-methoxyloxapine) was
added to each specimen, calibrator, control and blank matrix, and all were extracted into a solvent after being made
basic with sodium hydroxide. The extraction solvent from
each was analyzed on a Model 6890 gas chromatograph
(Agilent Technologies, Santa Clara, California) with a
DB-17 capillary column and nitrogen specific detector. Samples that showed the presence of oxybutynin below the lowest calibration point (25% of target) were reported as less
than 25% along with the calculated values. These values
were quantitative estimates since they were outside of the
calibration range. Calibrators were linear over the concentration range and all calibrators back calculated to within
5% of the target value. Control values were within 5% of the
target value with a spread of 104% and 52%. The chromatograms were clean with no extraneous peaks or shouldering
peaks observed.
Acidity can affect oxybutynin solubility. The pH values of
solvent solutions were measured using a standard commercial laboratory pH meter. They were correlated with degradation rate.
Results are reported as measured concentrations in
g/ml and as a percent of the original target concentration of
125 g/ml. Comparisons and statistical analysis was done
using KaleidaGraph statistical software (Synergy Software,
Reading, Pennsylvania).
tap water only solutions decreasing by about 50%. There

were 5 notable values. Four solutions had a concentration of
less than 31.3 g/ml, which is the lowest calibration point at
25%, and they are reported as this value. Their estimated
quantitative values were lower between 21 and 30 g/ml but
these values were not believed to be as accurate because
they were outside of the calibration range. There was an
outlier value of 127.5 g/ml in the normal saline crush
solution (see table). This was clearly above the control concentration and it represented a 2% error that was within the
range of the detection methodology. For purposes of calculation it was treated as a control value of 125 g/ml.
By week 4 the solutions containing normal saline or gentamicin sulfate had decreased to an average of 83%. The
most precipitous decline was found in solutions that contained only tap water as solution. They decreased to a value
of 25% by week 3 and remained there for week 4. Solutions
with normal saline or gentamicin sulfate had similar results
and they were statistically different from each other.
When average concentrations during the 4 weeks were
plotted, this effect could be seen graphically (see figure)
When these values were compared statistically at 4 weeks,
there was no difference between the average value of the
solutions containing tap water and gentamicin, and those
with normal saline or normal saline and gentamicin
(p 0.71). In contrast, the comparison between tap water
only solutions and the other solutions was statistically significant (p 0.0001).
In regard to pH in our test solutions, the 0.9% normal
saline had a mean pH of 6.3 (range 6.1 to 6.4), the gentamicin sulfate in normal saline solution was 4.81 (range 4.80 to
4.85), the tap water solution was 8.6 (range 8.4 to 8.8) and
the gentamicin sulfate in tap water solution was 6.4 (range
6.2 to 6.5).
DISCUSSION
Intravesical administration is an effective alternative to oral
oxybutynin therapy in patients who practice CIC. It avoids
the side effects associated with liver metabolites and provides a way of increasing the oxybutynin dose beyond the
level tolerated when given orally. The inconvenience of pre-
RESULTS
The concentration of oxybutynin in all solutions decreased
with time (see table). By week 1 all had decreased with the
1769
Mean oxybutynin concentration with time. mcg, g
1770
DURABILITY OF INTRAVESICAL OXYBUTYNIN SOLUTIONS OVER TIME
paring the solution each time can be mitigated by preparing

it in batches ahead of the time of use. Our data suggest that
the degradation of oxybutynin when in solution with normal
saline or gentamicin sulfate may only have a minimal clinical effect. The 20% decrease in concentration would be the
equivalent of a decrease from the instillation of a 5 mg tablet
to that of a 4 mg tablet. Whether there is any significant loss
of effect clinically remains to be proven but it seems unlikely.
The rapid degradation of oxybutynin in solutions with tap
water only may be related to its solubility and dissociation
state. Oxybutynin has a pKa of 8.04.11 It solubility in aqueous solution is inversely related to pH. Solubility at pH 1 is
77 mg/ml, which decreases to 0.8 mg/ml at pH 6 and is only
0.012 mg/ml at pH greater than 9.6.11 Using the HendersonHasselbalch equation, pH pKa log ([A-]/[HA]), we calculated the relative ratio of dissociated to nondissociated
forms of oxybutynin in these solutions.12 In tap water solutions with a pH of 8.6 the dissociated form was 3.6 times
more common than the nondissociated form. In contrast, in
the tap water with gentamicin sulfate solution the nondissociated form was 43.5 times more common than the dissociated form. Decreased solubility and a more dissociated
state may make oxybutynin less available in solution and
easier to degrade.
Tap water pH varies greatly among the communities
around Ann Arbor, Michigan. The Ann Arbor Water Utilities
Department reported a pH of 9.3 (range 9.2 to 9.5) measured
at the point that water is piped into circulation. It was
slightly more acidic at 8.6 when it left the laboratory tap.
Other communities that report water pH in Michigan show
a range of 7.1 to 7.5 (Battle Creek mean 7.5, range 7.4 to 7.7;
Grand Rapids mean 7.5, range 7.2 to 7.9; and Warren mean
7.1, range 6.8 to 7.4).13 The wide variability and its possible
effect on durability preclude the use of tap water alone as a
solvent when making up oxybutynin solutions.
Storing homemade solutions from which repeat aliquots
are drawn raises concerns about bacterial contamination.
While we did not specifically study this concern during this
project, data are available on solutions with gentamicin
sulfate, which is commonly used as an intravesical irrigant
in patients with neurogenic bladder. Solutions of gentamicin
sulfate at the concentrations used in this project have been
stored at room temperature up to 2 months without any loss
of efficacy against common pathogens, such as Pseudomonas
aeruginosa, Escherichia coli, Enterobacter cloacae, Proteus
mirabilis and Klebsiella penomoniae.10
CONCLUSIONS
Homemade solutions of oxybutynin degrade with time. Degradation is worst in solutions containing tap water only.
They showed a greater than 70% loss by 4 weeks. Solutions
with normal saline or gentamicin sulfate as an additive were
preferable. They retained about 80% of their original starting concentration at 4 weeks. The clinical significance of the
20% degradation remains to be demonstrated. There was no
significant difference between crushing and passively dissolving tablets when making up the solution. Oxybutynin
stability and solubility are improved at lower pH solutions.
Local and regional tap water pH was highly variable. making it unsuitable for use alone when creating solutions. We
recommend that tap water should be avoided as solvent if
possible or combined with gentamicin sulfate to make it
more acidic. If these recommendations are followed, we believe that solutions of oxybutynin can be stored and used up
to 4 weeks.
ACKNOWLEDGMENTS
Oxybutynin samples were measured at National Medical
Services Testing Laboratory, Willow Grove, Pennsylvania.

NS normal saline
pKa dissociation constant
REFERENCES
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
Mulcahy JJ, James HE and McRoberts JW: Oxybutynin

chloride combined with intermittent catheterization in
the treatment of myelomeningocele patients. J Urol 1977;
118: 95.
Diokno AC and Lapides J: Oxybutynin: a new drug with analgesic and anticholinergic properties. J Urol 1972; 108: 307.
Buyse G, Waldeck K, Verpoorten C, Bjork H, Casaer P and
Andersson KE: Intravesical oxybutynin for neurogenic
bladder dysfunction: less systemic side effects due to reduced first pass metabolism. J Urol 1998; 160: 892.
Massad CA, Kogan BA and Trigo-Rocha FE: The pharmacokinetics of intravesical and oral oxybutynin chloride. J Urol
1992; 148: 595.
Madersbacher H and Knoll M: Intravesical application of oxybutynin: mode of action in controlling detrusor hyperreflexia. Eur Urol 1995; 28: 340.
Kaplinsky R, Greenfield SP, Wan J and Fera M: Expanded
follow-up of intravesical oxybutynin chloride use in children
with neurogenic bladder. J Urol 1996; 156: 753.
Connor JP, Betrus G, Fleming P, Perlmutter AD and Reitelman
C: Early cystograms can predict the response to intravesical
instillation of oxybutynin chloride in myelomeningocele patients. J Urol 1994; 151: 1045.
Prasad KV and Vaidyanathan S: Intravesical oxybutynin chloride and clean intermittent catheterization in patients with
neurogenic vesical dysfunction and decreased bladder capacity. Br J Urol 1993; 72: 719.
Palmer LS, Zebold K, Firlit CF and Kaplan WE: Complications
of intravesical oxybutynin chloride therapy in the pediatric
myelomeningocele population. J Urol 1997; 157: 638.
Wan J, Kozminski M, Wang SC, Faerber GJ, McGuire EJ,
Bloom DA et al: Intravesical instillation of gentamicin sulfate: in vitro, rat, canine, and human studies. Urology 1994;
43: 531.
ONeil MJ, Smith A, Heckelman PE, Obenchain JR Jr,
Gallipeau JAR and DArecca MA: Oxybutynin, entry
7024. In: Merck Index. Whitehouse Station, New Jersey:
Merck Co 2006; p 1245.
Stryer L: Biochemistry, 2nd ed. San Francisco: WH Freeman
and Co 1981; chapt 2, pp 39 40.
Water Quality Reports. Michigan Department of Utilities and
Water 2005.
Tumor in Bladder Reservoir After Gastrocystoplasty

Miguel Castellan, Rafael Gosalbez, Marcos Perez-Brayfield, Patrick Healey, Ruth McDonald,
Andrew Labbie and Thomas Lendvay
From the Division of Pediatric Urology, Miami Childrens Hospital and Jackson Memorial Hospital, Department of Urology, University of
Miami, Miami, Florida, and Childrens Hospital and Regional Medical Center, Seattle, Washington
Purpose: To our knowledge the risk of malignancy in patients with previous bladder augmentation with stomach is
unknown. We report 3 cases of gastric adenocarcinoma and 1 of transitional cell carcinoma after augmentation cystoplasty
with stomach with long-term followup.
Materials and Methods: Between August 1989 and August 2002, 119 patients underwent augmentation cystoplasty with
stomach at our 2 institutions (University of Miami School of Medicine, and Seattle Childrens Hospital and Regional Medical
Center). Medical records, urodynamic studies, radiographic imaging and laboratory evaluations were reviewed retrospectively and cases of malignancy were analyzed in detail.
Results: Four male patients had carcinoma after augmentation gastrocystoplasty. Preoperative diagnosis was neurogenic
bladder in 3 patients and posterior urethral valve in 1. Three patients had gastric adenocarcinoma, while the other had poorly
differentiated transitional cell carcinoma. Each case progressed to malignancy more than 10 years after augmentation (11,
12, 14 and 14 years, respectively).
Conclusions: Patients who undergo bladder augmentation with a gastric remnant are at increased risk for malignancy,
probably similar to that in patients with enterocystoplasty. Therefore, they require close long-term followup. Patients should
be followed annually with ultrasound, and cystoscopy should be performed annually starting 10 years after gastrocystoplasty
unless they have abnormal ultrasound, hematuria or another cancer risk factor. Any suspicious lesions should be biopsied,
especially at the gastrovesical anastomotic site.
Key Words: bladder; carcinoma, transitional cell; bladder neoplasms; stomach; transplants
here are many reports of malignant degeneration in

urinary diversions using isolated intestinal segments as
well as in conduits, cystoplasties, continent reservoirs
and ileal ureter replacement.1,2 Cancer following bladder augmentation is a recognized risk factor. The reported incidence of
malignancy after augmentation cystoplasties using colonic or
ileal segments is reported to be 1.2% to 5.5%.13 The bladderintestinal junction is the most common location of the reported
tumors and most have been adenocarcinoma of the bladder,
although TCC was reported in 39% of patients.2,3
In 1988 gastric segments for bladder augmentation were
popularized by Adams et al.4 This has been recommended in
children with metabolic acidosis, renal failure or deficient
bowel segments suitable for bladder augmentation. To our
knowledge the risk of malignancy in patients with previous
bladder augmentation with stomach is unknown. A theoretical concern is that there is a higher incidence of carcinoma
in native stomach relative to ileum. We report 4 cases of
carcinoma after augmentation gastrocystoplasty. Only 3
other cases with malignant degeneration following gastrocystoplasty are reported in the literature.5 8

Between August 1989 and August 2002, 119 patients underwent augmentation cystoplasty with stomach at 2 institutions, including 29 at University of Miami School of Medicine, and 90 at Seattle Childrens Hospital and Regional
Medical Center. The medical records, urodynamic studies,
0022-5347/07/1784-1771/0
radiographic imaging and laboratory evaluations were reviewed retrospectively. A reservoir tumor developed in 3 of 29
patients (10.3%) in whom gastrocystoplasty was done at University of Miami School of Medicine and 1 of 90 (1.1%) in whom
gastrocystoplasty was done at Seattle Childrens Hospital and
Regional Medical Center. We gave particular attention to these
4 patients with tumor after gastrocystoplasty.
RESULTS
Patient age at augmentation was 6, 13, 18 and 22 years, respectively. Four male patients had carcinoma after augmentation gastrocystoplasty. Preoperative diagnoses were neurogenic bladder due to spina bifida in 3 cases and posterior
urethral valves in 1. No patient had a history of smoking
exposure. None of these patients had a post-augmentation
history of recurrent urinary tract infection, lithiasis in the
reservoir or hematuria-dysuria syndrome. Patient 4 in our
series had an additional risk factor for tumor formation since
he was a renal transplant recipient on immunosuppressive
medical therapy. In 2 patients cystoscopy was done 1 year
before the diagnosis of bladder tumor but it did not identify the
reservoir tumor.
Each case progressed to malignancy during 10 years after
augmentation, including 11, 12, 14 and 14 years, respectively. Table 1 lists patient characteristics, final surgery,
tumor stage, adjuvant therapy and outcome. In case 1 the
tumor was difficult to diagnose with ultrasound, CT, cystogram and cystoscopy because it presented with thickening of
1771
Vol. 178, 1771-1774, October 2007

Printed in U.S.A.
DOI:10.1016/j.juro.2007.05.100
1772
TUMOR IN BLADDER RESERVOIR AFTER GASTROCYSTOPLASTY

TABLE 1. Patient characteristics, final surgery, adjuvant therapy and outcome
Pt 1
Age at surgery (yrs)

Diagnosis
Associated disease
22
Spina bifida
None
Cystoscopy before bladder At 1 yr severe wall inflammation,

tumor diagnosis
no identifiable mass
Bladder tumor diagnosis Ureterovesical junction obstruction
of solitary kidney
Final surgery
Radical cystoprostatectomy, PLND,
partial pubectomy, proximal
urethrectomy, ileal conduit
Latency (yrs)
12
Metastatic evaluation
Local pos margin, distal neg
Bladder TNM stage
Adjuvant therapy
Outcome (mos)
Pt 2
T3/N0/M0
Radiotherapy, chemotherapy
(5-fluorouracil leucovorin)
Disease-free after radiotherapy
chemotherapy (20)
Pt 3
18
13
Spina bifida
None
Spina bifida
None
At 1 yr no identifiable mass None

Macroscopic hematuria
Pt 4
6
Posterior urethral valves
End stage renal disease/living
related kidney transplant
None
Macroscopic hematuria
Severe hydronephrosis of
transplanted kidney
Radical cystoprostatectomy, Radical cystoprostatectomy, Exploratory laparotomy, biopsy
PLND, ileal conduit
PLND, ileal conduit
of perivesical lymph nodes
omental studs
14
11
14
Neg
Neg
Pos lungs, lymph nodes,
peritoneal cavity
T1/N0/M0
T2/N0/M0
T4/N2/M1
None
None
Chemotherapy (carboplatin
taxol)
Disease-free postop (14)
Disease-free postop (36)
Dead after chemotherapy
initiation (1)
All patients were male.
the bladder wall without protrusion inside the lumen of the

reservoir (linitis plastica type). In 3 of the 4 patients the
tumor was localized at the gastrovesical anastomosis.
Figure 1 shows cystoscopy of the bladder tumor at the gastrovesical anastomotic site in patient 2. In patient 4 the tumor
involved the entire reservoir. Figure 2 shows pelvic CT in
this patient. Patient 3 in our series had positive urinary
cytology suggestive of bladder TCC and final pathological
findings revealed gastric adenocarcinoma.
Table 2 lists final pathological results. Three patients had
gastric adenocarcinoma, while the last male had poorly differentiated TCC. Figures 3 and 4 show pathological images
of patients 1 and 4, respectively.
DISCUSSION
The development of malignancy after enterocystoplasty is
well documented.13 A review of the literature in 2004 revealed that a neoplasm developed in 45 patients with previous bladder augmentation.1 There is an interval to the
formation of cancer in the augments since augmentation is
FIG. 1. Cystoscopy shows bladder tumor at gastrovesical anastomotic site.
frequently performed in childhood and malignancy is not

identified until adulthood. Many pediatric urologists in the
United States do not follow their patients with augmentation into adulthood, so that there is concern that cystoscopic
or cytological surveillance protocols are not performed.
The estimated latency between bladder augmentation
and tumor development in different series is long (mean 19
to 22 years).13 In 2004 in a review of the literature by
Austen et al 5 of 30 patients with previous ileocystoplasty
had neoplasm with a latency period of less than 10 years.1
Most cases occurred in adults with other risk factors, such as
tobacco use. The most common reason for bladder augmentation in these patients was urogenital tuberculosis.
In patients with a short latency period between operation
and tumor development there is an unresolved question of
whether the tumor was coincidental or whether it developed
due to the augmented bladder. The most common types of
tumors at this site are adenocarcinoma and TCC. In 2004
Soergel et al reported their series of children and adolescents who underwent augmentation cystoplasty.2 At a minimum of 10 years of followup 3 of 260 patients (1.15%) with
ileocecal (2) and cecal (1) bladder augmentation had TCC.
We report 4 cases of gastrocystoplasty in patients with
neurogenic bladder (3) and posterior urethral valves (1),
FIG. 2. Pelvic CT reveals severe gastrocystoplasty wall thickening.

Note Foley catheter inside bladder reservoir.
1773
TABLE 2. Histological tumor findings in bladder reservoir after gastrocystoplasty

Pt No.
Pathological Findings
Localization
Extension
Invasive, poorly differentiated signet ring cell adenoca
Gastrovesical anastomosis
Moderately differentiated gastric adenoca
Invasive, poorly differentiated gastric adenoca
Poorly differentiated TCC
Whole reservoir
Gastric flap, extension into muscularis propria

deep bladder tissues
Gastric flap, extension into bladder lamina
propria
Gastric flap, invasion of bladder muscularis
propria
Whole reservoir, deep tissue invasion pos
perivesical lymph nodes
complicated by adenocarcinoma in 3 and TCC in 1. Our

series includes patients in whom gastrocystoplasty was done
at 2 institutions, including 3 of 29 (10.3%) at Miami Childrens Hospital/University of Miami School of Medicine and
1 of 90 (1.1%) at Seattle Childrens Hospital and Regional
Medical Center.
During the last 20 years gastrocystoplasty has been performed as an alternative to ileocystoplasty or colocystoplasty
to avoid complications associated with the use of intestinal
segments.9 However, the long-term risks of neoplasms in the
setting of gastrocystoplasty are not well known. Qui et al
reported on a 73-year-old woman who had TCC 14 years
following gastrocystoplasty.5 In 2005 Baydar et al reported
the first case of our series,8 while Esquena Fernandez reported on a 27-year-old male with gastric adenocarcinoma
on the stomach side of a bladder augmentation.6 Recently
Balachandra et al reported on a 64-year-old man with a
history of spina bifida in whom adenocarcinoma arose in the
gastric mucosa 10 years after gastrocystoplasty.7
One patient in our series had an additional risk factor
since he was a renal transplant recipient on immunosuppressive medical therapy. The other 3 patients did not have
any other identifiable risk factors, such as tobacco use, recurrent urinary tract infection or urinary stones. Bladder
augmentation appeared to be an independent risk factor for
bladder carcinogenesis in this set of patients. In 3 cases
carcinoma arose near the surgical anastomosis and in the
other the tumor involved the entire reservoir and it was
impossible to define the exact initial location. In all patients
the neoplasm was associated with extensive transitional cell

metaplasia and dysplasia of the gastric mucosa.
Animal models of augmentation cystoplasty have been
used to evaluate the development of malignancy with the
histological abnormalities commonly observed at the enterovesical junction.10,11 The exact mechanism is unknown
but it was proposed that inflammatory changes at the anastomotic site may induce growth factor production and increased cellular proliferation. Buson et al found papillary
hyperplasia at the anastomotic site in more than 50% of
animals that survived 1 year after gastrocystoplasty.10
Close et al reported that transitional cell hyperplasia and
metaplasia were found in the anastomotic line in 12 rats
that survived more than 21 months after gastrocystoplasty.11 These changes are not specific for malignant transformation but they are characteristic of premalignant
changes that occur in other epithelia.
One patient had positive urinary cytology suggestive of
bladder TCC and final pathological results revealed gastric
adenocarcinoma. Urine cytology has not been recommended
for screening because of mucus and other contaminants.3
The role of urinary cytology in patients with previous bladder augmentation is still unclear. Interestingly cystoscopy
was done in 2 cases in our series a year before the diagnosis
of gastric carcinoma and it was reported to be normal. Even
with close surveillance the diagnosis of malignancy in the
augmented bladder remains a challenge. It is possible that
bladder wall thickening compared to previous imaging may
be the only indication of possible malignancy. Unfortunately
FIG. 3. Case 1. Signet ring cell adenocarcinoma arising in gastric

flap. Note signet ring cells in lamina propria surrounding gastric
gland. Reduced from 40.
FIG. 4. Case 4. Poorly differentiated tumor containing sheets of

epithelioid cells with large vesicular nuclei and prominent nuclei
with abundant mitotic figures. Reduced from 200.
1774
the specificity of this finding is poor and it may be related

more to cystitis due to acid irritation or infection.
Soergel et al recommended endoscopic surveillance of all
patients with a history of enterocystoplasty beginning 10
years after the initial procedure.2 In our series none of the
patients had a reservoir tumor before 10 years postoperatively. Because 2 patients were diagnosed with hydronephrosis, we still recommend annual surveillance by routine
ultrasound. All patients should undergo cystoscopy annually
starting 10 years after gastrocystoplasty unless they have
abnormal ultrasound, hematuria or another cancer risk factor. Any suspicious lesions should be biopsied, especially at
the gastrovesical anastomotic site.
CONCLUSIONS
Patients who undergo bladder augmentation with a gastric
remnant are at an increased risk for malignancy, probably
similar to that in patients with enterocystoplasty. Therefore,
they require close long-term followup. Patients should be followed annually with ultrasound, and cystoscopy should be
performed annually starting 10 years after gastrocystoplasty
unless they have abnormal ultrasound, hematuria or another
cancer risk factor. Any suspicious lesions should be biopsied,
especially at the gastrovesical anastomotic site.

CT computerized tomography
PLND peritoneal lymph node dissection
TCC transitional cell carcinoma
REFERENCES
1.
2.
3.
4.
5.
6.
Austen M and Kalble T: Secondary malignancies in different

forms of urinary diversion using isolated gut. J Urol 2004;
172: 831.
Soergel TM, Cain MP, Misseri R, Gardner TA, Koch MO and
Rink RC: Transitional cell carcinoma of the bladder following augmentation cystoplasty for the neurogenic bladder.
J Urol 2004; 172: 1649.
North AC and Lakshmanan Y: Malignancy associated with the
use of intestinal segments in the urinary tract. Urol Oncol
2007; 25: 165.
Adams MC, Mitchell ME and Rink RC: Gastrocystoplasty: an
alternative solution to the problem of urological reconstruction in the severely compromised patient. J Urol 1988; 140:
1152.
Qiu H, Kordunskaya S and Yantiss RK: Transitional cell
carcinoma arising in the gastric remnant following gastrocystoplasty. Int J Surg Pathol 2003; 11: 143.
Esquena Fernandez S, Abascal JM, Tremps E and Morote J:
Gastric cancer in augmentation gastrocystoplasty. Urol Int
2005; 74: 368.
7.
Balachandra B, Swanson PE, Upton MP and Yeh MM: Adenocarcinoma arising in a gastrocystoplasty. J Clin Pathol
2007; 60: 85.
8. Baydar DE, Allan RW, Castellan M, Labbie A and Epstein JI:
Anaplastic signet ring cell carcinoma arising in gastrocystoplasty. Urology 2005; 65: 1226.
9. DeFoor W, Minevich E, Reeves D, Tackett L, Wacksman J and
Sheldon C: Gastrocystoplasty: long-term followup. J Urol
2003; 170: 1647.
10. Buson H, Castro Diaz D, Manivel C, Jessurun J, Dayanc M and
Gonzalez R: The development of tumors in experimental
gastroenterocystoplasty. J Urol 1993; 150: 730.
11. Close CE, Tekgul S, Ganesan GS, True LD and Mitchell ME:
Flow cytometry analysis of proliferative lesions at the gastrocystoplasty anastomosis. J Urol 2003; 169: 365.
EDITORIAL COMMENT
These authors present a retrospective case series of 4 patients in whom malignancy developed after gastrocystoplasty at 2 large institutions with a high volume of bladder
augmentations. In a 13-year period 119 patients underwent
gastrocystoplasty, of whom 3 had gastric adenocarcinoma
and 1 had poorly differentiated TCC. In 3 patients the tumor
was found at the gastrovesical anastomosis, while TCC was
disseminated throughout the bladder and augment. Tumors
were detected between 11 and 14 years after bladder augmentation. The authors recommend yearly endoscopic surveillance 10 years after enterocystoplasty, especially since
imaging and cytology do not seem sensitive enough to reveal
all malignancies.
The findings are certainly concerning, especially since, as
the authors mention, many patients do not receive regular
surveillance and, therefore, they are at increased risk. The 2
patients with negative cystoscopy 1 year before the tumor
developed illustrate the crucial importance of regular surveillance.
Therefore, the care of our patients does not end after the
successful operation but it must be extended to lifelong
observation. We have the responsibility to follow our patients and enroll them in a surveillance program. This is
especially important, considering the young age of our patients at surgery who need continuous care even after we
stop our practices. A nationwide data base could enlist all
patients and enroll them in an accepted surveillance protocol to ensure appropriate lifelong followup after enterocystoplasty.
Dominic Frimberger
Pediatric Urology
Childrens Hospital of Oklahoma
Oklahoma City, Oklahoma
The Use of Botulinum Toxin A Injection for the Management of

External Sphincter Dyssynergia in Neurologically Normal Children
Israel Franco,*, Lori Landau-Dyer, Ginger Isom-Batz, Therese Collett and Edward F. Reda
From the Section of Pediatric Urology, New York Medical College, Valhalla and Department of Urology, Lenox Hill Hospital (GIB),
New York, New York
Purpose: Botulinum toxin A has previously been used for neurogenic and nonneurogenic urgency and urge incontinence. We
evaluated the effects of sphincteric botulinum toxin A injection in a series of neurologically normal children with evidence of
external sphincter dyssynergia with various voiding problems documented by abnormal voiding electromyography as well as
voiding cystourethrography to assess its effectiveness for eliminating post-void residual urine.
Materials and Methods: We retrospectively reviewed the charts of 16 dysfunctional voiders who underwent botulinum
toxin A injection to the external sphincter between 2002 and 2006, including 1 to 3 injections in 14, 1 and 1, respectively. Of
19 injections 17 were performed with 300 U to the sphincter, while 2 of 19 were done with 200 U. Two patients also received
100 U injected into the detrusor. Mean patient age at surgery was 9.0 years (range 6 to 16). Preoperative clinical data were
recorded, including medications, electromyography, uroflowmetry with post-void residual urine, ultrasound and voiding
cystourethrography. Before botulinum toxin A injection medical therapies had failed in all patients, including -blockers in
100%, biofeedback in 100%, oxybutynin in 33% and tricyclics in 3 (20%). One patient was on intermittent catheterization. All
patients were refractory to bowel regimens and timed voiding. Postoperative parameters consisted of medications, symptoms
and post-void residual urine. In the 3 males the resolution of epididymitis symptoms and prevention of recurrence were
evidence of success.
Results: Before treatment patients experienced symptoms of urge incontinence (14 of 16), recurrent urinary tract infections
(66%), voiding postponement (45%) and epididymitis (3 of 16). All patients had external sphincter dyssynergia, as documented by preoperative electromyography or voiding cystourethrography. Average preoperative post-void residual urine was
107 cc (range 49 to 218). Two patients who underwent preoperative voiding cystourethrography had unilateral grade 1 reflux.
Of the 16 children 12 (75%) were dry at the first postoperative visit. The remaining 2 patients had decreased enuresis and
13 of 16 were dry at the second postoperative visit. The last patient became dry after treatment for attention deficit disorder
was initiated. Average initial postoperative post-void residual urine volume was 43 cc (range 0 to 141) and the average best
postoperative visit post-void residual urine was 8 cc (range 0 to 26). Uroflow data revealed no difference in uroflow before or
after injections. Neuropsychiatric problems were present in 9 of the 16 patients, including depression in 4, anxiety in 3 and
attention deficit disorder in 2.
Conclusions: Before our study in the pediatric literature doses between 50 and 100 U were used. We used a significantly
higher dose with increased efficacy and no increased morbidity. Endoscopic botulinum toxin A injection of the external
sphincter appears to be a safe and efficacious way to treat refractory nonneurogenic voiding dysfunction in children with
external sphincter dyssynergia. Long-term followup is necessary and repeat endoscopic injections may be required in select
patients.
Key Words: bladder; botulinum toxin type A; urinary incontinence; endoscopy; urinary bladder, overactive
otulinum toxin A was first discovered in 1897 by van

Ermegam.1 It has been used in neurology for severe
muscle spasticity for many years and by ophthalmologists for strabismus. Its use was first introduced into urology by Dykstra et al in 1988 for DSD in spinal cord injured
patients.2 Subsequent studies of sphincter spasticity in
adult patients had significant positive results, in particular
decreased PVR.3 8
The use of botulinum toxin A was first described as treatment for ESD in 1997 by Steinhardt et al in a neurologically
* Correspondence: 19 Bradhurst Ave., Suite 2575, Hawthorne,

New York 10532 (telephone: 914-493-8628; FAX: 914-493-8564;
e-mail: Isfranco@optonline.net).
Financial interest and/or other relationship with sanofi-aventis.
0022-5347/07/1784-1775/0
normal child.9 They injected a total of 20 U in 4 quadrants at

the external sphincter in a 7-year-old girl with recurrent
UTIs and wetting as well as dramatic urethral dilatation
due to severe ESD. After 18 months the child was infectionfree and dry. At the American Academy of Pediatrics, Section of Urology meeting in 2001 Maizels et al presented a
series of 20 females with lazy bladder and external sphincter
spasticity who were treated with botulinum toxin A on a
prospective basis.10 At 2 years of followup the patients had
had no recurrent UTIs compared with a pre-procedural average of 3 infections per year. However, voiding diaries
revealed no statistical differences in voided volumes or voiding intervals. Maizels et al favored botulinum toxin A injection with 50 U and concomitant urethral dilation. Their
results showed subjective improvements in daytime/night-
1775
Vol. 178, 1775-1780, October 2007

Printed in U.S.A.
DOI:10.1016/j.juro.2007.03.185
1776
BOTULINUM TOXIN A FOR EXTERNAL SPHINCTER DYSSYNERGIA
time urine flow and they postulated that intravesical pressure after botulinum toxin A was likely decreased. Although
the number of UTIs was reduced, Maizels et al did not
achieve total continence in the patients in their study. Although this study appears to have been flawed by the fact
that the children underwent urethral dilation along with
botulinum toxin A injections, making it impossible to determine which procedure caused the improvements, it further
strengthens the foundation for the use of botulinum toxin A
injections in neurologically normal children.
More recently a larger series of 20 patients presented by
Radojicic et al indicated that DSD treatment is clearly
helped by the use of botulinum toxin A injections in neurologically normal children.11 A recent report by Mokhless
et al indicated successful treatment in 10 children with what
the authors classified as nonneurogenic neurogenic bladder
dysfunction.12 Nine of the 10 children were on intermittent
catheterization, of whom 8 resumed normal urination after
treatment.
We evaluated whether treatment in children in whom
who all medical therapy as well as biofeedback had failed
would be helped by botulinum toxin A injections to the
external sphincter. In many of these children this was the
only viable treatment option left.
GlycoLax or mineral oil with or without sensors. All patients had been on a timed voiding regimen with or without
a vibratory watch.
The injection technique in males consisted of a 3.7Fr
Deflux needle passed through a 10Fr Storz offset endoscope. The external sphincter was visualized and the area
from the sphincter to the verumontanum was injected at the
3, 6, 9 and 12 oclock positions in equal amounts with diluted
botulinum toxin A. In females the cystoscope was placed in
the urethra. Using a 22 or 23 gauge needle we placed the
needle beside the urethra with the endoscope in position
with care taken to ensure that the needle was adjacent to
the urethra. We injected again in 4 quadrants as we pulled
the needle out.
Post-injection parameters that were evaluated included
medications, symptoms and PVR. PVR was measured using
a SDU-350xl ultrasound machine (Shimadzu, Kyoto, Japan)
with the built-in protocol to measure bladder volume. In the
3 males resolution of epididymitis symptoms and prevention
of recurrence were the parameters that indicated success.
The maximum botulinum toxin A dose was a total of 300 U
in almost all patients. The average dose injected was 8.85
U/kg (range 11 to 7.58).
RESULTS

From 2002 through 2006 there were 4,201 visits for urinary
incontinence (International Classification of Diseases, revision 9, 788.30) and 1,121 for urinary retention (International Classification of Diseases, revision 9, 788.20 and
788.21) to our offices. Of these patient visits we retrospectively identified the charts of 16 dysfunctional voiders who
underwent botulinum toxin A injection to the external
sphincter between 2002 and 2006, including 1 to 3 injections
in 14, 1 and 1, respectively. A total of 19 injection procedures
were performed, including 17 with 300 U to the sphincter
and 2 with 200 U. Two patients also received 100 U injected
into the detrusor. Mean patient age at surgery was 9.0 years
(range 6 to 16). Average followup was 21 months (range 1.5
to 39). Only 1 patient had a followup of less than 6 months.
Preoperative clinical data recorded in all patients including
medications, uroflowmetry with EMG, bladder ultrasound
with PVR calculations and VCUG.
The reasons to be considered a candidate for botulinum
toxin A injection were recurrent bouts of epididymo-orchitis
in males with evidence of ESD confirmed on uroflowmetry/EMG and validated on VCUG. In female patients the
primary reason for inclusion was recurrent UTIs with increased PVR, and evidence of ESD confirmed on uroflowmetry/EMG and validated on VCUG. External sphincter
detrusor-sphincter dyssynergia was defined as inappropriate contraction of the external sphincter along with concomitant detrusor contraction, as confirmed on uroflowmetry/
EMG and/or micturition cystography.
Before any patient received botulinum toxin A injection
all medical and behavioral therapies were exhausted. Medical therapy included -blockers in all 16 patients (100%) for
internal sphincter dyssynergia and/or urgency, biofeedback
in all 16 (100%), anticholinergics in 5 (33%) and tricyclics in
3 (20%). One patient was on intermittent catheterization
due to recurrent UTIs caused by persistently high PVR. All
patients were treated with a bowel program that included
Before treatment 14 of 16 patients experienced urge incontinence symptoms, 66% had recurrent UTIs, 45% had voiding postponement and 3 had epididymitis. All 16 patients
(100%) had ESD, as documented by preoperative EMG or
VCUG. Average preoperative PVR was 107 cc (range 49 to
218). Of the 16 children (75%) 12 were dry at the first
postoperative visit at week 2. Of the remaining 4 children all
except 1 had decreased enuresis, documented as a reduction
in the number of wetting accidents and/or the volume of
wetting. Of the 16 children 13 were dry at the second postoperative visit. Two of these 3 patients became dry after
imipramine was added to the treatment regimen to control
urge incontinence and the remaining patient became dry
after treatment for ADD was initiated.
All 3 males with recurrent epididymo-orchitis and testicular pain responded immediately with the resolution of pain
and no further episodes of epididymo-orchitis. One patient
experienced 2 additional episodes of epididymo-orchitis,
which occurred at least 6 months after botulinum toxin A
treatment. This was the same patient who was treated for
ADD. After the initiation of ADD treatment he remained
free of epididymo-orchitis for 1 year but it recently recurred
due to poor compliance with ADD medications. Average
initial postoperative PVR was 43 cc (range 0 to 141) and the
average best postoperative visit PVR was 8 cc (range 0 to 26)
(see table).
Uroflowmetry data after injection were not available on a
consistent basis for all patients, making statistical analysis
inconclusive. Neuropsychiatric problems were present in 9
of the 16 patients, including depression in 4, anxiety in 3 and
ADD in 2. The 2 patients who required multiple injections
have neuropsychiatric problems, that is significant problems
with depression in 1 and ADD in the other.
Initially after injections some patients experienced increased incontinence, which was transitory. Incontinence
was seen in children who had a tendency toward voiding
postponement. After injection they were unable to retain

PVR in all patients except males with epididymitis
PVR (ml)
Pt
UV
AP
AE
SW
SS
DB
CF
EM
NM
MB
KM
MJ
LB
Av
Students t test
p value
Before
Botulinum A
126.00
62.00
50.00
117.00
49.00
73.00
217.00
218.00
35.00
184.00
55.00
0
After
Botulinum A
9.00
7.00
5.00
6.00
8.00
16.00
5.00
0.00
7.00
26.00
0.00
0
Not available
107.8
8.0
0.00037
117.00
55.00
45.00
111.00
41.00
57.00
212.00
218.00
28.00
158.00
55.00
0
99.7
urine and experienced wetting at an increased rate due to

the lack of external sphincter contraction. This resolved
after the children were made aware that they could no
longer volitionally contract the external sphincter. We found
this to be a powerful tool to train voiding postponers, who in
general tend to show oppositional behavioral traits. No patient experienced any complaints of weakness that may have
been related to botulinum toxin A injections.
DISCUSSION
Botulinum toxin A is produced by the facultative anaerobe
Clostridium botulinum. Botulinum toxin A acts by inhibiting acetylcholine at the presynaptic cholinergic junction.
Inhibited acetylcholine release results in regionally decreased muscle contractility and muscle atrophy at the injection site. The chemical denervation that ensues is a reversible process since axons resprout in approximately 3 to 6
months. The toxin acts at the neuromuscular junction at the
external sphincter to block vesicle transport of acetylcholine
in essence producing chemical denervation. Clinical effects
begin within 5 to 7 days and are reversible because of terminal resprouting occurs within six months.13 The clinical
success of botulinum toxin A is supported by laboratory
research showing marked decreases in the release of labeled
norepinephrine and acetylcholine in botulinum toxin A injected rat bladders and urethras.14 While therapeutic effects
of inhibiting acetylcholine release is obvious, blocking norepinephrine release may provide clinical benefit by inhibiting sympathetic transmission in smooth muscle dyssynergia. This is why some of the patients whom we treated who
had combined internal and external dyssynergia responded
well to botulinum toxin A injections.
Botox A is the most widely used version for human
medicinal purposes in the United States. Investigational use
in animals is generally done with botulinum D toxin. First
approved for use in patients with strabismus, Botox remains
Food and Drug Administration approved only for strabismus
and cosmetic wrinkle therapy. Despite its limited Food and
Drug Administration approval practitioners of many specialties have embraced the toxin to counter various maladies. Most recently off-label use was popularized for migraine headaches and myofascial pain.
1777
Botulinum toxin A has also been used to inject the bladder to decrease detrusor hyperactivity. Studies in adult spinal cord injured patients15 and children with spina bifida16
indicated success with multiple injections throughout the
bladder floor. A recent study by Hoebeke et al indicated that
it is useful and safe in children with nonneurogenic OAB.17
Botulinum toxin A injections to the detrusor have been
performed for nonneurogenic OAB in symptomatic adults
with some success.
A drawback of this treatment is the need for re-treatment
since the probable underlying cause is not in the bladder but
elsewhere. On the other hand, botulinum toxin A injections
for internal sphincter dyssynergia could possibly be beneficial in the regimen that is necessary to treat OAB. Elimination of the dyssynergic voiding pattern could possibly help
eliminate the detrusor hypertrophy that is commonly associated with OAB. Botulinum toxin A injection produces reversible chemical sphincterotomy, which avoids a major surgical procedure with its attendant risks. These benefits of
botulinum toxin A injection were seen in several of our
patients who had no other treatment options. One girl was
on intermittent catheterization and unable to empty the
bladder, leaving a PVR of 250 cc at a time. This child is
completely dry, and she has had no accidents and voids to
completion a year and a half after injection. A boy was
offered augmentation cystoplasty to manage intractable
wetting and severe DSD, leading to chronic epididymitis. He
was treated with multiple injections, thereby avoiding a
urinary diversion procedure with its associated morbidity.
The use of botulinum toxin A in human pathology has
focused on end organ therapies. It was postulated that this
type of toxin isolation would decrease morbid side effects.
Doses of 50 to 300 U per dose have been documented in the
literature by different specialists. Riccabona et al used a
dose of 10 U/kg to a maximum of 300 U,18 while SchulteBaukloh et al used a dose of 12.5 U/kg to a maximum of
300 U.19 In the study by Radojicic et al the dose was set at
50 U for weights less than 40 kg, making the maximum dose
injected 1.25 U/kg.11 In the study by Mokhless et al the dose
of 50 U was used in children younger than 6 years, while all
others received a dose of 100 U.12 In 1 child in this study a
total of 300 U were injected in a 3-month period without side
effects. The rationale for this pattern of injection was not
described but it is obvious that this patient benefited from
the higher dose. The fact that the injections were performed
during 3 months before the effect of botulinum toxin A would
dissipate indicates that the higher treatment dose was necessary to achieve success. In the study by Radojicic et al PVR
was greater than 10 ml in 11 of 20 patients,11 while in our
cohort we found a PVR of greater than 10 ml in only 2 of 12.
The highest PVR noted was 26 cc, while in the study by
Radojicic et al PVR was as high as 100 cc. When we compared PVR data in the study by Mokhless et al we found that
average PVR at 6 months was 75 cc,12 while the average was
8 cc in our study. Superficially it appears that the higher
doses used in our study may be more efficacious when considering PVR data.
The lethal dose of botulinum toxin A is 39 to 49 U/kg.20 If
this dose is extrapolated to a 70 kg human, it would be
2,800 U. To achieve lethality the dose must be injected
intravascularly. Therapeutic botulinum toxin A is never injected intravascularly and it is used at much lower doses
than 3,000 U (range 50 to 300). Most urological applications
1778
do not exceed 300 U. In many situations the botulinum toxin

A dose in children has been based on the recommendation in
the neurological literature of a dose of 10 to 12 U/kg, which
in fact is not based on evidence.21
Obvious fears of toxicity due to higher doses are something that worries many investigators, leading them to use
the lower doses. The only side effects documented in the
urological literature are isolated muscle weakness, which
uniformly resolves within 1 to 3 months. None of these side
effects have been documented in children and the patients
had neurogenic voiding dysfunction, unlike our cohort. The
use by Chancellor and Smith of 200 U botulinum toxin A in
patients with multiple sclerosis without any neurological
complications22 is closer to our experience since many of our
patients were older and almost adult size.
The dose and injection volume used in this study were
much higher than in the previously reported studies by
Mokhless12 and Radojicic11 et al. We used an average dose of
8.85 U/kg or a maximum of 300 U per patient. The 300 U per
patient maximum that has been imposed in the urological
literature is an arbitrary value and in some of our patients
this would have been below the 10 to 12 U/kg dose in the
neurological literature as well as the doses used by Riccabona18 and Schulte-Baukloh19 et al in the pediatric urological literature.
We chose an endoscopic approach to botulinum toxin A
injection to maximize its placement at the external sphincter in males, while in the females we injected periurethrally
with the endoscope in the urethra to visualize the injection
and ensure ideal periurethral placement. The volume of
injection (10 cc) was higher than in other studies to maximize the dispersion of botulinum toxin A to a greater area
but was less than that used by Hoebeke et al in their patients.17 Again, this larger dispersion volume poses a theoretically higher risk of potential neurological side effects,
which were not identified in this series. This greater dispersion area could have beneficial effects on the bladder neck as
well as on the pelvic floor musculature. We found that there
were no issues with persistent incontinence that would be
associated with bladder neck dysfunction combined with
external sphincter relaxation. All leakage that occurred was
limited to voiding postponement activity and after this was
suppressed no patient continued to experience leakage. Also,
there were no episodes of stool incontinence, which would
indicate any overall laxity of the pelvic floor mechanism. It is
unfortunate that post-injection VCUG was unavailable on
these children (due to parental opposition) to test the hypothesis that botulinum toxin A can have an effect on norepinephrine based contractility of the bladder neck. It is
clear that some of these children had mixed internal and
external sphincter dyssynergia since many were on -blockers, in addition to anticholinergics.
It is possible that external sphincter denervation by botulinum toxin A can have an effect on the sensory feedback
loop of the micturition reflex. With this feedback inhibition
in effect the overactive guarding reflex is taken out of the
loop and the child can easily be retrained using biofeedback
to void appropriately. Whether this overactive guarding reflex is a result of detrusor instability due to recurrent UTIs
or learned behavior, or possibly to a link with dysfunction in
the frontal lobes, is irrelevant when considering the efficacy
of botulinum toxin A, allowing for proper sphincter relaxation. Possible blockage of sensory pathways could explain
the relief of pain that patients experienced in the study by

Zermann et al, in which they injected botulinum toxin A in
11 men with chronic prostatitis.23 They found a marked
decrease in PVR as well as improved flow rates but also a
reduction of pain.
Two of our patients who required multiple injections have
significant neuropsychiatric disorders, requiring medication.
Many other patients in this group had neuropsychiatric problems, requiring medications to treat the disorders. Urgency
and urge incontinence were the overwhelming symptoms seen
in this group with neuropsychiatric problems. When the neuropsychiatric problems were controlled, urgency disappeared
and the need to reflexively contract the external sphincter
would be abated, allowing easy retraining of voiding using
biofeedback techniques. The 2 more refractory cases responded
to botulinum toxin A injection with decreased PVR and improved flow rates after injection. Incontinence and PVR returned after the botulinum toxin A wore off or when they were
off their medications, indicating that the control of neuropsychiatric problems is an integral part of their care.
CONCLUSIONS
Before our study in the pediatric literature doses between 50
and 100 U were used. We used a significantly higher dose
and volume than reported in other series when injecting
botulinum toxin A into the sphincter with what appears to
be increased efficacy and no increased morbidity. Endoscopic
botulinum toxin A injection of the external sphincter appears to be a safe and efficacious way to treat refractory
nonneurogenic voiding dysfunction in children with ESD.
Long-term followup is necessary and repeat endoscopic injections may be required in select patients.
On the horizon it would be interesting to evaluate these
children in a randomized, prospective clinical trial. Our assumptions are drawn from subjective results and post-therapy PVR measurements. Ideally it would be useful to perform videourodynamics in all of these patients at uniform
intervals after injection. This would give us the opportunity
to standardize our results. Unfortunately there are few parents who will agree to invasive testing like videourodynamics after their child has shown improvement.

ADD
DSD
EMG
ESD
OAB
UTI
VCUG
attention deficit disorder

detrusor-sphincter dyssynergia
electromyography
external sphincter dyssynergia
overactive bladder
voiding cystourethrography
REFERENCES
1.
2.
van Ermengem E: Ueber einen neuen anaeroben Bacillus and

seine Beziehungen zum Botulisms. Ztsch Hyg Infekt 1897;
26: 1.
Dykstra DD, Sidi AA, Scott AB, Pagel JM and Goldish GD:
Effects of botulinum A toxin on detrusor-sphincter dyssynergia in spinal cord injury patients. J Urol 1988; 139: 919.

3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
15.
16.
17.
18.
19.
Dykstra DD and Sidi A: Treatment of detrusor-sphincter dyssynergia with botulinum A toxin: a double-blind study.
Arch Phys Med Rehabil 1990; 71: 24.
Schurch B, Hodler J and Rodie B: Botulinum A toxin as a
treatment of detrusor-sphincter dyssynergia in patients
with spinal cord injury: MRI controlled transperineal injections. J Neurol Neurosurg Psychiatry 1997; 63: 474.
Petit H, Wiart E, Gaujard E, LeBreton F, Ferriere JM,
Lagueny A et al: Botulinum A toxin treatment for detrusorsphincter dyssynergia in spinal cord disease. Spinal Cord
1998; 36: 91.
de Seze M, Petit H, Gallien P, de Seze MP, Joseph PA, Mazaux
JM et al: Botulinum A toxin and detrusor sphincter dyssynergia: a double-blind lidocaine-controlled study in 13
patients with spinal cord disease. Eur Urol 2002; 42: 56.
Phelan MW, Franks M, Somogyi GT, Yokoyama T, Fraser MO,
Lavelle JP et al: Botulinum toxin urethral sphincter injection to restore bladder emptying in men and women with
voiding dysfunction. J Urol 2001; 165: 1107.
Schulte-Baukloh H, Weiss C, Stolze T, Herholz J, Sturzebecher
B, Miller K et al: Botulinum-A toxin detrusor and sphincter
injection in treatment of overactive bladder syndrome: objective outcome and patient satisfaction. Eur Urol 2005; 48:
984.
Steinhardt GF, Naseer S and Cruz OA: Botulinum toxin: novel
treatment for dramatic urethral dilatation associated with
dysfunctional voiding. J Urol 1997; 158: 190.
Diaz-Soldano D, Maizels M, Houston JTB, Palmer JS, Richards I and Firlit C: Problems with urinary control. Presented at annual meeting of Section of Urology, American
Academy of Pediatrics, San Francisco, California, October
20 24, 2001.
Radojicic ZI, Perovic SV and Milic NM: Is it reasonable to treat
refractory voiding dysfunction in children with botulinum
A toxin? J Urol 2006; 176: 332.
Mokhless I, Gaafar S, Fouda K, Shafik M and Assem A: Botulinum A toxin urethral sphincter injection in children with
nonneurogenic neurogenic bladder. J Urol 2006; 176: 1767.
Smith CP and Chancellor MB: Emerging role of botulinum
toxin in the management of voiding dysfunction. J Urol
2004; 171: 2128.
Smith CP, Franks ME, McNeil BK, Ghosh R, de Groat WC,
Chancellor MB et al: Effect of botulinum toxin A on the
autonomic nervous system of the rat lower urinary tract.
J Urol 2003; 169: 1896.
Schurch B, Sthrer M, Kramer G, Schmid DM, Gaul G and
Hauri D: Botulinum-A toxin for treating detrusor hyperreflexia in spinal cord injured patients: a new alternative to
anticholinergic drugs? Preliminary results. J Urol 2000;
164: 692.
Schulte-Baukloh H, Michael T, Sturzebecher B and Knispel
HH: Botulinum-a toxin detrusor injection as a novel approach in the treatment of bladder spasticity in children
with neurogenic bladder. Eur Urol 2003; 44: 139.
Hoebeke P, De Caestecker K, Vande Walle J, Dehoorne J, Raes
A, Verleyen P et al: The effect of botulinum-A toxin in
incontinent children with therapy resistant overactive detrusor. J Urol 2006; 176: 328.
Riccabona M, Koen M, Schindler M, Goedele B, Pycha A,
Lusuardi L et al: Botulinum A toxin into the detrusor: a
safe alternative in the treatment of children with myelomingocele with detrusor hyperreflexia. J Urol 2004; 171:
845.
Schulte-Baukloh H, Michael T, Schobert J, Stolze T and
Knipsel H: Botulinum-a toxin in children detrusor hyperreflexia due to myelomingocoele: preliminary results Urology 2002; 59: 325.
1779
20.
Scott AB and Suzuki D: Systemic toxicity of botulinum toxin by

intramuscular injection in the monkey. Mov Disord 1988; 3:
333.
21. Jorgensen B: Editorial comment. J Urol 2006; 176: 330.
22. Chancellor M and Smith CP: One surgeons experience in 50
patients with botulinum toxin injection into the bladder
and urethra. J Urol, suppl., 2002; 167: 249, abstract 981.
23. Zermann D, Ishigooka M, Schubert J and Schmidt RA:
Perisphincteric injection of botulinum toxin type A. A treatment option for patients with chronic prostatic pain? Eur
Urol 2000; 38: 393.
EDITORIAL COMMENTS
These authors report excellent symptom improvement after
botulinum toxin A injection into the external sphincter of
neurologically normal children with ESD. This adds to the
growing literature that demonstrates the safe and effective
role of botulinum toxin for the treatment of various types of
voiding dysfunction. Whereas many earlier reports were
restricted to patients with neurological impairment, this
study confirms the applicability of this form of treatment to
those with more functional nonneurological disorders.
One may question though whether the dose of 300 U
botulinum toxin A used in this study is overkill. The authors
note that prior reports of sphincteric injection showed the
use of 50 to 100 U and results were inferior to those reported
in the current study. While it may be true that doses higher
than 50 U are necessary, the authors present no data to
suggest that 300 U are optimal. Perhaps 150 or 200 U may
achieve the same result. Given the high cost per unit of
Botox and the potential morbidity, especially in children,
intermediate doses should be compared and studied to
delineate the true optimal dose. Furthermore, many references that the authors cite to support the use of 300 U
involve multiple injections into the detrusor spread over a
wide area and they may not be comparable to sphincteric
injections.
Almost all patients had followup greater than 6 months
with the majority not requiring repeat injection. This is
contrary to what has been noted for sphincteric injections in
patients with neurogenic DSD, in whom the effect appears to
last 4 to 7 months. The authors hypothesize that botulinum
toxin A injection into the sphincter of these neurologically
normal patients may actually break the loop and allow the
restoration of normal voiding reflexes even after the botulinum toxin A wears off. While this is provocative, other recent
studies described a similar pattern in patients with idiopathic
voiding dysfunction with a number of patients achieving longterm benefit and not requiring repeat injection.1
Now that a positive response to this type of therapy has
been established, further study to delineate some of the
unanswered questions (optimal dose, optimal patient, etc)
should be performed to allow expansion of this therapy to
even more patients in a cost-effective, efficient manner.
Howard B. Goldman
Section of Voiding Dysfunction and Female Urology
Glickman Urological Institute
Cleveland Clinic Foundation
Cleveland, Ohio
1.
Kuo HC: Recovery of detrusor function after urethral botulinum

A toxin injection in patients with idiopathic low detrusor
contractility and voiding dysfunction. Urology 2007; 69: 57.
1780
These authors report the successful treatment of 1 group of

patients who were previously treated unsuccessfully. They
compare their results with those in similar studies. Again, it
is necessary to use uniform terminology and standardized
procedures in this field.
1) I would like to emphasize the importance of using
International Childrens Continence Society terminology.
The authors use (or cite other authors who use) various
terms, including DSD, nonneurogenic neurogenic bladder
dysfunction, dysfunctional voiding, etc. I think that the appropriate term according to the International Childrens
Continence Society proposal is dysfunctional voiding1 and
the use of other terms can lead to unnecessary confusion.
2) The authors emphasize the use of larger doses of Botox
than in previous series, making a direct connection to better
PVR results. PVR is a parameter that is highly dependent on
measurement conditions and the comparison may be significant if the measurement is performed in the same way and
under similar conditions.
3) There are 2 commercially available preparations of
botulinum A toxin on the world market, that is Dysport
and Botox. One U Botox is not bioequivalent to 1 U Dysport,
so it is not adequate to simply exchange 1 dose for the other.
In everyday practice we use 4 U Dysport as equivalent to
1 U Botox.2 When comparing results, it is necessary to
emphasize which preparations are compared as well as the
dilution and application method.
Successful treatment of these long-term patients must be
recognized and all of us who believe in this method of treatment must be happy. However, we must be careful when
unreservedly accepting findings (ours or those of others) that
do not result from prospective, randomized, double-blind studies. One is eager to easily accept opinions and facts confirming
ones ideas, whether true or false, no matter how hard and
honest ones endeavors are to find the scientific truth.
Zoran Radojicic
University Childrens Hospital
University of Belgrade
Belgrade, Serbia
1.
2.
Neveus T, von Gontard A, Hoebeke P, Hjalmas K, Bauer S,

Bower W et al: The standardization of terminology of lower
urinary tract function in children and adolescents: report
from the Standardization Committee of the International
Childrens Continence Society. J Urol 2006; 176: 314.
Ranoux D, Gury C, Fondarai J, Mas JL and Zuber M: Respective
potencies of Botox and Dysport: a double blind, randomized,
crossover study in cervical dystonia. J Neurol Neurosurg
Psychiatry 2002; 72: 459.
REPLY BY AUTHORS
The post-void residuals obtained in this study were garnered
in the same fashion in all patients. Patients drank fluids
as desired to the point that they had a maximal urge to

void and then a post-void residual measurement was done
immediately after voiding using the same ultrasound machine and technician each time. In no case did a child wait
more than 2 minutes to have the post-void residual measured. By being scrupulous with this we believe that we
eliminated the possibility of excessive upper tract drainage into the bladder and possibly kept the post-void residuals low. Our decreased post-void residuals were confirmed by the elimination of urinary tract infections in
this group of patients. If these patients had increased
post-void residuals they would have continued having
UTIs as did those in other studies.
We agree that the ideal dose has not been defined to treat
this condition. We may have used a dose that may have been
overkill as evidenced by the fact that 2 patients responded to
treatment with 200 U botulinum toxin A but more studies
need to be done in a prospective fashion to determine the
ideal dose. What is obvious from the literature is that 50 U
is not the appropriate dose either. We chose 300 U based on
prior reports of dosage regimens based on weight, and we
chose the higher end of the scale without adverse effects.
Reports of muscle weakness have only been seen in neurologically impaired patients, indicating that the populations
may be different. Whether the dose is spread over a large
area or is concentrated in a more localized area as we did in
the sphincter makes a difference. When multiple injections
(10 to 20) are made in the detrusor botulinum toxin A is
more likely injected intravascularly since each injection
risks injury to a blood vessel. Sphincteric injection in 4 sites
reduces the risk of intravascular injection, thereby reducing
the risk of dissemination of the toxin intravascularly and the
risk of systemic absorption.
Another reason that our results are superior to those of
other studies is most likely due to the intensive nature of the
adjuvant therapy used in our patients. It was obvious that if
the underlying neuropsychopathology was not adequately
addressed patients required repeat injection. Treating the
sphincteric dysfunction solely will only lead to limited success in some patients and abject failure in others as seen in
this and other studies. The sphincteric dyssynergia will persist (and we believe that in these patients it is a neurogenic
bladder disturbance according to the International Childrens Continence Society nomenclature recommendations)
and will not be cured. Correction of OAB in this group of
patients was an essential part of the overall treatment,
without which there could not be any form of sphincteric
reeducation. In the most recalcitrant patients with anticholinergic resistant OAB treatment with centrally active drugs
targeted to the behavioral problems was successful, and we
have become even more aggressive in treating the neuropsychiatric problems. Taking this approach in the 2 years since
the data were gathered for this study we have seen a decrease in the need to use botulinum toxin A.
Kidney II
Incidence Based Fetal Urological Counseling Using
the Virtual Pediatric Urology Registry: Importance of
Insignificant Fetal Pyelectasis (Sonographically Evident Renal Pelvis)
Antonio H. Chaviano,* Max Maizels, Elizabeth B. Yerkes, Earl Y. Cheng, Jennifer Hagerty,
Theresa Meyer and William E. Kaplan
From the Division of Urology, Childrens Memorial Hospital (AHC, MM, EBY, EYC, JH, TM, WEK) and Feinberg School of Medicine,
Northwestern University (AHC, MM, EBY, WEK), Chicago, Illinois
Purpose: Since 1985, counseling for fetal renal pelvic dilatation has been done to determine whether there is need for
newborn urological evaluation. This is likely if the anteroposterior width of the renal pelvis exceeds categorical cutoffs, ie 4
mm or greater before gestational age 33 weeks, or 7 mm or greater after 33 weeks. Cases below these categorical cutoffs are
deemed not to merit newborn testing. We examined our fetal registry to determine the incidence of urological pathology in
cases deemed not to merit newborn testing.
Materials and Methods: Since 1980, we have prospectively input fetal ultrasound and postnatal followup data into
customized Virtual Pediatric Urology Registry software. The Virtual Pediatric Urology Registry compares index case findings
against those archived in the registry and then outputs the incidence of newborn diagnoses. Diagnoses are grouped as having
limited or extensive urological care.
Results: The Virtual Pediatric Urology Registry has 1,128 cases registered and data on 2,292 fetal ultrasound studies that
were done between gestational ages 12 and 43 weeks (average SD 29.3 7). There are measurements of anteroposterior
pelvic width for 1,712 cases. Pediatric data include ultrasound for 2,596 cases, diuretic renal scan for 449 and voiding
cystourethrogram for 574. Surgery was done for renal/bladder obstruction or vesicoureteral reflux in 358 cases (32%). Mean
followup was 9.8 months old (range 1 day to 14 years). Of the total of 1,128 fetal cases 148 (13%) showed anteroposterior pelvic
width below categorical cutoffs, so that they were deemed not to merit newborn testing. However, the Virtual Pediatric
Urology Registry incidence based method identified that extensive urological care extended to 30 of these 148 cases (20%).
There were 31 cases identified at less than gestational age 33 weeks, which showed newborn urological pathology in 11 (35%),
including hydronephrosis in 4, surgery in 3, vesicoureteral reflux in 2, solitary kidney in 1 and death in 1. There were 117
cases identified at gestational age 33 weeks or greater, which showed newborn urological pathology in 19 (16%), including
vesicoureteral reflux in 8, hydronephrosis in 7 and surgery in 4.
Conclusions: We found that about 13% of cases of fetal renal pelvic dilatation were insignificant because the measurement
was below currently accepted cutoffs that merit postnatal followup. However, 20% of these cases in fact showed extensive
urological care needs. The Virtual Pediatric Urology Registry provides an array of likely newborn diagnoses in neonates.
Counseling by the incidence based method is more effective than by the current cutoff method.
Key Words: kidney, fetus, prenatal diagnosis, hydronephrosis, ultrasonography
etal sonography is the most efficient method of detecting urological anomalies and predicting the need for
immediate urological care in a newborn. Fetal sonographers and urologists examine ultrasound results for signs
of pyelectasis, otherwise known as renal pelvic dilatation,
which is the most common cause of urological abnormalities
in infants. Currently diagnosing fetal pyelectasis is based on
the measurement of AP pelvic width, which is easily sized
using ultrasound and by evaluating that value in regard to
patient GA. Cases are divided into significant and insignificant with significance defined as 4 mm or greater at GA less
than 33 weeks or 7 mm or greater at 33 weeks.1 A diagnosis
of significant pyelectasis results in close fetal monitoring, a

* Correspondence: Division of Urology, Box 24, Childrens Memorial Hospital, 2300 Childrens Plaza, Chicago, Illinois 60614 (FAX:
773-880-3339; e-mail: achaviano@childrensmemorial.org).
0022-5347/07/1784-1781/0
newborn urological examination and followup urological

care.
The problem with this current prognostic system has
become apparent since patients who were originally diagnosed with insignificant results show urological conditions
after birth.2 In our practice we noted a number of patients
who were initially diagnosed as normal, who then returned
to our pediatric care after birth as their conditions ultimately required intervention. We addressed this problem by
examining our load of fetal cases in the last 25 years using
the VIPUR database.
We examined the group of patients with insignificant
pyelectasis and followed their care after birth. We used
VIPUR to examine the percent of cases deemed insignificant
that subsequently needed postnatal urological care. We hypothesized that a substantial number of insignificant cases
would show significant anomalies. We believed that our
1781
Vol. 178, 1781-1785, October 2007

Printed in U.S.A.
DOI:10.1016/j.juro.2007.03.197
1782
INCIDENCE BASED FETAL UROLOGICAL COUNSELING
results would indicate the need for a change in cutoff criteria

for the significant and insignificant determinations.

Study Design
We examined newborn outcomes in cases of fetal pyelectasis
less than 4 mm at GA less than 33 weeks or less than 7 mm
at 33 weeks or greater. We used incidence based methodology to evaluate cases showing AP pelvis measurement levels
that are normally dismissed by the categorical cutoff counseling method as insignificant and, therefore, considered not
to merit newborn urological examination. The methods are
described.
Inclusion Criteria
All cases with prenatal ultrasound showing pyelectasis
without a specific diagnosis and with newborn followup data
available were examined for the postnatal outcome.
Exclusion Criteria
Cases were excluded if fetal ultrasound showed an anomaly
of which the outcome was already predictable, including 1) a
specific anomaly, eg ureterocele or multicystic kidney,
2) detected pyelectasis that was not an isolated finding, eg
associated with myelomeningocele, or 3) an abnormal kidney that was not dilated, eg echogenic. Fetal cases with
known aneuploidy were excluded. The normal twin of an
affected fetus was not included in this study.
Definitions
We underscore the use of the acronym SERP.3 Briefly, while
the extent of fetal pyelectasis is the current gold standard to
predict the newborn outcome, the term is problematic because there is no consensus on the thresholds for newborn
urology pathology. Furthermore, the term pyelectasis fosters misconceptions regarding expected poor newborn outcomes. Therefore, we substitute the neutral term SERP as
an acronym.
Significant fetal renal pelvic dilatation as significant
SERP represents 4 mm or greater at GA less than 33 weeks,
or 7 mm or greater at GA 33 weeks or greater. Insignificant
fetal renal pelvic dilatation as insignificant SERP represents less than 4 mm at GA less than 33 weeks, or less than
7 mm at GA 33 weeks or greater.
For this study cases with a clinical diagnosis of SERP
were grouped as newborns needing limited or extensive care.
That is, limited care involved patients who had a normal
urinary tract on ultrasound and did not undergo further
testing; while extensive urological care involved those with
abnormal findings on postnatal ultrasound that led to further radiological investigation of the urinary tract,4,5 as
described. We regard a normal urinary tract on ultrasound
as when the kidneys did not show the likelihood of obstruction, as in SFU grade 2 or less hydronephrosis,4 when they
are not echogenic, and when they are within the normal
length range. In this database duplication anomalies with a
symmetrical appearance of the renal pelvis of the affected
renal unit are considered normal.
Maternal-Fetal Counseling Methods

There is currently 1 prevailing course for counseling fetal
urological cases, namely using the categorical cutoff method.
Cases showing significant SERP are counseled to undergo
followup pediatric care based on these categorical cutoffs.
There is no dialogue regarding likely newborn diagnoses.
We offer an alternative approach to counseling using an
incidence based method. This involves presentation to the
family of the predicted incidences of newborn diagnoses by
GA and the extent of SERP. The families along with their
caregivers can make personal decisions based on this information. The incidence is newly able to be predicted because
of the new availability of the large VIPUR database, as
described.
Description of VIPUR Database

The registry includes prenatal and newborn data acquired
from 1980 to 2005. Clinical data were input into an Access
database and then loaded into customized VIPUR software6
capable of exacting the frequency distribution of archived
information. Using VIPUR data SERP measurements from
index cases are compared to archived matched data of cases
showing a similar extent of SERP and GA with a known
postnatal outcome. The result is an array of predicted urological newborn diagnoses and their corresponding urological care (fig. 1).
The VIPUR database currently comprises 1,128 cases.
There are 2,292 fetal ultrasound studies done at between GA
12 and 43 weeks (average 29.3), providing a total of 1,712
measurements of fetal SERP. Pediatric data include 3,619
records of ultrasound (2,596), diuretic renal scan (449) and
VCUG (574). Surgery was performed for renal obstruction,
bladder obstruction or VUR in 358 children (32%). Mean
followup is 9.8 months (range 1 day to 14 years).
Performance of Fetal Ultrasound

All fetal ultrasound was performed by a registered diagnostic medical sonographer who was supervised by a maternalfetal medicine specialist. GA was determined by the best
obstetric estimate.
FIG. 1. About 87% of fetal cases in VIPUR show significant SERP

since they are above categorical cutoffs, so that newborn testing is
suggested. Of the cases 13% show insignificant SERP since they are
below such cutoffs and, therefore, newborn testing is not recommended by conventional categorical cutoff method.
1783
FIG. 2. For cases of significant SERP VIPUR incidence based counseling method shows predicted frequency distribution of newborn
urological diagnoses. At GA 30 weeks right kidney shows 13 mm and left kidney shows 9 mm SERP (A), corresponding to array of newborn
diagnoses (B) and (C). Sol., solitary. HN, hydronephrosis. GU, genitourinary. misc., miscellaneous.
Postnatal Evaluation
All newborns underwent urine examination for infection and
followup ultrasound, typically by 10 days of life, unless fetal
findings showed bilateral hydronephrosis. In that case evaluation was initiated in the newborn period.
Children with SFU grade 3 or greater hydronephrosis
underwent 1 or several tests, including periodic sonography (done after 10 days of life), WTR (done after 1 month
of life),5 furosemide enhanced sonography (done after age
3 months) and/or VCUG. For children with SFU grade 2 or
less hydronephrosis parental choice directed the performance of VCUG.
Test Interpretation
Renal obstruction was diagnosed when WTR showed delayed,
furosemide induced 99Tc-mercaptoacetyltriglycine drainage from
the renal pelvis as UPJO or from the ureter as ureterovesical
junction obstruction. Typically these findings were accompanied by a post-WTR ultrasound showing pathological dilatation. Retrograde/antegrade pyelography was done as necessary to make the diagnosis before surgery was pursued.
Azotemia was diagnosed as increased serum creatinine
above the predicted normal value for age. Excretory urogram was done to evaluate complex anatomy.
to merit postnatal testing (fig. 1). Counseling for these cases

of significant SERP by the conventional cutoff method only
provides information on whether to perform newborn urological testing, while using the VIPUR database provides
additional information as an array of newborn urological
diagnoses. Figure 2 shows a case example.
Insignificant SERP
The focus of this research was the remaining 148 of 1,128
cases (13%) that showed insignificant SERP measurements
(fig. 3). The VIPUR incidence based method identified that
extensive urological care extended to 30 of these 148 cases
(20%), which would have been be deemed not to merit newborn testing by the categorical cutoff method.
Insignificant SERP at GA Less Than 33 Weeks
This group comprised 11 of 31 patients (35%) who had hydronephrosis (4), surgery (3), VUR (2) and solitary kidney
(1), and 1 who died (fig. 4, A). Figure 5 shows a sample case
of VIPUR counseling.
Insignificant SERP at GA More Than 33 Weeks
This group comprised 19 of 117 patients (16%) who had VUR
(8), hydronephrosis (7) or surgery (4) (fig. 4, B).
Indications for Surgery

Pediatric urological surgery was done to relieve obstruction
or correct VUR. Surgery to address obstruction included
pyeloplasty for UPJO, ureteral reimplantation in for UVJO
or VUR, or transurethral resection/ablation of posterior urethral valves. The performance of postnatal surgery was determined by current empirical pediatric urological practice.
This practice was exclusively based on the results of the
mentioned radiographic tests, urine infection status, and
parent and pediatric caregiver preferences. The performance of surgery was not equivalent to the need for surgery
since this cohort comprised of children who were clinically
well.7 Patients with hydronephrosis in whom obstruction
was diagnosed but whose families chose not to elect surgery
were categorized as showing extensive urological needs. Institutional review board approval was obtained for this retrospective case-control study.
RESULTS
Significant SERP
Of 1,128 cases 980 (87%) showed significant SERP measurements above categorical cutoff levels, and so were considered
FIG. 3. Use of VIPUR registry to predict likely newborn outcomes in

those with insignificant SERP. Of 148 cases of insignificant SERP
30 (20%) showed array of newborn urological pathologies. Misc.,
miscellaneous. HN Gr, hydronephrosis grade.
1784
FIG. 4. VIPUR method predicted incidence of newborn diagnoses as

array for insignificant SERP at GA less than 33 weeks (A) and 33
weeks or older (B). Exp., death, Sol. Kid, solitary kidney. HN GR,
hydronephrosis grade.
DISCUSSION
We found that 87% of about 1,000 fetal cases of SERP in
VIPUR are at or above the categorical cutoff levels, and so
are considered significant SERP (fig. 1). The remaining 13%
of cases are below these levels and they are considered
insignificant SERP. The latter patients would have previously been dismissed from requiring pediatric urological
care. However, we found that about 20% of the latter group
showed newborn urological problems (fig. 3).
Maternal-fetal ultrasound screening is becoming more
routine in the United States and it is the current standard in
Europe. These screenings show that renal collecting system
dilatation (SERP) affects about 2% (between 1% and 5%) of
3 million pregnancies annually in the United States.8 This
translates to approximately 60,000 newly diagnosed cases.
When identified, the clinical significance of prenatal dilatation of the renal pelvis can become a concern for the welfare
of newborns. While cues such as renal parenchyma echogenicity, ureteral dilatation, bladder distention and decreased
amniotic fluid level predict an increased risk of newborn
pathology, most cases do not show these findings. Most often
there is isolated SERP.
In about 1990 the method for counseling on these findings
was first introduced.1,9 This method establishes which fetal
cases warrant postnatal testing. Namely, patients counseled
to undergo newborn testing are above the categorical cutoff
levels of 4 mm or greater SERP at GA less than 33 weeks, or
7 mm or greater SERP at GA 33 weeks or greater. This
method dismisses cases below these levels. This approach
has been evaluated exhaustively and it has been shown to
correlate with newborn pathology.10 13
However, to our knowledge the specific nature and likely
incidence of the newborn condition has not been previously
defined. The availability of VIPUR, which houses about
1,000 cases of fetal diagnosis with postnatal followup, now
provides a new dimension to such counseling. Namely parents are informed of the array of likely newborn diagnoses
for cases viewed as significant and also for those previously
viewed as insignificant, and so dismissed from newborn
evaluation (figs. 2, B and C, 3 and 4).
Predicting the likely newborn outcomes using the VIPUR
incidence based method is possible beginning from the mid
second trimester, when the kidneys first become evident, to
term. The dynamic nature of the likely outcomes provided is
inherent in the VIPUR counseling method. For example, for
different SERP levels at the same GA the VIPUR method
showed different likely outcomes. That is, the higher the
SERP, the more likely that there would be pathological
newborn outcomes. Namely, at GA 30 weeks there was a
79% likely pathological newborn urological outcome for 13

mm of SERP, while there was only a 43% likely pathological
newborn urological for 9 mm SERP (fig. 2, B and C). This
concept was also observed in the cohort of patients with
insignificant SERP at GA less than 33 weeks who showed a
higher incidence of pathological newborn outcomes than
those at GA 33 weeks or greater (35% vs 16%). Furthermore,
using VIPUR to predict newborn outcomes is not restricted
to cases at our institution because its use is available in a
World Wide Web based format.9
However, to our knowledge the current literature is devoid of clinical outcomes for newborns whose findings show
SERP at or below the categorical cutoff levels of 4 mm or
greater and 7 mm or greater of the renal pelvis at and before
GA 33 weeks or greater, respectively. Because the risk-tobenefit ratio for postnatal testing has dictated that such
cases did not merit newborn followup, pediatric urological
followup is not routine. Thus, while most such patients
remained well, anecdotal reports showed that such fetal
findings could lead to newborn pathology.2
To date there has not been a tool available to identify the
risk of pathology in cases of insignificant SERP. VIPUR helps
resolve this deficiency and it makes available likely pediatric
outcomes to families with all levels of SERP, whether viewed
as meriting or not meriting postnatal testing (figs. 1 to 5).
Specifically in our study we examined the incidence of clinical
outcomes of insignificant SERP in VIPUR to aid in prenatal
counseling. In the database 13% of all SERP cases were
deemed insignificant because the level of distention was below
categorical cutoff measurements. Using an extensive database
with prenatal findings and postnatal outcomes is a step toward
honing in on the clinical outcome of minimal levels of SERP
findings. Because these insignificant levels of distention do not
meet established criteria for significant dilatation, they are
dismissed as not needing further postnatal testing. In the
literature small case series allude to the possibility that any
level of prenatal renal pelvis distention may result in significant postnatal pathology. However, the incidence of this likelihood has previously been evasive. We believe that our study
is the first to provide the incidence for the likely newborn
urological problem when insignificant SERP is detected.
Through our database we are able to provide an array of likely
diagnoses and their incidence when confronted with any level
of SERP by GA.
The VIPUR method is not applicable to cases of fetal diagnosis in which the postnatal outcome is already accepted. For
example, cases of renal parenchymal abnormalities, such as
FIG. 5. Example of VIPUR method of incidence based counseling

when categorical cutoff method showed that newborn testing was
not warranted at GA less than 33 weeks. There was 3 mm SERP at
GA 19 weeks (A). After birth diuretic renal scan revealed left UPJO
(data not shown) and retrograde pyelogram demonstrated UPJ
stricture just before pyeloplasty (B). Dist, distance.

bilateral echogenic kidney, are expected to show azotemia postnatally. Additionally, cases of fetal diagnosis of ureterocele are
likely to require surgery to address pediatric urological concerns, while patients who have a fetal diagnosis of multicystic
kidney show solitary renal function.
Because VIPUR comprises cases referred to our institution in the last 25 years, the cases housed are biased since
they do not include cases that were not referred. This bias
likely results in an overestimate of newborn urological pathology. Current studies are under way to enroll consecutive
cases of SERP prospectively in major obstetric centers city
wide in Chicago.
This study demonstrates the need to reevaluate the prognostic indicators of urological anomalies during gestation. A
more accurate assessment of newborn outcomes would help
prepare parents and physicians who are involved with the
future needs of the child. We have already reported that
cases of fetal SERP (less than GA 24 weeks) that show
enlarged kidney length and bladder size undergo extensive
pediatric care significantly more often than when such features are normal.3 We plan to examine the relationship of
such features to cases at GA 24 weeks or greater.
REFERENCES
1.
2.
3.
4.
5.
6.
7.
CONCLUSIONS
The measurement is below currently accepted cutoffs that
merit postnatal followup. However, we found that about 13%
of SERP cases are insignificant because 20% of them in fact
show newborn urological problems. VIPUR provides an array of likely diagnoses affecting newborns. Counseling by
the incidence based method is more effective than by the
current cutoff method.
8.
9.
10.

AP
GA
SERP
SFU
UPJ
UPJO
VCUG
VIPUR
VUR
WTR
anteroposterior
gestational age
sonographically evident renal pelvis
Society for Fetal Urology
ureteropelvic junction
UPJ obstruction
Virtual Pediatric Urology Registry
well tempered renogram
1785
11.
12.
13.
Corteville JE, Gray DL and Crane JP: Congenital hydronephrosis: Correlation of fetal ultrasonographic findings
with infant outcome. Am J Obstet Gynecol 1991; 165:
384.
Morin L, Cendron M, Crombleholme TM, Garmel SH, Klauber
GT and DAlton ME: Minimal hydronephrosis in the fetus:
clinical significance and implications for management.
J Urol 1996; 155: 2047.
Maizels M, Wang E, Sabbagha RE, Dinsmoor M, Seshadri R
and Ginsberg N: Late second trimester assessment of pyelectasis (SERP) to predict pediatric urological outcome is
improved by checking additional features. J Matern Fetal
Neonatal Med 2006; 19: 295.
Maizels M, Reisman EM, Flom LS, Nelson J, Fernbach S, Firlit
CF et al: Grading nephroureteral dilatation detected in the
first year of life: correlation with obstruction. J Urol 1992;
148: 609.
Conway JJ and Maizels M: The well tempered diuretic renogram: a standard method to examine the asymptomatic
neonate with hydronephrosis or hydroureteronephrosis.
J Nucl Med 1992; 33: 2047.
Pregnancy corner. Available at www.childrensurology.org. Accessed XXX.
Palmer LS, Maizels M, Cartwright PC, Fernbach SK and
Conway JJ: Surgery versus observation for managing
obstructive grade 3 to 4 unilateral hydronephrosis: a
report from the Society for Fetal Urology. J Urol 1998;
159: 222.
Arger PH, Coleman BG, Mintz MC, Snyder HP, Camardese T
and Arenson RL: Routine fetal genitourinary tract screening. Radiology 1985; 156: 485.
Lee RS, Cendron M, Kinnamon MS and Nguyen HT: Antenatal
hydronephrosis as a predictor of postnatal outcome: a metaanalysis. Pediatrics 2006; 118: 586.
Aviram R, Pomeran A, Sharon R, Beyth Y, Rathaus V and
Tepper R: The increase of renal pelvis dilatation in the
fetus and its significance. Ultrasound Obstet Gynecol 2000;
16: 60.
Sairam S, Al-Habib A, Sasson S and Thilaganathan B: Natural history of fetal hydronephrosis diagnosed on midtrimester ultrasound. Ultrasound Obstet Gynecol 2001;
17: 191.
Gloor JM: Management of prenatally detected fetal hydronephrosis. Mayo Clin Proc 1995; 70: 145.
Ismaili K, Hall M, Donner C, Thomas D, Vermeyelen D and
Avni FE: Results of systematic screening of minor degrees
of fetal renal pelvis dilatation in an unselected population.
Am J Obstet Gynecol 2003; 188: 242.
Renal Function Outcomes in Patients Treated With

Nephron Sparing Surgery for Bilateral Wilms Tumor
Dana W. Giel, Mark A. Williams, Deborah P. Jones, Andrew M. Davidoff and Jeffrey S. Dome
From the Divisions of Pediatric Urology (DG and MW) and Pediatric Nephrology (DJ), University of Tennessee and Departments of
Surgery (AD) and Oncology (JD), St. Jude Childrens Research Hospital, Memphis, Tennessee
Purpose: Management of bilateral Wilms tumor represents a particular challenge in the consideration of long-term renal
function for affected patients. Aggressive surgical resection to prevent recurrence must be balanced with the desire to
preserve renal function. We evaluated our institutional experience with nephrological outcomes in patients treated with
nephron sparing surgery for bilateral Wilms tumor.
Materials and Methods: We identified all patients with synchronous bilateral Wilms tumors presenting to St. Jude
Childrens Research Hospital between October 1987 and February 2004. We also included patients with Wilms tumor
involving a solitary kidney presenting during the same period. A total of 17 patients were identified who underwent nephron
sparing surgery, including 16 with bilateral tumors and 1 with tumor in a solitary kidney. Institutional review board approval
was obtained to retrospectively review records and analyze outcomes based on long-term renal function, hypertension,
proteinuria, need for dialysis and indications for renal transplantation.
Results: Eight of the 17 patients initially underwent bilateral nephron sparing surgery and 9 initially underwent a
combination of nephrectomy and contralateral nephron sparing surgery. Two patients were eventually rendered anephric
following further resections secondary to local recurrence. Before the initiation of therapy all patients had normal baseline
creatinine clearance, which was calculated using the Schwartz formula. At a median followup from diagnosis of 72 months
(range 15 to 207) 1 patient had renal insufficiency and another 3 had renal failure requiring dialysis. One of the 3 patients
on dialysis died of metastatic Wilms and 2 await renal transplantation. None of the remaining patients had evidence of
proteinuria. Ten of the 17 patients (58.8%) had hypertension at diagnosis and 9 (52.9%) required antihypertensive medications at the most recent followup. The overall survival rate in this group of patients was 88.2% with no evidence of disease
in survivors at the most recent followup.
Conclusions: When combined with adjuvant radiation and/or chemotherapy, nephron sparing surgery provides an opportunity to preserve renal function, while maintaining excellent long-term oncological outcomes for patients with bilateral
Wilms tumor.
Key Words: kidney, nephrectomy, Wilms tumor
anagement of BWT represents a distinct challenge

of maintaining long-term renal function while not
compromising overall cure rates. Since establishing
that tumor control and resection is more manageable following preoperative chemotherapy, thereby allowing NSS and
the preservation of renal function without sacrificing cure,1
patients with BWT have been successfully treated with a
more conservative approach. Recent reports from NWTS
indicate that the incidence of renal failure in patients with
BWT without associated anomalies is 11.5% at 20 years.2
While single institution series cannot replicate the number
of patients identified in larger cooperative group trials, these
series provide the benefits of consistency in surgical approach and completeness of followup. We evaluated our in-

Supported by United States Public Health Service Childhood Solid
Tumor Program Project Grant CA 23099, Cancer Center Support
Grant 21766 from the National Cancer Institute and American
Lebanese Syrian Associated Charities.
* Correspondence: 3606 Aurora Circle, Memphis, Tennessee 38111
(telephone: 901-327-2977; FAX: 901-287-4094; e-mail: danagiel@
gmail.com).
0022-5347/07/1784-1786/0
stitutional experience with long-term renal function outcomes in patients treated with NSS for BWT.

Following approval by the institutional review board we
identified all patients presenting to St. Jude Childrens Research Hospital with a diagnosis of synchronous BWT between October 1987 and February 2004. Patients presenting
with WT involving a solitary kidney in this period were also
identified. All patients included in this study underwent
NSS of at least 1 renal unit. Medical records, operative
reports and pathology findings were reviewed. Patient age at
diagnosis, associated congenital syndromes or anomalies,
hypertension at diagnosis and initial renal function were
documented. Renal function was determined by estimating
the glomerular filtration rate using the Schwartz formula,
creatinine clearance (k height in cm)/serum creatinine,
where k equals 0.55 and 0.45 for children older and younger
than 18 months, respectively.3 Clinical outcomes based on
most recent followup data were further analyzed. The variables assessed included long-term renal function, hyperten-
1786
Vol. 178, 1786-1790, October 2007

Printed in U.S.A.
DOI:10.1016/j.juro.2007.03.183
RENAL FUNCTION OUTCOMES AND BILATERAL WILMS TUMOR

sion, proteinuria, need for dialysis and indications for renal
transplantation.
All patients were categorized as having stage V tumors
according to the NWTS staging system. However, each individual renal unit was assigned a local tumor stage by the
treating oncologist. All patients received preoperative chemotherapy with vincristine and dactinomycin with or without doxorubicin based on the highest stage assigned to either kidney. Details regarding the chemotherapy regimens
used at our institution were previously described.4 Following initial treatment with preoperative chemotherapy patients underwent surgical excision with NSS performed on
at least 1 renal unit. All cases of NSS were attempted partial
nephrectomies rather than enucleation when possible, with
inclusion of a rim of grossly uninvolved renal parenchyma
along with the resected tumor. However, the challenging
size and multifocality of many tumors led to marginal resection in some patients. Further postoperative treatment, consisting of continued chemotherapy with or without radiation, was documented. Any additional surgical procedures
performed for recurrence were also documented.
RESULTS
A total of 17 patients were identified, including 16 with BWT
and 1 with WT in a solitary kidney (33 affected renal units).
Table 1 lists demographic data on the study population.
Median patient age at diagnosis was 22 months (range 5 to
54). Of the renal units 69.7% were staged as I or II, although
2 children presented with distant metastases.
All patients were initially treated with 2 or 3-drug chemotherapy depending on the assigned stage. Following neoadjuvant chemotherapy all children underwent NSS of at
least 1 renal unit. Eight of the 17 patients initially underwent bilateral NSS and 9 initially underwent a combination
of nephrectomy and contralateral NSS. As mentioned, the
tumors that we observed in this patient population were
often quite large, especially radiographically, and in our
experience patients who presented with smaller tumors also
TABLE 1
No. Pts (%)
Sex:
M
F
Race:
White
Black
Age (mos):
012
1324
2536
3748
4960
Stage (individual renal units):
I
II
III
Histology:
Favorable
Anaplasia
Anomalies:
WAGR
Cryptorchidism
Hypospadias
Hemihypertrophy
Other
10
7
(59)
(41)
8
9
(47)
(53)
6 (35.3)
4 (23.5)
3 (17.6)
2 (11.8)
2 (11.8)
2 (6.1)
21 (63.6)
10 (30.3)
14 (82.4)
3 (17.6)
1 (5.9)
5 (29.4)
1 (5.9)
1 (5.9)
1 (5.9)
1787
presented with multifocal lesions throughout the affected

kidney. Because of these factors, estimating the percent of
parenchymal involvement was not attempted as we considered this method to be highly subjective and not reproducible for documentation purposes. However, for NSS to be
attempted all surgeons had to agree that at least a third of
the uninvolved parenchyma could be spared. Formal occlusion of the renal vasculature was not routinely performed,
although temporary manual compression was used when
needed. Two patients were eventually rendered anephric
following further resections/completion bilateral nephrectomies secondary to local recurrence. Notably patients who
experienced recurrence had negative surgical margins following the initial surgical resection. Importantly NSS was
not done when tumor histology was discovered to be unfavorable. Table 1 lists a breakdown of tumor histology in our
patients. Following all surgical resections 23 of 33 renal
units were ultimately spared for a 70% renal salvage rate for
this patient population.
Before the initiation of any therapy all patients had normal baseline renal function, as estimated using the
Schwartz formula. At diagnosis 10 of 17 patients (58.8%)
presented with hypertension and all with hypertension required medications to control blood pressure. Although it
was not statistically significant, there was a trend toward an
increased incidence of hypertension at diagnosis in male
patients and in black patients. Six patients (35.3%) were
diagnosed with at least 1 GU anomaly, including WAGR in
1 and a WT1 abnormality in 1 that was not classified as
WAGR or Denys-Drash syndrome (table 1).
At a median followup from diagnosis of 72 months (range
15 to 207) 3 of 17 patients had ESRD requiring dialysis,
including the 2 rendered anephric. One of these patients
died of metastatic disease and the other 2 await renal transplantation. Another patient had mild renal insufficiency
with a serum creatinine of 1.6 mg/dl. Notably he was an
adolescent on creatine supplements and the nephrologists
seemed to think that this may have falsely elevated creatinine levels.
Table 2 lists the clinical details of patients with renal
function abnormalities. In all patients who had ESRD the
condition developed within 5 years of initial surgery. In
addition, all 3 patients with renal failure received abdominal radiation secondary to tumor stage or histology. Notably
patient 11 had discordant histology with focal anaplasia
only on the side of complete nephrectomy. The tumor resected via NSS was diagnosed with favorable histology on
careful review. All of the remaining 13 patients had normal
renal function using the Schwartz formula and none had any
evidence of proteinuria (see figure).
At most recent followup 10 of 17 patients (58.8%) had
hypertension, although this group with hypertension at followup did not consist of the same 10 patients who initially
presented with hypertension. In patients with hypertension
at followup there was no difference in gender or race with
regard to the incidence of hypertension. Nine of the 10
children with hypertension required medications to control
blood pressure.
Two of the 3 patients (66.6%) with ESRD presented with
GU anomalies, including the child with the WT1 anomaly.
When specifically considering the subset with GU anomalies, ESRD developed in 2 of the 6 patients (33.3%) with GU
anomalies in our study. Notably the child with WAGR has
1788

TABLE 2
Pt 7
Age (mos)
Race
Sex
Renal unit stage
Diagnosis
followup
hypertension
GU anomaly
Pt 9
Pt 11
Pt 17
12
Black
M
III/III
Yes
54
Black
F
III
Yes
22
Black
F
II/II
Yes
18
White
M
II/II
No
Dysplastic kidney
None
None
Surgery
Cryptorchidism, hypospadias,
WT1 deletion
Bilat NSS
Nephrectomy
NSS, completion
nephrectomy
Nephrectomy NSS
Histology
Favorable
Nephrectomy
NSS, salvage
NSS, completion
nephrectomy
Favorable
Favorable
Renal function
ESRD
ESRD
Focal anaplasia on
nephrectomy side,
favorable on NSS
side
ESRD
Mild chronic renal

insufficiency
All patients underwent abdominal external beam radiotherapy.
normal renal function at a current followup of 42 months

since diagnosis. With regard to the entire patient population
the overall survival rate was 88.2% with no evidence of
disease in all survivors at the most recent followup.
DISCUSSION
Although synchronous BWT accounts for only 5% to 7% of all
Wilms tumors, they represent a particular challenge to practitioners involved in treating these patients. There is a need
to balance the desires of positive oncological outcomes
achieved through aggressive surgical resection and adjunctive therapies with the need to maintain adequate long-term
renal function for this pediatric population of potential cancer survivors. Because studies have demonstrated that patients with BWT are at higher risk for renal failure,2,5 the
management of BWT has successfully evolved from primary
renal extirpation to renal preserving surgical approaches
augmented by preoperative chemotherapy. Bishop and Hope
first recommended a conservative surgical approach with

initial biopsy and chemotherapy before surgical resection.6
In 1979 NWTS-3 formally recommended initial biopsy and
chemotherapy before surgical resection of bilateral tumors.7
Since that time, multiple studies have confirmed the beneficial effects of neoadjuvant chemotherapy followed by conservative surgical resection in amenable, favorable histology
BWT cases,1,8 11 demonstrating that preoperative chemotherapy allows tumor shrinkage, leading to easier resection
with preservation of unaffected parenchyma as well as decreased tumor spillage. This approach has resulted in an
overall decrease in the incidence of renal failure in BWT, as
evidenced by the fact that the renal failure rate in NWTS-1
and NWTS-2 was 16.4% compared with 3.8% in NWTS-4.5
Renal failure is often multifactorial in its etiology, resulting from recurrent tumor requiring further resection, an
inadequate amount of remaining renal parenchyma and the
potential for hyperfiltration injury, nephrotoxic effects of
Changes in renal function, estimated by calculating creatinine clearance (CrCl) using Schwartz formula, (k height in cm)/serum creatinine
creatinine clearance, where k 0.55 and 0.45 for patients older and younger than 18 months, respectively. F/U, followup.

chemotherapy and radiation, and a congenital predisposition to renal function abnormalities.2,5 The 17.6% incidence
of renal failure in our patient population slightly exceeds the
overall incidence of renal failure in most recent NWTS estimates for patients with bilateral tumors,2 although the
small number in our single institution series and the higher
proportion of patients with GU anomalies in our series may
explain this difference. However, our findings are consistent
with those by Breslow et al, in that our patients with GU
anomalies appeared to be at higher risk for renal function
abnormalities.2 Although our single patient with WAGR
continues to have normal renal function, his followup is only
42 months and he may still be at risk for future renal
insufficiency. As Breslow et al noted, patients with WAGR or
other GU anomalies may experience renal failure relatively
late during followup. Therefore, we recommend that all patients treated for BWT, especially those with a predisposition toward future ESRD, should be followed in the long
term to monitor overall renal function. These patients
should be followed with routine blood pressure evaluations,
urinalysis to monitor proteinuria, serum creatinine measurement and glomerular filtration rate estimation, at a
minimum. Furthermore, they should be counseled to continue to be monitored even after they leave pediatric oncological followup and transition into care by our adult medical
colleagues.
Although the number of patients in our series is small
compared with that in larger cooperative trials, our single
institution series provides the benefit of consistency with
regard to surgical approach. All patients in our series were
operated on by the same surgical team and NSS was performed with a uniform technique. Furthermore, surgical
decisions (the decision to perform total nephrectomy vs NSS)
were made using the same criteria for each patient. The
renal salvage rate achieved in our series is comparable to
that in NWTS-4.11 Furthermore, our overall survival rates
are slightly better than those in other single institution
series8 10 and in larger cooperative group trials.1,11
While NSS leads to the possibility of positive margins,
which causes concern for potentially higher recurrence rates
from an oncological standpoint, it was shown that positive
surgical margins do not invariably lead to local recurrence,11
reflecting the chemosensitivity and radio sensitivity of WT.
Rather, the ability to leave the patient with functioning
renal parenchyma in those with favorable histology BWT
has given physicians the opportunity to maintain cure rates
without sacrificing overall renal function. Thus, we are no
longer forced to trade the disease of childhood cancer for that
of adult chronic renal failure.
1789

BWT
ESRD
GU
NSS
NWTS
WAGR
WT
bilateral WT
end stage renal disease
genitourinary
nephron sparing surgery
National Wilms Tumor Study
Wilms-aniridia syndrome
Wilms tumor
REFERENCES
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
Kumar R, Fitzgerald R and Breatnach F: Conservative surgical management of bilateral Wilms tumor: results of the
United Kingdom Childrens Cancer Study Group. J Urol
1998; 160: 1450.
Breslow NE, Collins AJ, Ritchey ML, Grigoriev YA, Peterson
SM and Green DM: End stage renal disease in patients
with Wilms tumor: results from the National Wilms Tumor
Study Group and the United States Renal Data System.
J Urol 2005; 174: 1972.
Schwartz GJ, Haycock GB, Edelmann DM and Spitzer A: A
simple estimate of glomerular filtration rate in children
derived from body length and plasma creatinine. Pediatrics
1976; 58: 259.
Dome JS, Liu T, Krasin M, Lott L, Shearer P, Daw NC et al:
Improved survival for patients with recurrent Wilms tumor: the experience at St. Jude Childrens Research Hospital. J Pediatr Hematol Oncol 2002; 24: 192.
Ritchey ML, Green DM, Thomas PR, Smith GR, Haase G,
Shochat S et al: Renal failure in Wilms tumor patients: a
report from the National Wilms Tumor Study Group. Med
Pediatr Oncol 1996; 26: 75.
Bishop HC and Hope JW: Bilateral Wilms tumor. J Pediatr
Surg 1966; 1: 476.
DAngio GJ, Breslow N, Beckwith JB, Evans A, Baum H,
deLorimier A et al: Treatment of Wilms tumor. Results of
the Third National Wilms Tumor Study. Cancer 1989; 64:
349.
Kubiak R, Gundeti M, Duffy PG, Ransley PG and Wilcox DT:
Renal function and outcome following salvage surgery for
bilateral Wilms tumor. J Pediatr Surg 2004; 39: 1667.
Cooper CS, Jaffe WI, Huff DS, Canning DA, Zderic SA,
Meadows AT et al: The role of renal salvage procedures for
bilateral Wilms tumor: a 15-year review. J Urol 2000; 163:
265.
Millar AJ, Davidson A, Rode H, Numanoglu A, Hartley PS,
Daubenton JD et al: Bilateral Wilms tumors: a singlecenter experience with 19 cases. J Pediatr Surg 2005; 40:
1289.
Horwitz JR, Ritchey ML, Moksness J, Breslow NE, Smith GR,
Thomas PR et al: Renal salvage procedures in patients with
synchronous bilateral Wilms tumors: a report from the
National Wilms Tumor Study Group. J Pediatr Surg 1996;
31: 1020.
EDITORIAL COMMENT
CONCLUSIONS
Although BWT can be challenging, preoperative chemotherapy often allows the preservation of renal parenchyma via
NSS. Our data confirm that NSS can be used successfully for
amenable BWT with favorable histology. With this conservative approach patients are offered a chance for cure as
well as the opportunity to maintain renal function into
adulthood.
These authors report their single center experience with

renal preservation strategies for children with BWT. Their
findings are consistent with those of prior group studies.
While the renal failure rate reported is somewhat higher
than that in NWTS reports (17.6%), this may be explained
by their limited sample size and a higher percent of patients
with GU anomalies, as indicated by the authors. They report
an impressive overall survival rate of approximately 88%.
Larger studies from the Society of Pediatric Oncology, and
1790
NWTS-3 and NWTS-4 indicated a 10-year overall survival of

69% and 78%, respectively. Again, sample size may explain
this difference.1
The real question now is what is next for children with
BWT. Outcomes (survival, chemotherapy toxicity and renal
function rates) for children with BWT have not been as good
as those in children with unilateral tumors of similar pathological findings and stage. A major limitation is the lack of
any scientific protocol for children with BWT. As we anticipate the release of a new BWT protocol from Childrens
Oncology Group, we can ponder some central questions regarding the treatment of these patients.
The usefulness of preoperative chemotherapy for patients
with BWT is widely accepted. However, the need for preoperative biopsy is controversial. Data suggest that preoperative biopsy is often inaccurate and the risk of misdiagnosis is
extremely low. In addition, biopsy results in an increase in
local tumor stage. Should we perform preoperative biopsies?
New biomarkers, such as chromosome 1p and 16q abnormalities, continue to enhance our ability to target patients
at risk for poor outcome. Will these biomarkers be helpful in
children with BWT?
Initial chemotherapy for children with BWT in Society of
Pediatric Oncology studies has consisted of 3 drug therapy
with vincristine, dactinomycin and doxorubicin. While limited data are available, it appears that relapse rates can be
lowered using a 3-drug regimen up front. This may decrease
the need for radiation therapy and its toxicity. Should
3-drug therapy up front be standard?
Post-nephrectomy histology after chemotherapy has been
shown to have prognostic value and it may help guide ther-
apy. Should we tailor post-nephrectomy treatment based on

pathological results?
Shamberger2 and Hamilton3 et al reported that often
when patients with BWT respond poorly to initial chemotherapy, prolonging the duration of therapy is not beneficial
due to increasing toxicity and the risk of renal insufficiency.
Early biopsy is recommended in patents with tumors that do
not respond to chemotherapy and presumably a subset of
patients would benefit from early nephrectomy. Furthermore, frequent assessments of resectability should be done
to decrease the duration of chemotherapy. What should the
timelines be for surgery?
No previous study has specifically focused on patients
with BWT. Hopefully we will soon have new protocol that
will address these questions and allow us to improve eventfree survival for patients with BWT.
Fernando Ferrer
Pediatric Urology
Connecticut Childrens Medical Center
University of Connecticut
Hartford, Connecticut
1.
2.
3.
Coppes MJ, de Kraker J, van Dijken PJ, Perry HJ, Delemarre

JF, Tournade MF et al: Bilateral Wilms tumor: long-term
survival and some epidemiological features. J Clin Oncol
1989; 7: 310.
Hamilton TE, Green DM, Perlman EJ, Argani P, Grundy P,
Ritchey ML et al: Bilateral Wilms tumor with anaplasia:
lessons from the National Wilms Tumor Study Group. J Pediatr Surg 2006; 41: 1641.
Shamberger RC, Ritchey ML, Hamilton TE and Green DM:
Bilateral Wilms tumors with progressive or nonresponsive
disease. J Pediatr Surg 2006; 41: 652.
Which is BetterRetroperitoneoscopic
or Laparoscopic Dismembered Pyeloplasty in Children?
Stephen J. Canon,* Venkata R. Jayanthi and Gregory J. Lowe
From the Section of Urology, Columbus Childrens Hospital, Ohio State University, Columbus, Ohio
Purpose: Groups at multiple institutions have documented the efficacy of minimally invasive repair of ureteropelvic junction
obstruction with a retroperitoneoscopic or laparoscopic approach. To our knowledge no group has compared the 2 operative
procedures directly at a single institution.
Materials and Methods: The records of 49 consecutive patients with a history of retroperitoneoscopic pyeloplasty or
transperitoneal laparoscopic pyeloplasty for ureteropelvic junction obstruction were reviewed retrospectively, of whom
29 underwent attempted retroperitoneoscopic pyeloplasty and 20 underwent laparoscopic pyeloplasty. Retroperitoneoscopic pyeloplasty cases were performed first in the series before changing to the laparoscopic pyeloplasty approach.
Retroperitoneoscopic pyeloplasty was performed using an anterolateral approach with retroperitoneal balloon distention. Laparoscopic pyeloplasty repair was performed using a transmesenteric approach for left ureteropelvic junction
obstruction or after right colon mobilization for right repairs. Dismembered pyeloplasty was performed over a stent
using 5-zero polydioxanone suture. Stents were placed antegrade or retrograde based on anatomy and presenting
symptoms. Parameters studied were patient age, operative time, postoperative analgesic requirement during hospitalization, hospital stay and success rate.
Results: No difference was observed between the 2 groups in patient age, success rate, hospital stay or analgesic narcotic
requirement. Average operative time for retroperitoneoscopic pyeloplasty was significantly longer than for laparoscopic
pyeloplasty (239.1 vs 184.8 minutes). Overall success rates were also statistically equivalent (25 of 27 retroperitoneoscopic and 19 of 19 laparoscopic pyeloplasties) with incomplete followup in 1 patient in the retroperitoneoscopic
pyeloplasty group and 1 in the laparoscopic pyeloplasty group. Three children, including 2 with retroperitoneoscopic and
1 with laparoscopic pyeloplasty, had transient urinary extravasation postoperatively, which was related to poorly
positioned stents. Five patients in the retroperitoneoscopic group and 1 in the laparoscopic group underwent balloon
dilation for indistinct but persistent postoperative flank pain with equivocal radiological findings. There were no major
complications following either technique.
Conclusions: In our experience no major difference exists between the retroperitoneoscopic and laparoscopic approaches
for correcting ureteropelvic junction obstruction. The difference in operative time likely reflects the learning curve for
laparoscopic suturing and dissection. Currently we prefer the laparoscopic approach because of the larger working space
for suturing, the perceived ease of antegrade stent placement and the subjective improvement in cosmetic outcome. The
2 techniques should be considered equal with regard to the successful correction of ureteropelvic junction obstruction.
Key Words: ureter, kidney, ureteral obstruction, laparoscopy
roups at several centers have investigated treatment

for UPJO in children using minimally invasive techniques within the last decade.110 The success rate
for UPJO treatment using a dismembered technique is between 83% and 100% at short-term followup (table 1). Regardless of the technique used advocates for minimally invasive UPJO treatment contend that there is less
postoperative pain and an improved cosmetic outcome without sacrificing the successful resolution of obstruction.
LP and RP have been performed for UPJO at different
institutions with successful and comparable results
(table 1).110 Proponents for RP or LP argue that advantages
exist for 1 of the respective approaches but to our knowledge
no direct comparison between the 2 techniques exists. RP
* Correspondence: Section of Urology, Columbus Childrens Hospital,

700 Childrens Dr., Columbus, OH 43205 (e-mail: canons@chi.
osu.edu).
0022-5347/07/1784-1791/0
was initially performed at our institution but, because of

concern for cosmetic outcomes and the limited working
space, our preference subsequently shifted to LP in the
latter half of this series. This evolution in operative technique provided a unique opportunity to compare the outcomes of the 2 approaches performed at the same institution.
METHODS
The records of 49 patients with UPJO who treated with RP
or LP at our institution were reviewed retrospectively. Retroperitoneoscopy was performed using an anterolateral approach (fig. 1). Briefly, a 10 mm incision was made in the
midaxillary line. After blunt dissection into the retroperitoneum a balloon dilator was placed within and the retroperitoneum was distended with 200 to 400 cc room air. A 10 mm
camera port was introduced and retroperitoneoscopy was
1791
Vol. 178, 1791-1795, October 2007

Printed in U.S.A.
DOI:10.1016/j.juro.2007.03.200
1792
RETROPERITONEOSCOPIC OR LAPAROSCOPIC DISMEMBERED PYELOPLASTY

TABLE 1. Transperitoneal and retroperitoneal laparoscopic dismembered pyeloplasty results
References
LP:
Tan and Roberts1
Schier2
Tan3
Casale et al4
Metzelder et al5
Reddy et al6
Kutikov et al7
Present series
RP:
Yeung et al8
El-Ghoneimi et al9
Bonnard et al10
Present series
No. Procedures
No. Conversions
No. Intraop Complications
% Treatment Success
No. No Followup
Mean Followup (mos)
1*
2
18
19
46
16
8
20
1
0
0
0
2
0
0
0
1
0
2
0
1
0
0
0
83
100
87
94
95
100
100
100
0
0
2
0
0
0
0
1
6
3
Less than 6
12
29
3
6
9.9
13
22
22
29
1
4
2
1
0
0
0
0
92
100
96
93
0
0
0
1
10
12.7
24
39.1
* Only 1 of 6 patients underwent dismembered pyeloplasty with a result not distinguishable from the total outcome of 83%.
Including only patients with dismembered pyeloplasty.
performed. Under direct vision 5 mm ports were placed and

dissection begun. Occasionally a fourth port was placed to
assist with peritoneal retraction if inadvertent pneumoperitoneum developed.
Laparoscopic procedures were performed by initially
placing a 5 mm umbilical camera port using an open technique. Two 3 to 5 mm working ports were then placed in the
epigastrium and in the ipsilateral lower quadrant. Right
repairs rarely needed a fourth port for liver retraction. In
each group percutaneous hitch sutures were used to lift and
fix the renal pelvis out of the renal fossa.
Dismembered pyeloplasty using interrupted or running
5-zero polydioxanone suture was performed in each group
over a ureteral stent that was placed antegrade or retrograde based on anatomy and presenting symptoms. Antegrade stenting was done by first percutaneously placing a 16
gauge angiocatheter, through which a guidewire was passed
down the ureter into the bladder. The stent was then passed
over the guidewire into the proper position.
Ureteral stents were removed 4 to 6 weeks postoperatively. The successful resolution of UPJO was based on the
resolution of presenting symptoms, ie flank pain, and improvement in hydronephrosis on followup renal ultrasound.
FIG. 1. Anterolateral approach to RP
In equivocal cases nuclear scintigraphy was performed. Success was defined as symptom resolution and/or improvement
in hydronephrosis without the need for reoperative pyeloplasty. Parameters evaluated were patient age, operative
time, postoperative analgesic use, hospital stay and success
rate. Postoperative analgesic requirements were quantified
and compared using the number of doses of oral and intravenous narcotic medications ingested while recovering in
the hospital that were given on an as needed basis, as
determined by the patient.
SPSS was used to perform statistical analysis of the RP
and LP groups. Chi-square analysis was used to compare
categorical values and Student t test was used to compare
numerical variables with p 0.05 considered statistically
significant.
RESULTS
The series consists of 49 patients with UPJO. The initial 29
patients with a mean age of 8.6 years (range 2 to 19, median
7) in the study underwent RP and the subsequent 20 patients with a mean age of 6.5 years (range 8 months to 15
years, median 6 years) underwent LP. There was a single
open conversion in a patient in the RP group who was
diagnosed with UPJO following blunt renal trauma and
underwent conversion to open repair due to extensive retroperitoneal adhesions.
Ultimately successful resolution of UPJO was observed in
25 of 27 patients (93%) in the RP group and in 19 of 19
patients (100%) in the LP group. Two patients had incomplete postoperative followup, including 1 in the RP group
and 1 in the LP group. Five patients in the RP group and 1
in the LP group underwent balloon dilation for indistinct yet
persistent postoperative flank pain with equivocal radiological findings (fig. 2). Five of the 6 patients undergoing balloon dilation achieved symptom resolution, while in 1 in the
RP group with persistent flank pain treatment was subsequently deemed to have failed. Two patients in the RP
group underwent reoperative pyeloplasty, including the 1
with balloon dilation. One patient had a widely patent
anastomosis but persistent high insertion, while the other
had small, fixed narrowing of the UPJ. Average followup
for the RP and LP groups was 39.1 and 9.9 months,
respectively (p 0.001).
1793
TABLE 3. Students t test comparison of first vs second half RP

and LP operative time
No. Pts
FIG. 2. Select images of patient 4, who underwent balloon dilation.

A, retrograde pyelogram before RP. B, postoperative retrograde
pyelogram showed minimal abnormality. C, balloon dilation demonstrated no identifiable waste.
Multiple parameters were evaluated and compared for

the 2 groups. The RP and LP groups were comparable in
patient age and sex (p 0.19 and 0.68, respectively). The
difference in hospital stay approached but did not attain
statistical significance at an average of 2.8 vs 2.1 days for
the RP vs LP groups (p 0.08, table 2). Average postoperative codeine use for the RP and LP groups was 4.3 and 3.1
doses per patient, respectively (p 0.23). The average postoperative morphine requirement for the RP and LP groups
was 1.9 and 1.0 doses per patient, respectively (p 0.056,
table 2). Although a slightly higher narcotic requirement
was observed in the RP group, this did not achieve statistical
significance. Average operative time for RP was significantly
longer than that for LP (239.1 vs 184.8 minutes, p 0.002,
table 2). There was no statistical difference in UPJO resolution between the 2 groups (p 0.225).
No major intraoperative or postoperative complications
occurred in either group and there was no difference in the
rate of minor complications (p 0.95). Three children had
transient urinary extravasation postoperatively, including 2 in the RP group and 1 in the LP group, which was
attributable to poor ureteral stent placement in each.
Associated urinary extravasation resolved with successful
endoscopic repositioning of the ureteral stent. One patient
with RP had postoperative pancreatitis, manifesting as
vomiting and mildly increased pancreatic enzymes on
postoperative day 4. These symptoms resolved with observation and intravenous fluids for a short period. The first
patient with LP had omental prolapse beside a Penrose
drain, which required repair 1 week after surgery. Another patient with LP experienced an immediate febrile
urinary tract infection, which resolved with observation
and antibiotic therapy. No small bowel obstructions or
bowel injuries occurred.
There was no improvement in mean operative time for RP
when comparing mean operative time for the initial and
latter halves of this group (table 3). However, there was
improvement in mean operative time for patients in the LP
group for the first and second halves (216.1 vs 156.7 minutes, p 0.003, table 3). Three patients who underwent
RP:
First 14*
Second 14
LP:
First 10
Second 10
p Value
Mean SD Operative Time (SE) (mins)
0.74
237.5 67.928 (19.609)

240.5 35.656 (9.529)
0.003
216.1 36.119 (12.040)

156.7 37.810 (11.956)
* Two patients who underwent stone retrieval were excluded.

One patient who underwent stone retrieval was excluded.
intraoperative flexible nephroscopy for stone retrieval during pyeloplasty, including 2 in the RP group and 1 in the LP
group, were excluded from operative time analysis.
DISCUSSION
Continued interest in minimally invasive treatment for
UPJO has inspired new questions about the optimal laparoscopic approach for treating this disease. Laparoscopic techniques through the retroperitoneum and peritoneal space
have allowed successful UPJO correction (table 1). Series
evaluating RP and series evaluating LP demonstrate comparable success rates in the short term (table 1).110 Although proponents for each of the 2 approaches contend that
advantages exist for the RP or the LP technique, to our
knowledge no direct comparison between the 2 methods
exists.
Retroperitoneoscopy is typically performed via a lateral
or posterior approach. Because we had little prior training
and experience, a modified anterolateral approach was used.
This approach allows one to access the UPJ anteriorly, enabling easy detection of accessory vessels, if present. Furthermore, for surgeons with little experience surgical exposure is quite similar to that of standard pyeloplasty via a
flank incision. Lastly, if open conversion is needed, as it was
required once in our series, the working ports can easily be
connected, leading to a standard flank incision. The main
disadvantage is that there is greater potential for inadvertent peritoneal leakage compared to other RP approaches,
although this did not preclude the successful completion of
repair in any of our patients.
The most important parameter to measure is successful
UPJO correction. There was no difference in the ultimate
outcome between RP and LP in the avoidance of reoperative
pyeloplasty. Early in the series some patients underwent
postoperative balloon dilation. In each case the procedure
was performed for vague but persistent postoperative flank
pain with equivocal radiographic findings. Some may argue
that these cases should in fact have been deemed failures. If
cases that underwent dilation were labeled failures, there
TABLE 2. Students t test comparison of RP and LP

Variable
p Value
Mean SD RP
Mean SD LP
Mean SD Totals
Age (yrs)
Hospital stay (days)
Followup time (mos)
Total codeine dose (1 mg/kg)
Total morphine dose (0.1 mg/kg)
Operative time (mins)
0.19
0.08
0.001
0.23
0.056
0.002
8.64.9
2.81.6
39.18.4
4.33.1
1.91.9
239.151.9
6.54.2
2.10.2
9.97.1
3.11.4
1.01.0
184.847.2
7.74.7
2.51.3
26.916.5
3.82.6
1.51.6
216.256.3
1794
still would be no statistical difference in outcomes between

the RP and LP groups (21 of 27 or 78% vs 18 of 19 or 95%,
p 0.14).
We believe that this seemingly large number of patients
undergoing balloon dilation was a result of our apprehension
with equivocal radiological findings in patients after a new
procedure. Indeed, if patients had undergone open repair
and presented in the immediate postoperative period with
such symptoms or radiographic findings, normally we would
have done nothing. However, due to the newness of the
procedure we were uncomfortable and purposefully cautious. Most of these children had relatively trivial retrograde
findings at dilation (fig. 2). Four of these 6 patients were
among the first 11 in the series. Five of 6 patients remained
symptom-free, which in conjunction with the trivial findings
on retrograde pyelography raises the question of whether all
of these patients truly needed balloon dilation in the first
place.
Lam et al recently compared the retroperitoneoscopic
technique for nephrectomy to that of the transperitoneal
technique.11 In their series 22 patients underwent laparoscopic nephrectomy and 10 underwent retroperitoneal nephrectomy. A lower median postoperative morphine equivalent requirement was observed in the retroperitoneal group
(110 vs 280 mcg/kg) and average operative time was decreased in the retroperitoneal group (65 vs 95 minutes).
Although there appeared to be a higher morphine requirement in the LP group, analgesic use analysis did not control
for the difference in hospital stay for LP vs RP (2 vs 1.2
days), which might explain the apparent difference. Another
factor to consider in this series is that transperitoneal laparoscopic nephrectomy was the initial technique performed in
the preliminary 22 patients. In the retroperitoneal group
the decreased operative time and the decreased conversion to open rate (5 conversions in the laparoscopic group
and none in the retroperitoneal group) reflected improvement in minimally invasive techniques. Thus, it is possible that the findings in this study were actually a function
of improved technique, rather than of any actual benefit to
the approach.
Two trends noted by Lam et al11 were also present in the
current study, including 1) the initial technique used in this
retrospective study required a longer operative time and 2)
FIG. 3. Scars after RP
FIG. 4. Scars after LP
patients undergoing the initial technique were noted to require more analgesic medications. In the current study the
RP technique required an average of 239.1 minutes, while
the LP technique required an average operative time of
184.8 minutes (p 0.002). Indeed, longer operative times in
the RP group in our series may be related to the fact that we
had little prior laparoscopic experience and we were refining
these techniques during the development of the series. Interestingly average operative time for the first 10 patients
with LP was significantly longer than for the second 10 with
LP (216.1 vs 156.7 minutes, p 0.003, table 3). However,
there was no statistical difference between second half RP
and first half LP operative times (240.5 and 216.1 minutes,
respectively, t 1.6, p 0.114), suggesting that the learning curve may not become level until after approximately 40
such minimally invasive procedures. These findings also
demonstrate the continued improvement in laparoscopic
technique with time at our institution.
Neither group had any major complications, such as
bowel or vascular injuries. Because the risk of intestinal or
great vessel injury with laparoscopic surgery is less than
1%,12,13 any discernible difference in such major complications between RP and LP would be difficult to delineate and
they will most likely remain theoretical. One patient in the
RP group had postoperative pancreatitis but no direct causal
relationship could be established. Also, the first patient in
the LP group experienced omental prolapse beside the indwelling drain left in 1 port site. Because of this complication, drain use in the LP group was subsequently abandoned.
Despite the lack of difference in objective parameters
measured in this series certain subjective benefits for LP
were observed in this series. Even in small children the
peritoneum provides a much larger working space for approaching the UPJO. This increased space allows much
greater ease with intracorporeal suturing than in the retroperitoneal approach. If antegrade stenting is chosen, it tends
to be less difficult transperitoneally because the larger working space allows a more precise site to be chosen for guidewire insertion so that a straight path to the ureter can be
established. Postoperatively urinary extravasation in this
series was directly related to poor stent placement and an

understanding of potential difficulties with stent position is
part of the learning curve for this procedure.
There was only 1 patient in the RP group who underwent
conversion to open pyeloplasty in our series. However, the
collective conversion rate for RP appeared higher than for
LP in the literature at 8 of 86 cases (9.3%) for RP vs 3 of 130
(2.3%) for LP (table 1). Massive hydronephrosis with limited
working space was often the reason for open conversion in
the RP series.8 10 In our experience limited working space
was not perceived to be a problem with LP.
Cosmesis is an issue that is hard to measure objectively.
We freely acknowledge the difficulty in objectively measuring cosmetic outcomes. Some may wonder why a 10 mm
camera port was used for pediatric RP. Most techniques for
RP access suggest a 10 mm incision with blunt dissection to
the retroperitoneum. Commonly smaller camera ports may
then be placed with the fascia closed next to the port to
prevent gas leakage. If a 10 mm skin incision is made, it
seems logical to use the largest port that will fit in that
incision to maximize optics and minimize gas leakage.
Figure 3 shows the scars following RP. LP allows the camera
port to be within the umbilicus, which is a natural scar. This
leaves only 2 and rarely 3 visible ports, that is the 3 to 5 mm
working ports (fig. 4).
CONCLUSIONS
Because of the trend toward minimally invasive urological
surgery for UPJO, consideration of the potential benefits of
RP and LP is warranted. We were unable to delineate any
major differences between the retroperitoneoscopic and
transperitoneal approaches to minimally invasive pyeloplasty, especially in terms of success. The data observed
following RP and LP in this series reinforce the idea that
minimally invasive treatment for UPJO in children is highly
successful in the short term with a low complication rate.
The answer to our original question of which is better (retroperitoneoscopic or laparoscopic dismembered pyeloplasty) in
children remains unclear at this time. Ultimately the choice
should rest on surgeon preference and experience. A controlled study of the 2 techniques with randomization would
be helpful for understanding whether any true difference
exists between RP and LP.

LP
RP
UPJ
UPJO
laparoscopic pyeloplasty
retroperitoneoscopic pyeloplasty
ureteropelvic junction
UPJ obstruction
REFERENCES
1.
Tan HL and Roberts JP: Laparoscopic dismembered pyeloplasty in children: preliminary results. Br J Urol 1996; 77:
909.
2. Schier F: Laparoscopic Anderson-Hynes pyeloplasty in children. Pediatr Surg Int 1998; 13: 497.
3. Tan HL: Laparoscopic Anderson-Hynes dismembered pyeloplasty in children. J Urol 1999; 162: 1045.
4. Casale P, Grady RW, Joyner BD, Zeltser IS, Figueroa TE and
Mitchell ME: Comparison of dismembered and nondismem-
5.
6.
7.
8.
9.
10.
11.
12.
13.
1795
bered laparoscopic pyeloplasty in the pediatric patient.

J Endourol 2004; 18: 875.
Metzelder ML, Schier F, Petersen C, Truss M and Ure BM:
Laparoscopic transabdominal pyeloplasty in children is feasible irrespective of age. J Urol 2006; 175: 688.
Reddy M, Nerli RB, Bashetty R and Ravish IR: Laparoscopic
dismembered pyeloplasty in children. J Urol 2005; 174:
700.
in the infant younger than 6 months: is it technically possible? J Urol 2006; 175: 1477.
Yeung CK, Tam YH, Sihoe JD, Lee KH and Liu KW: Retroperitoneoscopic dismembered pyeloplasty for pelvi-ureteric
junction obstruction in infants and children. BJU Int 2001;
87: 509.
El-Ghoneimi A, Farhat W, Bolduc S, Bagli D, McLorie G,
Aigrain Y et al: Laparoscopic dismembered pyeloplasty by a
retroperitoneal approach in children. BJU Int 2003; 92:
104.
Bonnard A, Fouquet V, Carricaburu E, Aigrain Y and ElGhoneimi A: Retroperitoneal laparoscopic versus open
pyeloplasty in children. J Urol 2005; 173: 1710.
Lam JP, MacKinlay GA, Munro FD and Aldridge LM: Endoscopic nephrectomy in children: is retro the way forward?
J Laparoendosc Adv Surg Tech A 2006; 16: 59.
Peters CA: Complications in pediatric urological laparoscopy:
results of a survey. J Urol 1996; 155: 1070.
Rassweiler J, Fornara P, Weber M, Janetschek G, Fahlenkamp
D, Henkel T et al: Laparoscopic nephrectomy: the experience of the laparoscopy working group of the German Urologic Association. J Urol 1998; 160: 18.
EDITORIAL COMMENT
These authors share with us their experience with the 2
approaches to LP. Their failure with the procedure was
defined as repeat pyeloplasty, although some patients required balloon dilation. They state that caution regarding
their equivocal radiological postoperative findings led them
to perform these secondary procedures. However, all of these
patients had vague yet persistent flank pain, which resolved
in 6 after dilation. Therefore, pain and not only equivocal
findings led to further intervention, which correctly should
be defined as failure since it added morbidity to the patients.
Although a causal relationship could not be identified in
the child with pancreatitis and it was dismissed, one must
keep an open mind that injury may have occurred during
surgery. Cautery injury? Blunt dissection injury? Retraction? Or was it drug induced? I believe that it teaches us to
be cautious and aware of our instruments and intraoperative surroundings that may be hiding just off the screen.
If the surgical outcome is equal compared to that of gold
standard open pyeloplasty, who really cares about the incisions. I agree with their conclusion that the approach truly
is surgeon preference. Pick the approach that gives your
patients the best possible operation that you can give them.
Better yet, pick the operation that you would have performed on your loved one if that surgeon had the same skill
sets as you.
Pasquale Casale
Miscellaneous
Corporeal Sparing Dismembered Clitoroplasty:
An Alternative Technique for Feminizing Genitoplasty
Joo L. Pippi Salle, Luis P. Braga, Nicanor Macedo, Nicolino Rosito and Darius Bagli
From the Division of Urology, Hospital for Sick Children, Toronto, Ontario, Canada
Purpose: Management for clitoral enlargement remains controversial. New understanding of clitoral function stimulated a
search for more conservative surgical approaches, such as recession or partial resection. However, these techniques risk
decreasing clitoral sensation or causing painful erections. Moreover, irreversibility continues to be the principal problem that
fuels patient, surgeon and societal anxiety in the management of this challenging developmental issue. We describe a new
technique, corporeal sparing dismembered clitoroplasty, that dismembers the corporeal bodies and preserves all clitoral
structures.
Materials and Methods: After obtaining full informed consent and institutional review board approval 8 consecutive
patients with clitoral enlargement underwent corporeal sparing dismembered clitoroplasty. Five girls had congenital adrenal
hyperplasia (Prader IV and V in 4 and 1, respectively), 1 had ovotesticular disorder of sexual differentiation and 2 had partial
androgen insensitivity syndrome. One pubertal girl was tested with warm, cold and pain clitoral stimulation before and after
surgery. For the clitoroplasty technique the glans and its neurovascular bundles are dissected from the corpora. The isolated
corpus is then completely divided starting at the bifurcation. Each separated hemicorpus is rotated inferior and lateral, to be
placed inside the labial scrotal folds. The glans is reduced by superficial excision of its epithelium and fixed to the pubic
attachments. Labia minora are constructed with preputial Byars flaps. Labioplasty and vaginoplasty are then routinely
performed.
Results: Eight patients 6 months to 13 years old underwent this procedure. Followup was 6 to 12 months. All patients
recovered well from surgery without early complications. The initial cosmetic result was good in all girls. The hemicorpora
were easily palpated inside their labia majora pouches, which retained the desired cosmetic appearance following feminizing
genitoplasty. All glans clitoris were preserved. The teenaged patient does not report painful erections. She has maintained
clitoral sensation and is satisfied with the cosmetic result.
Conclusions: Conservative reconfiguration of the female genitalia without removing genital structures is feasible in girls
with clitoral enlargement. The cosmetic appearance of the genitalia is acceptable, at least to the surgeon and parents, in that
the enlarged clitoris is hidden. The physiological consequences of the current operation and any surgery in the future to
reverse it are unknown. With these aspects in mind we believe that corporeal sparing dismembered clitoroplasty should be
incorporated into the armamentarium of surgeons involved in the treatment of clitoral enlargement and presented as an
option for feminizing genitoplasty.
Key Words: abnormalities, clitoris, genitalia, reconstructive surgical procedures
anagement of CE remains controversial because

of the nature of intervention in a structure so
vital to psychological body image and gender.
Initial efforts to correct CE consisted of partial or total
clitoridectomy based on the belief that permanent reduction of the enlarged clitoris was necessary to treat or
prevent gender dysphoria.1 However, understanding that
the clitoris has a crucial role in the development of female
sexuality has stimulated a search for more conservative
surgical approaches. This prompted the development of
techniques to bury or recess the clitoris.2,3 Unfortunately
fixation of the enlarged clitoris to the peripubic attachments
may lead to painful erections during sexual arousal.4 To avoid
this complication strategies for reduction clitoroplasty have
been described, including removal of the corporeal erectile
tissue with preservation of the neurovascular bundle to the
glans.5,6 Unfortunately these techniques risk decreasing clito-
Study received institutional ethics board approval.
0022-5347/07/1784-1796/0
ral sensation.7,8 We agree with others that maintaining all

clitoral structures is important to optimize postoperative clitoral sensation as well as minimize psychological trauma.9
Moreover, irreversibility continues to be the principal problem that fuels patient, surgeon and societal anxiety in the
management of this challenging developmental issue. In an
effort to optimize psychological and sexual function as well
as decrease anxiety surrounding the difficult surgical decision that patients and their families must make we developed a new technique, CSDC, that dismembers the corporeal
bodies and preserves all clitoral structures.

Following extended discussion of all available alternatives
and with full informed surgical consent 8 consecutive patients with clitoral enlargement underwent CSDC for feminizing genitoplasty. Institutional ethics board approval for
chart review was obtained. Furthermore, all patients inde-
1796
Vol. 178, 1796-1801, October 2007

Printed in U.S.A.
DOI:10.1016/j.juro.2007.03.167
CORPOREAL SPARING DISMEMBERED CLITOROPLASTY
1797
Size (cm)
Pt
No.Age
16 Mos
218 Mos
36 Mos
413 Yrs
56 Mos
614 Mos
712 Mos
810 Mos
Diagnosis
Clitoris
UGS
Prader
Vaginoplasty
Result
CAH
XY, ovarian/testicular DSD
CAH
XYDSD
CAH
CAH
XYDSD
CAH
3.5
3
3.5
4.5
3
3
3
3
3
2.5
3.5
3
3
2.7
3.5
2.0
4
4
5
4
4
4
4
4
PUM UF
PUM UF FF
Clitoroplasty vulvoplasty
PUM UF FF
PUM UF FF
PUM UF FF
PUM UF FF
PUM FF
Mild introital stenosis

Good
Good
Good
Good
Good
Good
Good
pendently consented to have their results published. All

except 1 patient had never previously undergone surgery. A
13-year-old girl with partial androgen insensitivity had undergone bilateral orchiectomy and introitoplasty 6 months
and 9 years previously, respectively. This patient was cooperative and accepted testing to evaluate preoperative and
postoperative clitoral sensation using cold, warm and vibration stimulation. The table lists patient characteristics.
Clitoroplasty Technique
The clitoris is degloved and the neurovascular bundles of the
glans are dissected off of the corpora cavernosa (fig. 1). This
dissection should start as ventral as possible, entering the
distinct avascular plane that exists between the albuginea
and the first layer of Bucks fascia. If the correct plane is
entered, it is possible to separate the neurovascular bundle
and the entire glans from the corpora without much bleeding
(fig. 2). The corporeal bodies are dissected down to their
bifurcation. Starting at the level of the bifurcation the corpora are divided with sharp dissection, entering a tightly
defined plane that exists in the midline between the 2 hemicorpora. Rigorous attention with sharp dissection in this
midline plane of the corpora accomplishes separation without significant bleeding (fig. 3). The lateral edges of the
opened albuginea of each hemicorpora are approximated
using fine absorbable sutures (fig. 4). A dartos pouch, similar
to that used for orchiopexy, is constructed in the labia majora. Each hemicorpus is rotated lateral and inferior to be
placed inside the respective labial pouch (fig. 5).
Glans reduction is then performed. This is accomplished
by superficial excision of the epithelium of the glanular
groove to stimulate future adhesions to the prepuce. An
effort is made to excise as little glans tissue as possible, only
FIG. 1. Outline of incisions to deglove clitoris
partially de-epithelializing it. The glans clitoris is then fixed

to the pubic bone approximately 1 cm above the original
bifurcation of the corpora. The prepuce is incised in the
midline (Byars flaps) to reconstruct the labia minora. It is
important to limit this incision, leaving enough intact skin
to produce an anatomical preputial clitoral hood. Vaginoplasty is performed by partial UGS mobilization. The posterior vaginal wall is constructed using a combination of an
inverted U cutaneous flap for the posterior medial aspect
(FF)10 and the redundant tissue from the divided UGS at the
6 and 12 oclock positions, as described by Rink.11 The latter
is rotated latero-inferior and used to build and enlarge the
lateral aspects of the vagina and introitus. Vulvoplasty is
then completed by bringing the lower aspect of the reconstructed labia minora to the lateral aspect of the exteriorized
vagina (fig. 6).
RESULTS
Eight patients 6 months to 13 years old underwent CSDC, as
described. Mean followup was 12 months (range 3 to 16). All
patients had an uneventful postoperative recovery and none
had postoperative bleeding or hematoma. The 2 hemicorpora
were easily palpated inside the labia majora pouches, which
retained the desired cosmetic appearance following feminizing genitoplasty. All glans clitoris were preserved. In 7 patients vaginoplasty was performed using PUM. One severely
virilized girl with CAH-Prader 5 and a hypoplastic high
vagina underwent CSDC and reductive vulvoplasty alone,
leaving vaginoplasty to be performed later (fig. 7). The 18month-old patient with ovotesticular DSD underwent laparoscopy and bilateral gonadal frozen section biopsy before
CSDC and vaginoplasty. Pathological examination confirmed the diagnosis of bilateral ovotestis. Another 12month-old girl with partial androgen insensitivity underwent
prior bilateral orchiectomy and pathological examination was
consistent with bilateral immature testes. This patient underwent CSDC and PUM of the 4 cm vagina located 2.5 cm
FIG. 2. Glans clitoris and its neurovascular bundle are separated

from corpora cavernosa. Dissection at tunica albugineal level allows
separation without much bleeding.
1798
FIG. 3. Two hemicorpora are separated in midline, starting at bifurcation.
FIG. 5. Each hemicorpus is rotated lateral and inferior to be placed

inside respective labial pouches.
DISCUSSION
from the bladder neck. In this particular patient partial
mobilization of the UGS would not allow the vagina to reach
the perineum, requiring division and separation to be
brought down. Its anterior and posterior walls were constructed using UGS and FFs, respectively. This maneuver
allowed the reconstructed vagina to reach the perineum
without tension (fig. 8). The 13-year-old pubertal patient
was tested postoperatively. She denied painful engorgement
of the labia or decreased genital sensitivity after surgery and
seemed to be satisfied with the cosmetic result. She remained on periodical vaginal dilation, which was discontinued after she became sexually active (fig. 9). Our first patient underwent posterior vaginal wall construction using
the redundant UGS exclusively, as previously described.12,13
This patient had introital stenosis and required introitoplasty because she retained a significant amount of urine in
the vagina. This procedure was successfully performed 1
year after CSDC. Following this case we started using a
combination of FFs and UGS flaps to construct the posterior
vaginal wall and enlarge the introitus, avoiding stenosis in
the remaining patients. The table lists patient characteristics, technical surgical details and results.
Surgical treatment in infants with CE remains controversial. There are also controversies regarding the time of and
need for clitoral reduction.14 However, leaving a grossly
enlarged clitoris untouched during childhood underscores
the psychological impact that this situation can cause to the
untreated child.13,15 Recent consensus statement on the
management of intersex disorders suggests that cosmetic
surgery in girls with severe virilization (Prader III to V)
should be performed in the first year of life, when appropriate, in conjunction with common UGS repair.16 On the other
hand, there are a number of publications reporting a loss of
sensation and decreased ability to achieve orgasm in women
who underwent various techniques for clitoroplasty.7,16,17
Although many of these complaints are related to vaginoplasty rather than to clitoral surgery, there remains a significant group of patients with decreased sensitivity in the
clitoris following clitoroplasty.7 Moreover, up to 5% of patients with CAH may have gender dysphoria later in life and
wish that they could revert decisions made by parents and
caregivers earlier in their lives.18
Although fully informed consent was obtained that explained the potential risks of clitoral surgery, in this report
FIG. 4. Opened albugineal layer of each hemicorpus is approximated with fine absorbable suture to prevent bleeding.
FIG. 6. Posterior vaginal wall is constructed using combination of

inverted U cutaneous flap (FF) and redundant UGS tissue (UGS
flap). Vulvoplasty is then completed.
1799
FIG. 7. Preoperative (left), transoperative (middle) and postoperative (right) views of severely virilized female with CAH and Prader 5.
Despite enlarged clitoris good cosmetic result was achieved with corporeal preservation.
we consciously do not present a detailed discussion about

gender identity issues, indications and timing for clitoral
surgery. We agree with others that irreversible genital surgery should be avoided and we present a conservative alternative technique.14,19 Most contemporary clitoroplasty techniques consist of excising the erectile tissue of the enlarged
clitoris. The consequence of such removal is largely unknown. It is possible that, as in males, clitoral sensation is
enhanced during erection of the corpora. Our technique attempts to spare this erectile tissue, which may have a role in
genital sensation. Although remarkable dissection is done
during this procedure, we believe that by doing the procedure correctly and respecting the tissue planes most clitoral
components will likely have their enervation and sensation
preserved. Starting the separation of the neurovascular bundle as ventral as possible, as routinely done in cases of
hypospadias, is an important step to minimize injury to
nerves.20 Dissection similar to the proposed clitoroplasty
technique is performed in some males with epispadias, in
whom the corpora are completely disassembled. We have
followed some of these adolescent males with epispadias who
underwent major penile reconstruction and maintained unchanged sensation as well as the ability to achieve erection,
ejaculation and orgasm. We believe that the technique presented has the potential to preserve clitoral sensitivity, similar to males with comparable interventions.
There is a possibility that these operated females could

have painful erections during sexual arousal. However, unlike previously described clitoral recession techniques in
which the clitoris was sutured to the pubis, we keep the 2
hemicorpora relatively free inside the labia majora. We hope
that this will permit a certain mobility of the corpora and
prevent painful erections. Furthermore, one could argue
that unpleasant labial engorgement could occur during clitoral erection. If this is the case, there would still be the
option of removing the 2 hemicorpora through labial incisions with the advantage of including the patient in this
difficult decision making process.
Moreover, irreversibility continues to be a dilemma for
patients, their families and the multidisciplinary team involved in their care. Using the preserved erectile tissue
would be potentially feasible in patients with gender dysphoria who desire phallic reconstruction.
Finally, we believe that it is important from the psychological standpoint that all clitoral parts are still in place but
simply redistributed in the perineum. This aspect may have
been underscored in importance compared to other procedures because some patients may manifest castration feelings after the removal of genital tissue.
We acknowledge that our technique still has limited,
short followup for drawing definitive conclusions and only 1
pubertal patient was suitable for objective assessment. How-
FIG. 8. View of 18-month-old patient with ovotesticular DSD who underwent PUM, vaginal separation and reconstruction using combination
of UGS flap and FF to reach perineum without tension.
1800
FIG. 9. View of 13-year-old girl with XY-DSD who underwent preoperative and postoperative testing for clitoral sensitivity with warm, cold
and vibratory stimulation. No change in sensation or painful labial engorgement was noted by patient after surgery.
ever, it is gratifying to see that she is satisfied with the

surgical results and does not report postoperative changes in
genital sensation or painful labial engorgement. Despite
good initial results we are aware that late complications
related to patient dissatisfaction can occur but they could be
potentially managed without the need for major interventions. Therefore, we do not see any disadvantages of CSDC
compared with the current techniques in use. Based on that
we believe that it is reasonable to propose this procedure as
an alternative technique to be presented to families when
surgery for CE is contemplated.
3.
4.
5.
6.
7.
8.
CONCLUSIONS
Conservative reconfiguration of the female genitalia is feasible in girls with CE and the anatomical structures of the
clitoris seem to survive after CSDC. The cosmetic appearance of the genitalia is acceptable, at least to the surgeon
and parents, in that the enlarged clitoris is hidden. The
physiological consequences of the current operation and any
surgery in the future to reverse it are unknown. With these
aspects in mind we believe that CSDC should be incorporated into the armamentarium of surgeons involved in the
treatment of CE and presented as an option for feminizing
genitoplasty.
9.
10.
11.
12.
13.
CAH
CE
CSDC
DSD
FF
PUM
UF
UGS
14.
congenital adrenal hyperplasia

clitoral enlargement
corporeal sparing dismembered clitoroplasty
disorders of sexual differentiation
Fortunoff flap
partial urogenital mobilization
urogenital sinus flap
urogenital sinus
15.
REFERENCES
1.
2.
Gross RE, Randolph J and Crigler JF Jr: Clitorectomy for

sexual abnormalities: Indications and technique. Surgery
1966; 59: 300.
Lattimer J: Relocation and recession of the enlarged clitoris
with preservation of the glans: an alternative to amputation. J Urol 1961; 86: 113.
16.
17.
18.
19.
20.
Randolph J and Hung W: Reduction clitoroplasty in females

with hypertrophied clitoris. J Pediatr Surg 1970; 5: 224.
Allen L, Hardy BE and Churchill BM: The surgical management of the enlarged clitoris. J Urol 1982; 128: 351.
Kogan SJ, Smey P and Levitt SB: Subtunical total reduction
clitoroplasty: a safe modification of existing techniques.
J Urol 1983; 130: 746.
Glassberg KI and Laungani G: Reduction clitoroplasty. Urology 1981; 17: 604.
Crouch NS, Minto CL, Liao LM, Woodhouse CRJ and Creighton
SM: Genital sensation after feminizing genitoplasty for congenital adrenal hyperplasia: a pilot study. BJU Int 2004; 93: 135.
Minto CL, Liao LM, Woodhouse CRJ, Ransley PG and Creighton
SM: The effect of clitoral surgery on sexual outcome in individuals who have intersex conditions with ambiguous genitalia: a cross-sectional study. Lancet 2003; 361: 1252.
Newman K, Randolph R and Parson S: Functional results in
young women having clitoral reconstruction as infants.
Fortunoff S, Lattimer JK and Edson M: Vaginoplasty technique for female pseudohermaphrodites. Surg Gynecol Obstet 1964; 118: 545.
Rink RC, Metcalfe P, Cain MP, Meldrum KK, Kaefer MA and
Casale AJ: Use of the mobilized sinus with total urogenital
mobilization. J Urol 2006; 176: 2205.
Pea A: Total urogenital mobilization: an easier way to repair
cloacas. J Pediatr Surg 1997; 32: 263.
Rink RC and Adams MC: Feminizing genitoplasty: state of the
art. World J Urol 1998; 16: 212.
Creighton S and Liao LM: Changing attitudes to sex assignment in intersex. BJU Int 2004; 93: 659.
Maharaj N, Dhai A, Wiersma R and Moodley J: Intersex conditions in children and adolescents: surgical, ethical, and legal
considerations. J Pediatr Adolesc Gynecol 2005; 18: 399.
Hughes IA, Houk C, Ahmed SF and Lee PA: Consensus statement
on management of intersex disorders. J Pediatr Urol 2006; 2:
148.
Warne G, Grover S, Hutson J, Sinclair A, Metcalfe S, Northam
E and Freeman J: A long-term outcome study of intersex
conditions. J Pediatr Endocrinol Metab 2005; 18: 555.
Dessens AB, Slijper F and Drop SLS: Gender dysphoria and
gender change in chromosomal females with congenital
adrenal hyperplasia. Arch Sex Behav 2005; 34: 389.
Kipnis K and Diamond M: Pediatric ethics and the surgical
assignment of sex. J Clin Ethics 1998; 9: 398.
Baskin LS, Erol A, Li YW and Liu WH: Anatomy of the neurovascular bundle: is safe mobilization possible? J Urol
2000; 164: 977.

EDITORIAL COMMENT
For a number of years now it has been clear that the long,
previously held presumption that gender imprinting did not
become fixed until ages 15 to 18 months is incorrect.1 Gender
imprinting begins in utero. The quandary faced by those
who treat DSD is how to measure how much imprinting has
occurred in an infant. That is an unresolved dilemma. From
the consensus statement it is quite clear that we know in a
couple of diagnoses exactly where we should be making the
gender assignment (reference 16 in article). In general females born with CAH will be happy in a female gender role.
Likewise, those with complete androgen insensitivity maybe
satisfactorily raised as female. However, there is no real
consensus on other DSDs.
There is a conflict between the desire not to burn any
bridges and preserve maximally the choices for a child and
the difficulty of parents raising a child whose genitalia look
frankly abnormal at every diaper change. That sets up
stresses for many families that simply are unacceptable.
1801
Accordingly I think that these authors are creative in

arriving at a potential way to perform feminizing genitoplasty that does indeed preserve options. While this is a
limited and preliminary report, I think that such innovative
steps should be encouraged. Further followup of these cases
will provide an answer as to whether this should become a
definitive part of our surgical armamentarium for DSD. It is
important for all of us caring for these children to keep an
open mind about innovations because it is clear that we still
have a long way to go in the management of these difficult
gender assignment and surgery issues.
Howard M. Snyder, III
Division of Urology
1.
Zderic SA, Carr MC, Canning DA and Snyder HM: Pediatric

Gender Assignment: A Critical Reappraisal. New York:
Kluwer-Plenum Press 2002.
Basic Science Research

Nerve Sparing Ventral Clitoroplasty Preserves
Dorsal Nerves in Congenital Adrenal Hyperplasia
Dix P. Poppas, Ariella A. Hochsztein, Rebecca N. Baergen, Emily Loyd, Jie Chen and Diane Felsen
From the Departments of Urology (Institute for Pediatric Urology) and Laboratory Medicine (RNB, EL), New York Presbyterian
Hospital-Weill Cornell Medical College, New York, New York
Purpose: Masculinization of the female genitalia observed in patients with classic congenital adrenal hyperplasia often
results in clitorimegaly. Reduction clitoroplasty is the most widely practiced corrective surgery for clitorimegaly, yet
reservations about surgical intervention exist based on fears of nerve destruction during surgical removal of excess erectile
tissue. In this study, we modified the reduction clitoroplasty and examined excised erectile tissue for the presence of dorsal
nerves.
Materials and Methods: We describe the development of the nerve sparing ventral clitoroplasty. Nerves were examined in
situ using optical coherence technology. In addition, erectile tissue removed from 27 female patients with congenital adrenal
hyperplasia was examined immunohistochemically for the presence of nerves by staining for neurofilament. Nerves outside
of the tunica albuginea were counted and measured. Tissue from 2 adult females was also examined by immunohistochemistry.
Results: Optical coherence technology visualized dorsal nerves in 3 patients with congenital adrenal hyperplasia (size 600
to 800 m). In 4 of 27 patients undergoing nerve sparing ventral clitoroplasty, no dorsal nerve branches were visualized in
excised erectile tissue. In another 18 patients 10 or fewer nerve branches were found. In patients who underwent nerve
sparing ventral clitoroplasty 92% of dorsal nerves detected were 90 m or less. In contrast, 88% of dorsal nerves found in the
2 adult specimens were 120 m or greater. The maximum nerve fiber size observed in patient specimens was significantly
smaller than the maximum nerve fiber size observed in control specimens.
Conclusions: Scarcity of large dorsal nerves in histological specimens excised using nerve sparing ventral clitoroplasty likely
reflects their preservation within the congenital adrenal hyperplasia patients. This preservation is vital to future somatosensory and motor function of the clitoris.
Key Words: adrenal hyperplasia, congenital; virilism; clitoris; immunohistochemistry; abnormalities
asculinized genitalia may develop in female patients with classic CAH in utero upon excess androgen exposure.1,2 This masculinization may be characterized by hyperpigmented or fused labioscrotal tissue,
enlarged labia minora, a urogenital sinus, and clitorimegaly.
Clitorimegaly ranges from an almost normal clitoris to an
almost normal penis, with any degree of intermediate variation. Treatment of clitorimegaly may ensure adequate sexual function, promote consistent gender identity development, and provide a normal clitoral esthetic.
As described by Hendren and Crawford,3 reduction clitoroplasty is the most widely accepted and practiced surgical
management for the treatment of clitorimegaly. It maximizes clitoral esthetic, while preserving potential sexual
function through neurovascular conservation.
To further optimize postoperative somatosensory and
erectile function following reduction clitoroplasty, we have
modified the procedure and developed the NSVC, to optimize
preservation of the dorsal nerves of the clitoris and maintain
erectile function. The dorsal nerves maintain both somatosensory and erectile capacity, which are essential to female
sexual function. The anatomical location of these important
nerves has only recently been described in great detail.4
0022-5347/07/1784-1802/0
These nerves have been documented to course dorsally along

the clitoral erectile bodies, most prominently at the 11 and 1
oclock positions, and send branches around the tunica until
the 5 and 7 oclock positions, between which nerves are
absent.4 In detailed descriptions of clitoral anatomy,
OConnell et al have found the dorsal nerves to be approximately 2 mm in diameter.5,6 They stated furthermore that
the diameter remains unchanged from infancy through
adulthood.
We determined if NSVC could preserve the dorsal nerves
in patients undergoing this procedure. To document this, we
took advantage of new OCT to determine the presence and
size of dorsal nerves, in situ. In addition, erectile tissue
excised from patients undergoing the modified clitoroplasty
was examined immunohistochemically for the presence of
nerves. Specific attention was paid to circumferential nervous tissue, reflecting the course of the dorsal nerve, with
the theory that absence of large nerves from excised tissue
indicated their preservation within the patients, which
would allow for future dorsal nerve function.
METHODS
We retrospectively reviewed the charts of female patients
with CAH, who had undergone a consistently performed
1802
Vol. 178, 1802-1806, October 2007

Printed in U.S.A.
DOI:10.1016/j.juro.2007.03.186
NERVE SPARING VENTRAL CLITOROPLASTY IN CONGENITAL ADRENAL HYPERPLASIA
1803
NSVC by a single surgeon at 1 institution between 1996 and

2004. This procedure was performed in 27 patients, ages 1.5
months to 18 years 11 months. Mean SD patient age was
2 years 8 months 11 months. Median patient age was 11
months. Of the patients 22 were 2 years or younger. Tissue
was also obtained from 2 adult female patients who had
undergone vulvar resection. Three patients underwent intraoperative optical coherence tomography to visualize dorsal nerves in situ.
Surgical Procedure
To begin the NSVC, a 5-zero polypropylene stitch is placed
in the glans clitoris and used as a traction suture. The
preputial adhesions are taken down bluntly and the clitoris
is re-prepared. A subcoronal circumferential incision is then
made 0.5 to 1 cm proximal to the glans clitoris. The incision
should not extend beyond Bucks fascia, to avoid neurovascular damage. The shaft skin is degloved back to the pubis
and sharply bisected in the dorsal midline to create skin
flaps for the development of labia minora. During the degloving, it is necessary to transect the ventral plate. This
significantly lengthens and straightens the erectile bodies.
This erectile body dissection is carried out to the crural
bifurcation between the dartos layer and Bucks fascia. Staying in this plane of dissection is critical to maintaining blood
flow to the bisected skin flaps and preserving the NVBs to
the glans clitoris, which reside between Bucks fascia and
the tunica albuginea (fig. 1).
With tension on the glans stitch to straighten the clitoris,
Bucks fascia is incised using 2 parallel incisions lateral to the
ventral midline in a longitudinal fashion (fig. 1). Using microinstruments and loop magnification; the lateral edges of Bucks
fascia are elevated and sharply dissected off of the tunica
albuginea circumferentially. This is best accomplished by
starting on the right side and working toward the dorsal mid-
FIG. 1. Anatomical diagram of cross-section of clitoral shaft. NVB

(A) with nerves (yellow areas), arteries (red areas) and veins (blue
areas), which are between Bucks fascia (D) and tunica albuginea
(B), corpora cavernosa tissue (C), initial ventral points of incision to
begin elevation of NVB and Bucks fascia off of tunica albuginea (E),
and vascular dartos layer (F).
FIG. 2. Anatomical drawing of clitoroplasty. NVB (A) and Bucks

fascia elevated off of tunica albuginea, tunica albuginea surrounding corpora cavernosa (B), and transection of corpora cavernosa (C)
1.5 to 2 cm distal to bifurcation.
line, and then moving to the left side and working toward the
midline. Once a plane of dissection has been obtained around
the entire corporeal bodies, the neurovascular bundles and
Bucks fascia can be easily mobilized for the entire length of the
corporeal bodies to approximately 1 cm to 2 cm distal to the
crural bifurcation. This allows for maximum preservation of
the neurovascular bundles. During the dissection, 1:100,000
papaverine solution is used to irrigate the neurovascular bundles, facilitating artery identification and preventing vasospasm, thrombosis and ischemia. Care is taken not to grab or
kink the neurovascular bundles. The neurovascular bundles
should be dissected off the corporeal bodies proximal to the
anticipated site of corporeal body transection. Next, the glans
clitoris is sharply dissected off of the distal corporeal bodies,
leaving the glans attached to Bucks fascia and exposing the
entire corporeal bodies (fig. 2).
The corporeal bodies are then transected approximately
1.5 to 2 cm distal to their bifurcation. By leaving some
erectile tissue distal to the bifurcation, clitoral erection during sexual arousal will be maintained, as well as allowing
support and elevation of the clitoris beneath the pubis. The
transected ends of the corporeal bodies are then oversewn
using a running 3-zero polydioxanone suture with a small
tapered needle.
At this point, the glans clitoris is secured to the end of the
corporeal bodies using 2 interrupted 4-zero Biosyn sutures
on a tapered needle. The dorsal aspect of the glans is carefully excluded from these sutures to avoid injuring the
nerves and blood supply as they enter the glans. The position of the NVBs is continuously monitored as the sutures
are tied, to preclude any damage from kinking or entrapment. The glans clitoris is never reduced in size. Once the
glans clitoris is secured, the previously bisected preputial
1804
skin flaps are rotated posteriorly on each side of the clitoris,

trimmed and sutured into place to create a glans hood and
labia minora. The remainder of the genital reconstruction is
completed, including any additional procedures necessary
for that particular patient.
OCT
OCT is a new imaging modality using near infrared light to
obtain high resolution images with 10 times greater resolution
than ultrasound. The image created by backscattered light is
interpreted with an interferometer, which produces the final
image. OCT has been shown to have potential for detecting
bladder cancer.7 We used the Niris Imaging System to evaluate the dorsal nerves in 3 patients undergoing NSVC.
Immunohistochemical Staining
Excised hypertrophic corporeal tissue was embedded in paraffin and routinely sectioned after formalin fixation. Immunoperoxidase staining was performed as routinely carried
out at our laboratory8 with antibody to neurofilament, antihuman neurofilament protein monoclonal mouse antibody
(Dako, Glostrup, Denmark), a structural component of peripheral nerves.9 Heated citrate buffer (pH 6.8) was used for
FIG. 4. a and b, immunohistochemical identification of nerves with

neurofilament immunoperoxidase. Immunoperoxidase stain for
neurofilament successfully stained peripheral nerve fibers on high
power view.
epitope retrieval. Following application of secondary and

tertiary antibodies and chromogen, sections were counterstained with hematoxylin, treated with ammonia solution,
and covered with crystal mount.
Quantifying Nerves
Nerves localized outside of the tunica albuginea representing dorsal nerve branches were counted and measured at 4
power using a micrometer. The frequency of nerve fibers of
varying sizes was calculated on each patient.
RESULTS
The major focus of this report was to determine if the dorsal
nerves are preserved following NSVC. In 3 patients we used
OCT to visualize the dorsal nerves prior to surgery. We
found that in the 3 patients, we could visualize the nerves
and the mean diameter obtained was 735 m (range 600 to
800) (fig. 3). Following surgery, we scanned the excised erectile tissue with OCT. After surgery, no nerves were visualized in the excised tissue.
We also used an immunoperoxidase stain to identify and
localize nerves in the excised tissue. Figure 4 demonstrates the
typical staining, in high magnification, of a dorsal neurovascular complex, indicating the efficacy of this staining procedure.
Figure 5, a shows a trichrome stained slide from a patient, (arrow B), in whom no staining was found outside of
the tunica albuginea (fig. 5, b, arrow B), where dorsal nerves
should be found. In contrast, cavernous nerve fibers were
well visualized in the removed tissue (fig. 5, b, arrow D).
FIG. 3. OCT image. a, typical dorsal nerves observed in situ (black

outline). b, same image without nerves outlined. Arrows indicate
nerve location.
FIG. 5. a and b, cross-sections of same erectile body removed during

clitoroplasty. a, area where NVB and Bucks fascia were located
prior to clitoroplasty. H & E. b, tunica albuginea. Immunohistochemical staining. c, corpora cavernosa tissue. d, positive staining
nerve tissue within corpora cavernosa. Note that no nerve tissue
remained outside of tunica albuginea, indicating that it was not
removed during clitoroplasty.

We examined tissue from 27 patients ages 1.5 months to
18 years 11 months. In the 27 patients, we found a mean of
10.7 2.1 nerves per specimen (range 0 to 47, median 8). In
4 of 27 patients, no nerves were found, and in 18 patients, 10
or fewer nerves were found. In 2 adult patients, we found 8
and 22 nerves per specimen.
The average dorsal nerve size was 59.5 35.7 m in
tissue removed from the 27 patients via NSVC. In the 2
adult patients, the average dorsal nerve size was 245.5
141.4 m. In the NSVC cases 92% of dorsal nerves (265 of
288) detected were 90 m or less. In contrast, 88% of dorsal
nerves found in the 2 adult specimens were 120 m or
greater. The maximum nerve size observed in any NSVC
patient sample was 180 m. In the adult patients, the maximum nerve size found was 630 m.
DISCUSSION
Masculinization of the female genitalia observed in patients
with CAH often results in clitorimegaly. Reduction clitoroplasty is the most widely practiced corrective surgery for
clitorimegaly, yet reservations about surgical intervention
exist, based on fears of nerve destruction during surgical
removal of excess erectile tissue that may result in sexual
dysfunction.
Dorsal nerve preservation is vital to future sexual function. Clitoral somatosensory function is performed predominantly by the dorsal nerve.10 In female animals, pelvic
nerve stimulation (into which the dorsal nerve leads) causes
genital swelling, increased clitoral pressure and engorgement, and increased corporeal diameter and length.11 With
functional and neuroanatomical considerations in mind, a
novel, nerve sparing ventral clitoroplasty was developed and
implemented by a single surgeon using a consistent surgical
protocol in female patients with clitorimegaly. Several
slight, yet deliberate, modifications were made to the surgical technique to maximize postoperative neuronal function.
Incisions within Bucks fascia were made ventrally (compared with a frequently used dorsal approach), just lateral to
the 6 oclock position, where neuronal tissue is sparse. Because nerve density is greatest dorsally and minimal ventrally, the risk of dorsal nerve transection and interruption
with suture was minimized. It should be noted that in a
study of hyenas, a natural animal model of a common urogenital sinus, it was shown that there is a fanning out of the
neurovascular bundle of the female clitoris.12 It might be
expected that in a virilized clitoris, the dorsal nerves might
be found in a more lateral position. Therefore, it is important
to make the incisions close to the 6 oclock position, as
described.
Another modification was leaving a portion of corporeal
tissue distal to the bifurcation within the patient. This allows for optimal esthetic and functional results. The cavernous nerve pierces corporeal tissue at the ventrolateral junction of the crural bodies,13 allowing the corporeal tissue to
raise the glans during intercourse, potentiating stimulation,
arousal, and engorgement. Other techniques often involve
crural body ligation near the pubic bone,14 causing local
nerve damage, whereas neuroanatomical mapping studies
suggest leaving more erectile tissue to avoid this possibility.4
Disruption of the cavernous nerve, as it enters the crural
bifurcation, and of its communicating branches with the
dorsal nerve is avoided with subtotal resection.
1805
Additionally, glans size was not reduced. The dorsal

nerve main trunks enter the glans dorsally,15 branching
extensively both medially and ventrolaterally in the glans
penis16,17 more extensively than in the shaft.18 In the clitoris, their greatest density is dorsally, just deep to the epithelium,5 and least ventrally.4 While neural tissue is sparsest in the glans ventrally and distally, the risk of sensory
loss via ventral resection is unknown.
The risk of sensory loss is minimized in operating at an
early age, as was done for most patients in this study. Age is
overwhelmingly the most important factor in sensory recovery,19 with less difficult recovery in children.20 Furthermore,
dorsal nerve neuroanatomy is comparable in adult and fetal
patients,10 so that surgical technique remains identical at
all ages.
The described modifications were designed to reduce loss
of the nerve tissue. However, objective data on nerve preservation is difficult to obtain, since it cannot be directly
measured or easily observed in vivo. Furthermore, the
smaller size of the clitoris makes it difficult to accurately
measure vasodilatory response to somatosensory stimulation, an objective measure often used for testing penile function. Several studies have focused on questionnaires and
stimulatory testing in patients at age appropriate times, but
results are difficult to quantify.
Therefore, in this study we determined if dorsal nerves
were preserved during surgery. In histological sections, we
visualized nerves with neurofilament. In the 27 patients
examined, 4 specimens showed no nerves, and the majority
showed 10 or less small nerves per specimen. Tissue from
age matched controls is difficult to obtain. However, we used
tissue from 2 adult females who underwent vulvar resection.
In these 2 patients, 8 and 22 nerves were found, respectively.
Baskin et al used the neuronal markers PGP 9.5 and
S-100 in a study of fetal clitoral tissue, from fetuses at 8 to
24 weeks of gestation, and in 2 patients who had undergone
feminizing genitoplasty.2 In their study, they noted that
nerves were noted along the shaft of the clitoris next to the
tunica albuginea in specimens from fetuses of all ages. In the
2 patients who had undergone genitoplasty, they similarly
noted nerves adjacent to the tunica albuginea. It is difficult
to compare their results to our study, since they did not
count nerves or indicate their size. However, the absence of
nerves in 4 patients, and the relatively small number of
small nerves found in this study, suggests that the NSVC
can optimize the preservation of the dorsal nerves.
To confirm these results, we used optical coherence technology to visualize the dorsal nerves. Figure 3 shows that
this technology allowed us to visualize the dorsal nerves in 3
patients. When the same tissue was visualized following
resection, no nerves were found in the excised tissue. We
found that the average nerve size was 735 m. The size of
the nerves found by OCT is similar to the largest nerve size
found in the adult patients (450 and 600 m, respectively).
In contrast, nerves of a much smaller size were found in the
patients, suggesting that these are not dorsal nerves, but
rather branches of the dorsal nerves entering the tunica and
corpora. Although the size of the nerves found even in the
adult samples in this study is smaller than the 2 mm reported by OConnell,5,6 the facts that we were able to visualize nerves before resection and the nerves were not
present in the excised tissue, suggests their preservation.
1806
CONCLUSIONS
15.
Dorsal nerve preservation via NSVC may maximally preserve sexual function. Postoperative sensory and other clinical data in these same patients in a tandem study provide
further evidence that NSVC conserves dorsal nerve fibers
and sexual function within these patients.
16.
17.
18.

CAH
NSVC
NVB
OCT
congenital adrenal hyperplasia

nerve sparing ventral clitoroplasty
neurovascular bundle
optical coherence technology
REFERENCES
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
New MI: An update of congenital adrenal hyperplasia. Ann NY

Acad Sci 2004; 1038: 14.
White PC and Speiser PW: Congenital adrenal hyperplasia due
to 21-hydroxylase deficiency. Endoc Rev 2000; 21: 245.
Hendren WH and Crawford JD: Adrenogenital syndrome: the
anatomy of the anomaly and its repairsome new concepts.
Baskin LS, Erol A, Li YW, Liu WH, Kurzrock E and Cunha GR:
Anatomical studies of the human clitoris. J Urol 1999; 162:
1015.
OConnell HE, Hutson JM, Anderson CR and Plenter RP:
Anatomical relationship between urethra and clitoris.
J Urol 1998; 159: 1892.
OConnell HE, Sanjeevan KV and Hutson JM: Anatomy of the
clitoris. J Urol 2005; 174: 1189.
Manyak MJ, Gladkova ND, Makari JH, Schwartz AM,
Zagaynova EV, Zolfaghari L et al: Evaluation of superficial bladder transitional-cell carcinoma by optical coherence tomography. J Endourol 2005; 19: 570.
Ito K, Chen J, El Chaar M, Stern JM, Seshan SV, Khodadadian
JJ et al: Renal damage progresses despite improvement of
renal function after relief of unilateral ureteral obstruction
in adult rats. Am J Physiol Renal 2004; 287: F1283.
Cai Z, Manavis J, Cash K, Thompson PD and Blumbergs PC:
Immunohistochemical staining of epoxy resin sections of
peripheral nerve. Appl Immunohistochem Molec Morphol
2005; 13: 292.
Yucel S and Baskin LS: Identification of communicating
branches among the dorsal, perineal and cavernous nerves
of the penis. J Urol 2003; 170: 153.
Giraldi A, Marson L, Nappi R, Pfaus J, Traish AL, Abdulmaged M et al: Physiology of female sexual function: animal
models J Sex Med 2004; 1: 237.
Baskin LS, Yucel S, Cunha GR, Glickman SE and Place NJ: A
neuroanatomical comparison of humans and spotted hyena,
a natural animal model for common urogenital sinus: clinical reflections on feminizing genitoplasty. J Urol 2006;
175: 276.
Baskin LS, Erol A, Li YW and Liu WH: Anatomy of the neurovascular bundle: is safe mobilization possible? J Urol
2000; 164: 977.
Snyder HM, Retik AB, Bauer SB and Colodny AH: Feminizing
genitoplasty: a synthesis. J Urol 1983; 129: 1024.
19.
20.
Baskin LS, Lee YT and Cunha GR: Neuroanatomical ontogeny

of the human fetal penis. BJU 1997; 79: 628.
Breza J, Aboseif SR, Orvis BR, Lue TF and Tanagho EA:
Detailed anatomy of penile neurovascular structures: surgical significance. J Urol 1989; 141: 437.
Yang CC and Bradley WE: Innervation of the human glans
penis. J Urol 1999; 161: 97.
Baskin LS, Erol A, Li YW and Cunha GR: Anatomical studies
of hypospadias. J Urol 1998; 160: 1108.
Rosen B, Lundborg G, Dahlin LB, Holmberg J and Karlson B:
Nerve repair: correlation of restitution of functional sensibility with specific cognitive capacities. J Hand Surg 1994;
19B: 452.
Lundborg G: A 25-year perspective of peripheral nerve surgery: evolving neuroscientific concepts and clinical significance. J Hand Surg 2000; 25A: 391.
EDITORIAL COMMENT
This study shows the surgical efficacy of the nerve preservation technique of modified clitoroplasty. Clitoral neuroanatomy has been studied in detail and important technical
modifications have been suggested to preserve the sensory
and motor roles of the clitoral dorsal nerves for optimal
sexual function and gratification (reference 4 in article). It is
a challenging task to find and study a virilized human clitoral sample, such as a complete pathological specimen. This
disadvantage is partly compensated for by the observation of
the naturally virilized phallus in female hyenas, Crocuta
crocuta. This hyena study showed that more lateral fanning
out of the dorsal nerves could be the case in the virilized
clitoris and a medial first incision on the ventral clitoris
seems wise for the best nerve preservation (reference 12 in
article).
As suggested in the current article, OCT is also a compelling strategy for nerve identification in phallic surgery as
long as we know where to look on the clitoris. However, we
are still unsure about the sensory and motor outcomes after
any type of clitoroplasty. We know that there is a lot of
redundancy among nerves in highly sensitized organs like
the genitalia (reference 10 in article).1,2 Is it really important to preserve every nerve in the genital area? Would it
make surgery more successful if we preserved the larger
nerves? Would it be wiser to try our best to protect landmarks known to be rich in nerve redundancy? I believe that
now the time has come to verify the functional outcomes of
nerve preserving surgical techniques for the genitalia.
Selcuk Yucel
Akdeniz University School of Medicine
Antalya, Turkey
1.
2.
Yucel S, De Souza A Jr and Baskin LS.: Neuroanatomy of the

human female lower urogenital tract. J Urol 2004; 172: 191.
Yucel S and Baskin LS.: Neuroanatomy of the male urethra and
perineum. BJU Int 2003; 92: 624.
The Effect of Oxybutynin on Structural

Changes of the Obstructed Guinea Pig Bladder
Jeroen R. Scheepe,* Bas W. D. de Jong, Katja P. Wolffenbuttel, Marlous E. Arentshorst,
Petra Lodder and Dirk J. Kok
From the Department of Pediatric Urology, Erasmus Medical Center, Sophia Childrens Hospital, Rotterdam, The Netherlands
Purpose: Oxybutynin is used clinically to lower intravesical pressure and detrusor overactivity. In vitro it inhibits stretch
induced bladder smooth muscle cell proliferation. We tested whether oxybutynin also prevents hypertrophic bladder changes
in vivo in a model of partial bladder obstruction.
Materials and Methods: Subvesical obstruction was induced in immature guinea pigs by a silver ring around the urethra.
Eight animals received 0.4 mg oxybutynin per kg body weight per day in 2 doses. Control groups were obstructed without
oxybutynin treatment or sham operated. Urodynamic pressure flow studies were performed at 1-week intervals for 10 weeks
in all animals under anesthesia with ketamine/xylazine. After 10 weeks the animals were sacrificed and the bladder was
removed for structural analysis with periodic acid-Schiff stain, in which the number of glycogen granules was also scored as
a measure of previous ischemia.
Results: Compared to the sham treated group obstructed animals had significantly higher intravesical pressure and detrusor
overactivity, lower compliance and increased contractility. Obstructed animals that received oxybutynin retained normal
intravesical pressure, detrusor overactivity and compliance. Their bladder contractility increased as in obstructed animals.
The oxybutynin group showed less collagen infiltration in the detrusor and fewer glycogen granules compared to those in
obstructed animals.
Conclusions: Our results demonstrate that oxybutynin has a protective effect on bladder function and structure. Prevention
of hypertrophic and ischemic bladder changes is an argument for an early start of oxybutynin treatment in children with
inborn neurogenic bladder dysfunction, such as spina bifida, or in patients with urethral valves.
Key Words: bladder; cholinergic antagonists; guinea pigs; muscle, smooth; hypertrophy
nticholinergics in combination with clean intermittent catheterization are the mainstays of current
treatment for neurogenic bladder dysfunction,1,2 a
disease often characterized by DO and functional bladder
outlet obstruction (detrusor-sphincter dyssynergia). Antimuscarinic drugs such as tolterodine or oxybutynin suppress DO and partly lower infravesical pressure. This in
combination with clean intermittent catheterization usually
results in a low pressure system, preserving renal function.
Also, in nonneurogenic bladder outlet obstruction, eg posterior urethral valves, a combination of detrusor hypertrophy,
DO, a low compliance bladder and high intravesical pressure
is observed. With current treatment progression of smooth
muscle cell hypertrophy and bladder wall fibrosis is still
common. To prevent the last treatment option, invasive
surgery, including bladder augmentation or any form of
urinary diversion, and its associated complications a better
understanding of the road to the end stage bladder and of
the way that compounds such as oxybutynin interfere is
needed.
Study received approval from the Erasmus University institutional animal care and use committee.
Supported by the Dutch Kidney Foundation.
* Correspondence: Erasmus Medical Center, Sophia Childrens
Hospital, Department of Urology and Pediatric Urology, P. O. Box
2060, 3000 CB Rotterdam, The Netherlands (telephone: 31-10-463
65 59; FAX: 31-10-463 68 02; e-mail: j.scheepe@erasmusmc.nl).
0022-5347/07/1784-1807/0
From the literature certain facts are known.3 6 An obstructed bladder increases in muscle volume to produce the
higher pressure needed to overcome obstruction. This muscle hypertrophy is accompanied by collagen deposition and it
may end in a thick walled, high pressure bladder. After this
stage the bladder may reverse in strength and mass. At
some point intrinsic bladder wall processes start to drive
hypertrophy. The start may be a decrease in bladder blood
flow due to high bladder pressure, inducing local hypoxia7,8
and activating anaerobic detrusor function.9,10 Chronic hypoxia induces hyperexcitation of the afferent nerves that are
involved in DO,11 fibrosis, collagen deposition and glycogen
storage.1116
In vitro relaxation of bladder smooth muscle by the anticholinergic agent oxybutynin15 or the smooth muscle cell
relaxant nitric oxide inhibits stretch induced bladder
smooth muscle cell proliferation.15,17 To our knowledge this
effect of oxybutynin has not yet been studied in vivo. We
evaluated the effect of oxybutynin on bladder structure with
emphasis on signs of hypoxia in a guinea pig model of partial
bladder obstruction and we related this effect to longitudinally followed bladder function.

A previously described guinea pig model for gradual urethral obstruction6,7 was used. We used immature male al-
1807
Vol. 178, 1807-1812, October 2007

Printed in U.S.A.
DOI:10.1016/j.juro.2007.03.187
1808
OXYBUTYNIN AND OBSTRUCTED GUINEA PIG BLADDER STRUCTURAL CHANGES
bino guinea pigs of the Hartley strain weighing approximately 250 gm.
Study Design
Data were obtained from each animal every week. Analysis
of the response to obstruction and treatment was based on
individual longitudinal data. In the reference groups (sham
operated and obstructed) this availability of longitudinal
data in individual animals allowed us to sacrifice animals
and obtain tissue at variable end points. Consequently the
number of animals in these groups decreased during the
experiment. The size of the oxybutynin group did not allow
this division during time end points. Thus, these animals
were followed for 10 weeks and sacrificed at the same end
point.
The sham treated group consisted of 9 animals that were
sham operated and 2 that were unoperated. Seven animals
underwent weekly urodynamic studies for up to 9 weeks.
Four animals were measured at a lower frequency. The
animals were sacrificed at different time points. From weeks
0 (before obstruction) to 9 the number of animals included in
analysis was 11, 10, 10, 8, 3, 3, 6, 7, 4 and 2, respectively. As
an individual reference each obstructed/oxybutynin treated
animal also underwent initial measurement before obstruction.
Obstructed Group
A total of 93 animals underwent obstruction after the first
urodynamic evaluation at day 0. They were measured
weekly or biweekly thereafter. From weeks 0 to 10 the
number of analyzed, obstructed animals was 93, 66, 76, 61,
55, 47, 46, 26, 14, 14 and 11, respectively. In the staining
procedures tissue from the group of 11 animals that completed 10 weeks of obstruction was compared to tissue from
sham operated and oxybutynin treated animals, which also
had 10 weeks of obstruction.
Operative Procedure
With the animal under ketamine/xylazine anesthesia the
peritoneal cavity was accessed via a lower vertical midline
abdominal incision. A silver jeweler jump ring with an internal diameter of 2.2 mm was placed around the bladder
neck above the prostate.
End Point Procedure
After the final urodynamic study the animal was sacrificed
and the bladder was removed immediately. Part was snapfrozen and stored at 80C and part was fixated in 10%
buffered formalin (pH 7.4) and embedded in paraffin.
Calculations
DO was assessed visually as the number of detrusor contractions of 10 cm H2O (a third of mean voiding pressure at
week 0) without urine flow during the filling phase. Maximum flow rate, associated pdetQmax, bladder compliance in
ml/cm H2O and bladder contractility were calculated from
the pressure flow plot, as described previously (fig. 1).5,6
PAS Stain and PAS Score
Cross-sections (4 m) covering all bladder wall layers were
stained with PAS. This procedure stains sugar moieties to
visualize the general structure (fig. 2). At what we defined as
score 0 linear structures such as cell membranes were visible. At higher scores granular staining appeared inside the
muscle cells. We counted the number of these granules and
the increasing number was defined as a score of 1 to 3. We
observed that concomitant with the increase in the number
of granules their location changed. At score 1 the granules
were located near the adventitium, at score 2 they were also
found in the middle of the muscle layer and at score 3 they
were noted throughout the muscle up to the urothelium. On
biochemical analysis a high count of PAS stained granules in
tissue correlated with a high glycogen content.16 Using
Oxybutynin Group
Eight animals were obstructed and received oxybutynin
from that time on for 10 weeks. Two animals were withdrawn from the experiment at week 1 due to the discomfort
of the operative wound and at week 5 due to failure to thrive,
respectively. Only the remaining 6 animals were used for
analysis. Up to the point that the 2 animals were removed
from the study their urodynamic parameters did not differ
from those of the rest of the group but all of their data were
excluded. The applied dose of oxybutynin was 0.4 mg/kg per
day divided over 2 doses.
Urodynamic Procedure
The animals were anesthetized using ketamine (43 mg/kg
intramuscularly) and xylazine (0.9 mg/kg intramuscularly).
Through a 24 gauge suprapubic catheter bladder pressure
was measured and the bladder was filled continuously with
sterile saline at a rate of 0.23 ml per minute. The flow rate
was measured with a T106 small animal flow meter ultrasound transducer (Transonic Systems, Ithaca, New York)
around the penis. The volume voided was collected in tubes
and used to calibrate the flow rate signal of each voiding.
After the last voiding residual fluid was measured.
FIG. 1. Pressure flow plot shows relation between bladder pressure

(pdet) and flow rate. Each curve represents 1 void. After 10 weeks of
obstruction curve of oxybutynin treated animal was still close to
curve of untreated animal after 1 week of obstruction. ml/s, ml per
second.
1809
of animals in the comparison changed from week to week, as

explained. The protocol was approved by the institutional
animal care and use committee of Erasmus University and it
was in line with European Union guidelines.
RESULTS
Pressure Flow Plots
Figure 1 shows pressure flow plots for an obstructed animal
and an obstructed/oxybutynin treated animal. In the obstructed animal bladder pressure increased with the duration of obstruction and the urine flow rate decreased. In the
oxybutynin treated obstructed animal pressure was only
slightly increased after 10 weeks and the urine flow rate was
increased greatly.
FIG. 2. Examples of lowest and highest PAS scores of 0 (left) and 3
(right), respectively. At PAS score 0 distinct line pattern of membranous staining, among others, was visible. At higher scores additional staining occurred as granules (right). At higher magnification
(bottom) these granules appeared to be inside muscle cells.
Raman spectroscopy we reported that these granules contain high amounts of glycogen.12
EVG
In the EVG procedure collagen stains red, elastin stains
black and the rest of the tissue stains yellow. Three investigators visually assessed the percent of red in the muscle
layer and scored this as 1 to 3.
Statistics
All values are expressed as the mean SEM. Statistical
significance was determined by the 2-sided unpaired Student t test with p 0.05 considered statistically significant.
Note that because of our specific study protocol, the number
Bladder Pressure
Compared to the sham treated group pdetQmax increased
significantly in the obstructed group (p 0.05, fig. 3). In
obstructed animals that received oxybutynin pdetQmax did
not differ from that in the sham treated group and it was
significantly lower than in the obstruction group (p 0.05).
In a previous publication we divided the group of obstructed animals according to the PAS score (0 to 3) of the
bladder tissue at the end point.14 Figure 3 shows the average end point bladder pressure for these PAS score groups.
For the group with a PAS score of 1 the average end point
bladder pressure was 47 cm H2O. Average bladder pressure
in the oxybutynin group remained below this level for all 10
weeks.
Figure 4 shows individual longitudinal data. In 4 of 6
oxybutynin treated animals pdetQmax remained close to normal for the whole period. Their bladder tissue had a PAS
score of 1. In 1 animal there was a transient increase in
pdetQmax at week 8 up to the level of the obstructed group. In
this animal the PAS score was 2. In 1 animal pdetQmax was
FIG. 3. Effect of oxybutynin on pdetQmax during 10 weeks of obstruction. Solid lines represent average pdetQmax for animals with PAS score
of 0 to 3, respectively. Values on which these lines are based were taken from de Jong et al.14 Pas, PAS. Asterisk indicates significant
difference vs obstruction (p 0.05).
1810
FIG. 4. Values of pdetQmax in individual oxybutynin treated animals. Each symbol represents individual animal. Solid lines represent average
pdetQmax for groups with PAS (pas) score of 1 to 3, respectively.14
continuously increased during the last 3 weeks. In this animal the PAS score was 3.
Urine Flow Rate
The urine flow rate was higher in the oxybutynin group than
in the obstructed and sham operated animals throughout
the test period (fig. 5). Compared to the sham treated group
the difference attained significance from week 1 and thereafter. Compared to the obstructed group the difference was
significant at weeks 2, 3 and 5 to 7.
Contractility
Compared to the sham treated group bladder contractility
increased significantly in obstructed animals and animals
that received oxybutynin (p 0.05, fig. 6). The latter 2
groups did not differ from each other.
pared to that in the sham group (p 0.05). In oxybutynin

treated animals compliance remained normal and it
was significantly higher than in the obstruction group
(p 0.05).
DO
In sham operated animals DO was rare (fig. 8). Obstruction induces DO. This starts the first week after obstruction and increases in severity the following weeks. In
obstructed animals receiving oxybutynin some DO occurred but significantly less than in the obstructed group.
During the final weeks DO increased in the oxybutynin
group. This increase was caused entirely by the 2 animals
in which suppression of high bladder pressure failed during those weeks.
Compliance
Bladder compliance in sham operated animals fluctuated
around 1 during the whole study period (fig. 7). In obstructed animals compliance decreased significantly com-
EVG
The mean EVG score of 1.7 0.2 for oxybutynin treated
animals was significantly lower than the score of 2.4 0.2
for 10-week obstructed animals and significantly higher
than the score of 1.1 0.1 for sham operated animals
FIG. 5. Effect of oxybutynin on flow rate during 10 weeks of obstruction.
FIG. 6. Effect of oxybutynin on contractility during 10 weeks of

obstruction. Wmax, bladder contractility. Asterisk indicates significant difference vs sham treatment (p 0.05).
1811
(each p 0.05). Thus, oxybutynin treated animals showed

less collagen infiltration into the bladder muscle.
DISCUSSION
Oxybutynin is a drug with nonspecific antimuscarinic properties used in the clinic to suppress bladder overactivity in patients with obstructive bladder disease. In our model oxybutynin prevented obstruction induced increases in bladder
pressure and instability, and preserved bladder compliance,
although it did not prevent the increase in bladder contractility. Unexpectedly the flow rate was increased by oxybutynin.
The latter may point to some unknown effect on sphincter or
urethral relaxation and it warrants further investigation.
It was not known what effect long-term oxybutynin treatment in vivo has on bladder structure. Bladder smooth muscle cells respond to mechanical stretch by changes in cell
proliferation and protein synthesis.18,19 It is well documented that several smooth muscle relaxants, including
oxybutynin, can suppress detrusor muscle cell proliferation
in vitro.15 We observed that oxybutynin also decreases collagen infiltration into the bladder muscle in vivo. Less hypertrophy and less matrix deposition may explain part of the
maintenance of good compliance under oxybutynin. Improved muscle relaxation during the filling phase could be a
second mechanism. We also found that under oxybutynin
treatment less glycogen accumulation occurs in the muscle
cells. This may be a sign that oxybutynin treatment has
decreased hypoxia in the bladder wall and it may be related
to the DO reduction.
Several investigators have suggested that decreased
bladder wall blood flow leading to hypoxia links high intravesical pressure to DO.11,16,20 The afferent nerves in the
bladder wall that regulate DO may be damaged by oxidative
stress during reperfusion.11 Muscle cells may reduce oxidative stress by turning to the anaerobic metabolism of glycogen. Bladder muscle cells that more often experience high
bladder pressure for prolonged times have more glycogen
stores.16 In vitro strips of muscle cells that originally do not
express overactivity become overactive when they are first
depleted of glycogen.
It was previously reported that the granules stained by a
PAS procedure are intracellular glycogen stores.12,16 We
report that during obstruction guinea pig detrusor cells
build up these PAS stained glycogen granules. Obstructed
animals that maintained low bladder pressure, good compliance and no DO during oxybutynin treatment did not show
this buildup. In 2 animals in which the suppression of high
bladder pressure and overactivity transiently fails glycogen
FIG. 8. Effect of oxybutynin on DO during 10 weeks of obstruction
stores the buildup occurred. In the review of Brading et al it

was mentioned that bladder blood flow completely ceases
when bladder pressure exceeds 30 mm Hg (41 cm H2O).11
The 4 animals in our study with no glycogen buildup remained below this 41 cm H2O bladder pressure during the
whole period. The 2 animals that had glycogen buildup
surpassed this value. Why the suppression by oxybutynin
failed in these animals is unclear. It may have been due to
inadequate drug intake for several days or to some unknown
oxybutynin resistance. Since the effect was transient in 1
animal, we favor the first explanation. However, this highlights the fact that, although the bladder in these animals
may have seemed normal at the surface with low pressure,
no overactivity and good compliance it could revert to a
situation in which it becomes damaged rapidly, probably
because of increased contractility.
During obstruction bladder contractility increases because more effort is needed to empty the bladder against
obstruction. Obstructed animals treated with oxybutynin
showed the same increase in contractility and had to
overcome the same obstruction. Increased flow in the oxybutynin group suggests that there was some over compensation.
Such an effect could underlie the experience in patients
receiving long-term oxybutynin treatment, in whom unexplained and sudden deterioration of bladder function
can occur. Possibly in these patients oxybutynin treatment for some reason temporarily fails and the increased
contractility of that bladder allows it to produce high
pressures that turns on the cycle of decreased blood flow,
hypoxia, nerve damage and overactivity. A question for
future study is whether treatment with compounds that
aim to improve blood microcirculation as a single compound or in combination with oxybutynin would keep the
bladder wall in better shape and further decrease the
chance of ischemia.
CONCLUSIONS
FIG. 7. Effect of oxybutynin on compliance during 10 weeks of

obstruction. Asterisk indicates significant difference vs sham treatment (p 0.05).
Our results demonstrate that oxybutynin might have a protective effect on bladder function. Preventing hypertrophic
bladder changes related to ischemia is an argument for an
early start of oxybutynin treatment in children with inborn
neurogenic bladder dysfunction, such as spina bifida, or in
patients with urethral valves.
1812

9.

BOO
CIC
DO
EVG
PAS
pdetQmax
bladder outlet obstruction

detrusor overactivity
elastic van Gieson stain
periodic acid-Schiff
bladder pressure during maximal flow
10.
11.
12.
REFERENCES
1.
2.
3.
4.
5.
6.
7.
8.
Kasabian NG, Bauer SB, Dyro FM, Colodny AH, Mandell J and
Retik AB: The prophylactic value of clean intermittent
catheterization and anticholinergic medication in newborns
and infants with myelodysplasia at risk of developing urinary tract deterioration. Am J Dis Child 1992; 146: 840.
Dik P, Klijn AJ, van Gool JD, de Jong-de Vos van Steenwijk CC
and de Jong TP: Early start to therapy preserves kidney
function in spina bifida patients. Eur Urol 2006; 49: 908.
Steers WD and De Groat WC: Effect of bladder outlet obstruction on micturition reflex pathways in the rat. J Urol 1988;
140: 864.
Levin RM, Haugaard N, OConnor L, Buttyan R, Das A, Dixon
JS et al: Obstructive response of human bladder to BPH vs.
rabbit bladder response to partial outlet obstruction: a direct comparison. Neurourol Urodyn 2000; 19: 609.
Wolffenbuttel KP, Kok DJ, Minekus JP, van Koeveringe GA,
van Mastrigt R and Nijman JM: Longitudinal urodynamic
follow-up of experimental urethral obstruction in individual
guinea pigs. Neurourol Urodyn 2001; 20: 699.
Kok DJ, Wolffenbuttel KP, Minekus JP, van Mastrigt R and
Nijman JM: Changes in bladder contractility and compliance due to urethral obstruction: a longitudinal followup of
guinea pigs. J Urol 2000; 164: 1021.
Ghafar MA, Shabsigh A, Chichester P, Anastasiadis AG,
Borow A, Levin RM et al: Effects of chronic partial outlet
obstruction on blood flow and oxygenation of the rat bladder. J Urol 2002; 167: 1508.
Levin RM, OConnor LJ, Leggett RE, Whitbeck C and
Chichester P: Focal hypoxia of the obstructed rabbit bladder wall correlates with intermediate decompensation.
Neurourol Urodyn 2003; 22: 156.
13.
14.
15.
16.
17.
18.
19.
20.
Kato K, Lin AT, Haugaard N, Longhurst PA, Wein AJ and

Levin RM: Effects of outlet obstruction on glucose metabolism of the rabbit urinary bladder. J Urol 1990; 143: 844.
Lin AT, Monson FC, Kato K, Haugaard N, Wein AJ and Levin
RM: Effect of chronic ischemia on glucose metabolism of
rabbit urinary bladder. J Urol 1989; 142: 1127.
Brading A, Pessina F, Esposito L and Symes S: Effects of
metabolic stress and ischaemia on the bladder, and the
relationship with bladder overactivity. Scand J Urol
Nephrol, suppl., 2004; 215: 84.
de Jong BW, Bakker-Schut TC, Coppens J, Wolffenbuttel KP,
Kok DJ and Puppels GJ: Characterization of collagen infiltration by Raman spectroscopy. Vibrational Spectrosc 2003;
32: 57.
Migally N, Tucker A and Zambernard J: Glycogen localization
in pulmonary vascular smooth muscle of chronically hypoxic rats. J Submicrosc Cytol 1982; 14: 247.
de Jong BWD, Wolffenbuttel KP, Arentshorst ME, Lodder P
and Kok DJ: Detrusor glycogen reflects the functional history of bladders with partial outlet obstruction. Unpublished data.
Park JM, Bauer SB, Freeman MR and Peters CA: Oxybutynin
chloride inhibits proliferation and suppresses gene expression in bladder smooth muscle cells. J Urol 1999; 162: 1110.
Pessina F, Solito R, Maestrini D, Gerli R and Sgaragli G: Effect
of anoxia-glucopenia and re-superfusion on intrinsic nerves
of mammalian detrusor smooth muscle: importance of glucose metabolism. Neurourol Urodyn 2005; 24: 389.
Johansson R, Pandita RK, Poljakovic M, Garcia-Pascual A, De
Vente J and Persson K: Activity and expression of nitric
oxide synthase in the hypertrophied rat bladder and the
effect of nitric oxide on bladder smooth muscle growth.
J Urol 2002; 168: 2689.
Adam RM, Eaton SH, Estrada C, Nimgaonkar A, Shih SC,
Smith LE et al: Mechanical stretch is a highly selective
regulator of gene expression in human bladder smooth
muscle cells. Physiol Genomics 2004; 20: 36.
Park JM, Borer JG, Freeman MR and Peters CA: Stretch
activates heparin-binding EGF-like growth factor expression in bladder smooth muscle cells. Am J Physiol 1998;
275: C1247.
Mitterberger M, Pallwein L, Gradl J, Frauscher F, Neuwirt H
et al: Persistent detrusor overactivity after transurethral
resection of the prostate is associated with reduced perfusion of the urinary bladder. BJU Int 2007; 99: 831.
Clinical Research
Artificial Urinary Sphincters Placed After Posterior
Urethral Distraction Injuries in Children are at Risk for Erosion
Richard A. Ashley and Douglas A. Husmann*,
From the Department of Urology, Mayo Clinic, Rochester, Minnesota
Purpose: Management for posterior urethral disruption and concurrent bladder neck incompetence is controversial. Some
groups recommend treatment with a Mitrofanoff catheterizable stoma, while others advocate urethral reconstruction with
delayed placement of an artificial urinary sphincter. We report our experience with the latter strategy.
Materials and Methods: We reviewed the records of all patients with the above injury who were treated with end-to-end
urethroplasty followed by delayed bladder neck artificial urinary sphincter placement from 1986 to 2006.
Results: Five patients had videourodynamic evidence of bladder neck incompetence coexisting with traumatic posterior
urethral disruption. The etiology of bladder neck incompetence in all 5 patients was a known longitudinal tear through the
bladder neck that occurred at the time of trauma. Each patient underwent end-to-end urethroplasty. Six to 12 months later
the patients had persistent incontinence. Bladder function and urethral patency were documented by urodynamic, radiographic and endoscopic studies. A bladder neck artificial urinary sphincter was subsequently placed. Each operation was
technically demanding due to fibrosis in the pelvis and around the bladder neck. All patients were initially continent but
erosion of the artificial urinary sphincter into the bladder neck in 4, and the bladder neck and rectum in 1 occurred at a mean
of 3 years (range 6 months to 8 years).
Conclusions: Placement of a bladder neck artificial urinary sphincter for managing urinary incontinence due to concurrent
posterior urethral disruption and bladder neck incompetence is difficult and it risks delayed erosion. In this patient
population we would strongly consider urinary diversion with a Mitrofanoff catheterizable stoma.
Key Words: sphincter, artificial urinary; urethra; urinary incontinence; wounds and injuries
rethral injury occurs in approximately 10% of pelvic

trauma cases and on rare occasions it may be associated with longitudinal bladder neck laceration.1
The combination of these injuries is recognized as a harbinger for future difficulties with urinary incontinence.2 In patients with recognized bladder neck lacerations and concurrent posterior urethral disruptions we have primarily
repaired the bladder neck, placed a suprapubic tube and
subsequently performed delayed urethral reconstruction or
alternatively we have repaired the 2 injuries at the time of
insult. When followup after successful urethral reconstruction reveals persistent bladder neck incompetence, options
for management include the creation of continent urinary
diversion by Mitrofanoff formation with bladder neck closure or the reestablishment of urethral continuity with delayed AUS placement. Use of the AUS has not been extensively evaluated in the posttraumatic setting. We reviewed
AUS outcomes in children after pelvic trauma associated
with posterior urethral disruption and simultaneous bladder neck injury.

* Correspondence: Department of Urology, Mayo Clinic, 200 First
St. Southwest, Rochester, Minnesota 55905 (e-mail: husmann.
douglas@mayo.edu).
0022-5347/07/1784-1813/0

After receiving approval from the institutional review board
all patients who sustained traumatic pelvic fractures compounded by 1) posterior urethral distraction injury and 2) a
longitudinal tear through the bladder neck were identified
from 1986 to 2006. No other injuries occurred to other internal organs. All patients included in this study underwent
primary bladder neck repair with placement of a suprapubic
tube and delayed urethral reconstruction or immediate repair of the 2 injuries. Six to 12 months after urethral repair
videourodynamics and cystoscopy were performed. Patients
with persistent urinary incontinence subsequently underwent placement of a bladder neck AUS. The AUS was placed
according to previously described, standard protocols.3 The
outcomes in these patients were reviewed.
RESULTS
Patient Cohort
From 1986 to 2006 we identified 5 male patients with a
median age of 11 years (range 8 to 16) who experienced
pelvic trauma. All patients sustained a pelvic fracture with
traumatic distraction of the pubic rami, resulting in a posterior urethral disruption injury and a longitudinal laceration through the bladder neck. Three patients underwent
end-to-end urethroplasty and bladder neck repair the night
of the injury. Two other patients underwent immediate bladder neck repair with delayed correction of the urethral disruption. All urethral reconstructions were successful, as doc-
1813
Vol. 178, 1813-1815, October 2007

Printed in U.S.A.
DOI:10.1016/j.juro.2007.05.095
1814
ARTIFICIAL URINARY SPHINCTERS AND EROSION RISK
umented by cystoscopy and retrograde urethrography.

However, all 5 patients experienced posttraumatic urinary
incontinence. Videourodynamics confirmed bladder neck incompetence and complete urinary incontinence.
AUS Placement and Surgical Approach
All 5 patients underwent AUS placement 6 to 12 months
after the initial surgery. Each patient received an
AMS800 model placed at the bladder neck with a narrow
back 4.5 cm cuff and a 61 to 70 cm H2O balloon reservoir.
The abdominal approach was used in each procedure. Each
surgical case was hampered by significant posttraumatic
fibrosis in the pelvis and around the bladder neck. No urethral, bladder or rectal injuries occurred during dissection.
Each sphincter was left in the deactivated state for 6 weeks.
No immediate postoperative complications occurred.
AUS Outcomes
In the beginning all patients were completely continent
without any need for pads. With time unfortunately all AUS
devices eroded. Four patients experienced erosion of the
device into the bladder neck, while 1 had erosion of the AUS
into the bladder neck and rectum. Mean time to erosion was
3 years (range 6 months to 8 years) in all patients.
When this occurred the AUS was urgently explanted. The
1 child with coexisting bladder and rectal erosion required
AUS explantation in conjunction with diverting colostomy.
Six to 12 months after AUS explantation patients required
bladder neck closure along with formation of a continent
urinary stoma. Median followup in these patients after bladder neck closure and continent diversion was 5 years (range
2 to 13). No long-term sequelae were found in these patients
at the time of this report but our prior studies indicate
that erectile dysfunction will likely be a major concern in
the future.4
DISCUSSION
Management for posttraumatic urinary incontinence is a
difficult task and controversial subject in adult and pediatric
urology. This urological ailment is often a consequence of
longitudinal lacerations through the bladder neck and severe injury to the sacral plexus/pudendal nerve complex.2,4
No consensus guidelines have been established and management is at surgeon discretion. The options that we chose to
pursue were to perform primary or delayed repair of the
urethral and bladder neck injuries, followed by delayed insertion of an AUS at the bladder neck. We chose this location
since our patients were children and adolescents with little
perineal tissue for AUS protection, and because of concern
for a tenuous blood supply to the newly reconstructed urethral margin. We did not want to damage the repaired
urethra or risk AUS erosion by placing the device at the
bulbar region.
Unfortunately our data demonstrate that delayed placement of a bladder neck AUS was technically difficult and
carried an overwhelming risk of delayed erosion. Indeed, the
end results were quite devastating with 5 children undergoing difficult AUS explants and 1 requiring simultaneous
colostomy. All patients eventually underwent bladder neck
closure with formation of a continent stoma.
Our finding of 100% erosion (5 of 5 patients) is in contrast

to the erosion and infection rates quoted at between 2% and
20%, depending on patient comorbidities, in other series.59
However, our findings are similar to the higher erosion rates
found in the exstrophy-epispadias population with prior
bladder neck reconstruction.7,8,10,11 These patients, similar
to ours, have an aberrant pelvic vascular supply due to
congenital, traumatic and iatrogenic causes. AUS complications in children with bladder exstrophy following pelvic
surgery occur so frequently that some groups may consider
prior bladder neck repair to be a contraindication to placement of the device at this position. Our data support the
concept that children with prior bladder neck reconstruction
should not receive a bladder neck AUS.
The main limitation of this study is the small patient
cohort. However, to our knowledge this report represents the
only published series in which AUS placement has been
studied in the posttraumatic setting in children. The poor
long-term results that we report weigh heavily on our minds
and for that reason we could not recommend AUS placement
in a posttraumatic pediatric pelvis. Instead, we would support consideration for continent catheterizable urinary diversion in children with complete urinary incontinence due
to traumatic posterior urethral disruption and concurrent
bladder neck incompetence.
CONCLUSIONS
Pediatric pelvic trauma associated with posterior urethral
distraction defects and longitudinal bladder neck laceration
is a controversial topic. The tissues require extensive repair
and even after arduous labor incontinence may persist. To
resolve this one may consider placement of a bladder neck
AUS. However, we present 5 cases in which extremely difficult surgical dissection was done to place the device. We
report that the erosion rate was high and erosion caused
significant morbidity. Rather than place an AUS in a traumatized pelvis and risk further tissue damage, we recommend continent urinary diversion. Although this may preclude voiding via the urethra, it would prevent future
operations that may be even more devastating to self-esteem
and overall quality of life.

AUS artificial urinary sphincter
REFERENCES
1.
Koraitim MM: Pelvic fracture urethral injuries: the unresolved

controversy. J Urol 1999; 161: 1433.
2. Iselin CE and Webster GD: The significance of the open bladder neck associated with pelvic fracture urethral distraction defects. J Urol 1999; 162: 347.
3. Elliott DS and Barrett DM: Mayo Clinic long-term analysis of
the functional durability of the AMS 800 artificial urinary
sphincter: a review of 323 cases. J Urol 1998; 159: 1206.
4. Husmann DA, Wilson WT, Boone TB and Allen TD: Prostatomembranous urethral disruptions: management by suprapubic cystostomy and delayed urethroplasty. J Urol 1990;
144: 76.
5. Herndon CD, Rink RC, Shaw MB, Simmons GR, Cain MP, Kaefer
M et al: The Indiana experience with artificial urinary sphincters in children and young adults. J Urol 2003; 169: 650.
ARTIFICIAL URINARY SPHINCTERS AND EROSION RISK

6.
7.
8.
Barrett DM, Parulkar BG and Kramer SA: Experience with AS

800 artificial sphincter in pediatric and young adult patients. Urology 1993; 42: 431.
Castera R, Podesta ML, Ruarte A, Herera M and Medel R:
10-Year experience with artificial urinary sphincter in children and adolescents. J Urol 2001; 165: 2373.
Hafez AT, McLorie G, Bagli D and Khoury A: A single-centre
long-term outcome analysis of artificial urinary sphincter
placement in children. BJU Int 2002; 89: 82.
9.
1815
Raj GV, Peterson AC and Webster GD: Outcomes following

erosions of the artificial urinary sphincter. J Urol 2006;
175: 2186.
10. Decter RM, Roth DR, Fishman IJ, Shabsigh R, Scott FB and
Gonzales ET: Use of the AS800 device in exstrophy and
epispadias. J Urol 1988; 140: 1202.
11. Fishman IJ, Shabsigh R and Scott FB: Experience with the
artificial urinary sphincter model AS800 in 148 patients.
J Urol 1989; 141: 307.
Long-Term Continence Outcomes After Immediate Repair

of Pediatric Bladder Neck Lacerations Extending Into the Urethra
Jonathan C. Routh and Douglas A. Husmann*,
From the Department of Urology, Mayo Clinic, Rochester, Minnesota
Purpose: Traumatic bladder neck lacerations extending into the urethra are devastating injuries that occur more commonly
in children than in adults. It is unclear whether immediate repair of these injuries decreases long-term complications, such
as urinary incontinence. We report our long-term results with immediate operative repair of these injuries.
Materials and Methods: Since 1986, we have primarily repaired all individuals sustaining traumatic longitudinal bladder
neck lacerations extending into the urethra within 24 hours of injury. All patients were followed a minimum of 2 years.
Results: A total of 12 patients 2 to 16 years old sustained longitudinal bladder neck lacerations extending into the proximal
urethra. Median followup was 7 years (range 2 to 13). Postoperatively none of the 12 patients recovered complete urinary
continence. Periurethral bulking agents were attempted in all 12 patients with no long-term improvement. Eight patients
(75%) pursued additional surgery. Three boys underwent artificial urinary sphincter placement, of which all subsequently
eroded. Three girls underwent bladder neck reconstruction with fascial sling procedures, of whom 2 became continent but
experienced urinary retention, while 1 became partially continent. Five patients, including the 3 boys with artificial urinary
sphincter erosion, ultimately underwent bladder neck closure and continent diversion.
Conclusions: Traumatic longitudinal bladder neck and proximal urethral lacerations are devastating injuries fraught with
long-term complications and the need for additional surgery despite immediate surgical repair. Bladder neck closure and
continent diversion should be considered in girls with substantial traumatic urethral loss and in boys with persistent urinary
incontinence following primary repair.
Key Words: wounds and injuries; urethra; bladder; urinary sphincter, artificial
raumatic bladder neck lacerations are an uncommon

but serious injury that may occur following pelvic
fracture with longitudinal extension of the laceration
into the proximal urethra seen in up to half of patients.1,2
Previous reports demonstrated that bladder neck lacerations with urethral extension occur more commonly in children than in adults17 and they are significantly associated
with urinary incontinence and sexual dysfunction.13
Repair of combined longitudinal bladder neck and urethral injuries may be performed immediately or in delayed
fashion. Delayed repair is typically performed following temporary urinary diversion by suprapubic tube placement or
urethral catheterization. In our experience temporary diversion without acute reapproximation of the bladder neck and
urethral injury is associated with urinary extravasation
leading to pelvic urinoma, abscess, osteomyelitis or necrotizing fasciitis and urinary incontinence.1,3 In contrast to
our experience with delayed repair, immediate repair is
reported to result in the return of urinary continence.2 Since
1986, we have treated all patients with pelvic fracture and
associated bladder neck injury extending into the urethra
with immediate primary repair. We report our long-term
results with immediate operative repair of these injuries in
children.

* Correspondence: Department of Urology, Mayo Clinic, 200 1st
St. Southwest, Rochester, Minnesota 55905 (telephone: 507-2660191; FAX: 507-284-5591; e-mail: husmann.douglas@mayo.edu).
0022-5347/07/1784-1816/0

After receiving approval by the institutional review board
we identified and retrospectively reviewed the records of all
children sustaining traumatic pelvic fractures with associated longitudinal bladder neck lacerations extending into
the distal urethra between 1986 and 2006. During this era it
was our standard practice to primarily repair all traumatic
urethral injuries within 24 hours. All children were treated
by a single surgeon (DAH). Patients older than 18 years and
children with concurrent urethral distraction injuries were
excluded from study.
Urinary continence was assessed by pad count at the time
of physician interview 3 months following surgery. Continence was defined as the total lack of urinary leakage,
partial continence was defined as urethral leakage requiring
3 or fewer pads per day and diaper dependent incontinence
was defined as urethral leakage requiring greater than 3
pads per day.
RESULTS
Eight girls and 4 boys sustained longitudinal bladder neck
lacerations extending into the proximal urethra during the
study period. All patients had pelvic fractures with associated pubic diastasis. All patients underwent primary repair
within 24 hours of injury. Median patient age was 8 years
(range 2 to 16) and median post-trauma followup was 7
years (range 2 to 13). In all 4 boys the longitudinal bladder
neck laceration extended through the prostatic urethra and
1816
Vol. 178, 1816-1818, October 2007

Printed in U.S.A.
DOI:10.1016/j.juro.2007.05.094
CONTINENCE OUTCOMES AFTER PEDIATRIC BLADDER NECK LACERATIONS

into the bulbar urethra. Coexisting vaginal lacerations that
were contiguous with the vaginal injury were present in 4 of
the 8 girls (50%). One of these young girls also sustained a
separate distal urethral laceration, which was simultaneously repaired. Concurrent rectal injury requiring fecal
diversion occurred in 4 of the 12 children (33%). A pelvic
abscess developed postoperatively in 2 of these 4 patients, of
whom 1 also had necrotizing fasciitis. No patient had a
urethral stricture or urethrovaginal fistula during followup.
It should be noted that the 1 girl with the coexisting distal
urethral injury lost the distal 2 cm of urethra, resulting in a
hypospadiac female urethra with the meatus located approximately 2 cm from the vaginal introitus.
No patient showed complete urinary continence following
primary repair. Partial continence developed in 5 of the 12
children (42%). Less than 1 pad per day was required in 1
patient, while another 4 required 2 to 3 perineal pads per
day. Only 1 patient with partial continence was male. Diaper dependent incontinence was present in 7 of the 12 patients (58%), of whom 3 were male and 4 were female.
A single periurethral collagen injection was performed in
all 12 patients with no improvement in continence noted
after 3 months of followup. Therefore, additional surgery
was attempted in 8 of the 12 patients (75%). Four of the 5
partially continent patients chose not to pursue further surgical repair.
AUS placement was performed in 3 boys with diaper
dependant incontinence using an AMS800 device. The
AUS was placed around the bulbar urethra in all 3 patients
and underwent delayed activation, as previously described.8
All patients were initially dry and required no perineal pads
through the immediate postoperative period. Unfortunately
all of these patients were found to have AUS erosion through
the bulbar urethra between 3 and 8 years after placement.
These patients subsequently underwent AUS explantation,
secondary bladder neck closure and continent urinary diversion using a Mitrofanoff channel.
Three of the 8 female patients (38%) underwent modified
bladder neck reconstruction using a fascial sling. Two girls
were diaper dependent and 1 had partial continence with 3
pads per day. This surgery was performed with the specific
goal of attaining continence even at the expense of urinary
retention. In 2 of 3 patients (67%) this goal was attained.
These patients became continent postoperatively but required intermittent urethral catheterization. Partial continence developed in the remaining female patient, who required 1 perineal pad per day postoperatively. However, this
patient was able to void to completion. No further operative
measures were required in this group.
Primary bladder neck closure and continent urinary diversion was performed in 1 boy and 1 girl. In the female
patient this option was chosen due to the development of a
hypospadiac female urethral meatus following the traumatic injury. In this patient the urethral meatus was located
approximately 2 cm within the vaginal ostium, making visualization of the urethral meatus and intermittent catheterization via the urethra virtually impossible. In the male
patient the parents elected primary diversion after discussion of the potential risks of erosion and further surgery
associated with AUS placement. These patients have done
well from a continence standpoint and no further interventions have been required.
1817
DISCUSSION
Relatively few groups have explored the long-term ramifications of bladder neck lacerations with proximal urethral
extension.1 In the few available reports long-term complications are significant. From the current literature it is unclear whether these complications are due to the traumatic
injury or to the delayed manner in which the injury was
repaired.13 We describe our results with the practice of
immediate repair of all bladder neck lacerations with urethral extension. Immediate repair of the bladder neck injury
seemed to decrease the incidence of acute complications that
we had previously seen with delayed repair.1,3 Specifically
no patient had urinary extravasation leading to pelvic urinoma or osteomyelitis. However, we noted that patients with
concurrent rectal injuries may still have a pelvic abscess
and/or necrotizing fasciitis. Although we had hoped that
immediate repair would result in increased urinary continence compared to delayed repair, this was not our experience since all of our patients had urinary incontinence following initial repair. Indeed, 8 of the 12 children (75%)
required further operative intervention to achieve urinary
continence.
In this series we present 12 children followed up to 13
years after injury. As noted, all patients underwent initial
bladder neck and urethral repair, which was performed by a
single surgeon within 24 hours of injury. Despite immediate
repair no patients achieved complete continence. An attempt
to improve continence by injection of a urethral bulking
agent was performed in all patients, although unfortunately
no patient improved after a single injection. No patient
elected repeat injection. Eight of the 12 patients (75%) subsequently chose further surgery to achieve continence. In
the 3 boys in whom AUS placement was attempted the AUS
eroded in 100% after extended followup.
Our results with AUS placement merit further discussion. Previous groups have reported excellent results with
the AUS device in children.9 11 However, high erosion rates
have been clearly noted in patients who underwent previous
bladder neck and urethral reconstruction, presumably due
to aberrant or inadequate urethral vasculature following the
surgical procedure.9,12 We suspect that the delayed AUS
erosion that we noted were due to this reason. Specifically
we hypothesize that the scarring from the traumatic injury
and its surgical repair combined with posttraumatic alteration of the pelvic/urethral blood supply predisposed these
patients to delayed AUS erosion.
Three girls underwent a fascial bladder neck sling procedure, which was performed to improve continence. Two girls
became continent but urinary retention developed, while the
third achieved only partial continence. These procedures
were difficult to perform due to extensive scarring in the
bladder neck region. Despite the traumatic injury and difficult dissection no erosions occurred in this patient population during long-term followup. Ultimately 5 of our 12 patients (40%) required bladder neck closure and continent
diversion, including primary closure in 2 and secondary
closure in 3 after urethral AUS erosion. All patients are
continent and without sequelae. Because of this result, we
would recommend that all boys in whom primary repair fails
and any girls who have substantial urethral loss should be
considered for primary bladder neck closure and a continent
abdominal stoma as a primary treatment modality.
1818
CONTINENCE OUTCOMES AFTER PEDIATRIC BLADDER NECK LACERATIONS
To our knowledge this is the first study to describe the

long-term complications and management of pediatric longitudinal bladder neck lacerations extending into the urethra via immediate primary repair within 24 hours of injury.
However, our findings have limitations that bear mentioning. This is a single surgeon series of a small number of
patients and, thus, it is possible that other individual surgeons could achieve a better result. Notably our results are
in contrast to the recent report by Black et al, who noted that
1 of 2 children with concurrent bladder neck and urethral
injuries achieved continence following primary repair.2 The
reasons for this discrepancy are unclear, although the relatively limited sample size of the 2 studies certainly may have
a role. Beyond a mere difference in study power our results
may have been due to the severity of the bladder neck injury,
which is clearly difficult to quantify and compare between
the series, or they may have been due to the severity of
sphincteric denervation (pelvic nerve injury), which was not
assessed in either study. Finally, it should be noted that this
study was not designed to compare or contrast the efficacy of
primary vs delayed bladder neck and urethral repair. Although we believe in the usefulness of primary repair, in
part to decrease acute complications such as pelvic urinoma
or abscess, our data cannot be used to substantiate this
practice.
CONCLUSIONS
Traumatic longitudinal bladder neck lacerations extending
into the proximal urethra are devastating injuries. Even
with immediate primary repair 7 of our 12 patients (58%)
had diaper dependent incontinence and required further
operative intervention. In our experience a single injection
of urethral bulking agents did not improve the continence of
any patient. In girls fascial bladder neck slings, although
technically difficult, were effective for attaining continence
at the cost of postoperative urinary retention. In boys AUS
placement was universally associated with long-term erosion years after the initial injury. Overall 5 of the 12 patients
(more than 40%) with this severe and debilitating injury
ultimately required bladder neck closure and continent urinary diversion. Due to our experience we recommend primary bladder neck closure and the formation of a continent
abdominal stoma in girls who sustain substantial traumatic
urethral loss and in boys with persistent urinary incontinence following primary repair.

AUS artificial genitourinary sphincter
REFERENCES
1.
Husmann DA, Wilson WT, Boone TB and Allen TD: Prostatomembranous urethral disruptions: management by suprapubic cystostomy and delayed urethroplasty. J Urol 1990;
144: 76.
2.
Black PC, Miller EA, Porter JR and Wessells H: Urethral and

bladder neck injury associated with pelvic fracture in 25
female patients. J Urol 2006; 175: 2140.
3. Perry MO and Husmann DA: Urethral injuries in female subjects following pelvic fractures. J Urol 1992; 147: 139.
4. Bredael JJ, Kramer SA, Cleeve LK and Webster GD: Traumatic rupture of the female urethra. J Urol 1979; 122: 560.
5. Merchant WC 3rd, Gibbons MD and Gonzales ET Jr: Trauma
to the bladder neck, trigone and vagina in children. J Urol
1984; 131: 747.
6. Okur H, Kucikaydin M, Kazez A, Turan C and Bozkurt A:
Genitourinary tract injuries in girls. Br J Urol 1996; 78:
446.
7. Patil U, Nesbitt R and Meyer R: Genitourinary tract injuries
due to fracture of the pelvis in females: sequelae and their
management. Br J Urol 1982; 54: 32.
8. Elliott DS and Barrett DM: Mayo Clinic long-term analysis of
the functional durability of the AMS 800 artificial urinary
sphincter: a review of 323 cases. J Urol 1998; 159: 1206.
9. Castera R, Podesta ML, Ruarte A, Herrera M and Medel R:
10-Year experience with artificial urinary sphincter in children and adolescents. J Urol 2001; 165: 2373.
10. Hafez AT, McLorie G, Bagli D and Khoury A: A single-centre
long-term outcome analysis of artificial urinary sphincter
placement in children. BJU Int 2002; 89: 82.
11. Herndon CD, Rink RC, Shaw MB, Cain MP and Casale AJ:
Experience with non-cycled artificial urinary sphincters.
BJU Int 2004; 93: 1049.
12. Kryger JV, Spencer Barthold J, Fleming P and Gonzalez R:
The outcome of artificial urinary sphincter placement after
a mean 15-year follow-up in a paediatric population. BJU
Int 1999; 83: 1026.
EDITORIAL COMMENT
This is another good study from a leading group in the field
of trauma urology. The take-home lessons seem to be
1) repair bladder neck lacerations acutely in children because you will avoid complications (persistent urinoma, abscess and necrotizing fasciitis, etc) and 2) expect these patients with bladder neck lacerations to be ultimately
incontinent (100% of 12). This is welcome illumination of
how to treat a difficult problem even if it is disappointing to
hear that bladder neck injuries in children so often result in
incontinence.
Surgery to address the resulting incontinence is equally
problematic. Not surprisingly collagen injection into the
scarred and injured area failed. Three of 3 patients with an
AUS ultimately had erosion. The answer seems to be bladder neck closure and suprapubic diversion using a continent
catheterizable stoma. Incontinent females may be cured by
pubovaginal sling placement. From experience I can suggest
that these surgeries are often difficult in this previously
traumatized operative field. Referral to a center where there
is experience with such surgeries is probably prudent.
Richard A. Santucci
Detroit Receiving Hospital
Wayne State University School of Medicine
Detroit, Michigan
The Common Ileal Ureter: A New

Technique for Compliant Ureterocystoplasty
D. Frimberger, J. Klein and B. P. Kropp
From the Department of Urology, Childrens Hospital of Oklahoma, Oklahoma Health Science Center, Oklahoma City, Oklahoma
Purpose: Ureterocystoplasty is an effective technique for bladder reconstruction in patients with megaureter. Initial reports
were encouraging but later repeat augmentation with bowel was necessary in many patients. We evaluated whether repeat
augmentation after ureterocystoplasty could be avoided using two-thirds of each megaureter.
Materials and Methods: Ureterocystoplasty was performed in 6 patients using the distal two-thirds of the 2 ureters.
Continuity was reestablished by anastomosis of the remaining proximal ureters to a tubularized and tapered piece of ileum,
which was reimplanted in an antireflux manner into the reconstructed bladder. All patients underwent preoperative and
postoperative evaluation with ultrasound, creatinine, voiding cystourethrogram, nuclear renal scan and videourodynamic
testing.
Results: Patient age at ureterocystoplasty was between 7 and 15 years with a median followup of 45.3 months. Preoperative
videourodynamics demonstrated low capacity bladders with grade 5 vesicoureteral reflux and a poor mean bladder compliance of 7.4 ml/cm H2O. Bladder capacity increased up to 12-fold postoperatively with a mean compliance rate of 58 ml/cm H2O
and vesicoureteral reflux resolved in all patients. One patient required endoscopic incision of the reimplanted common ileal
ureter but no other complications occurred.
Conclusions: The common ileal ureter provided a long-term compliant reservoir without the need for future repeat
augmentation in all patients. Using standard urological techniques the complication rates remained low and recovery time
was similar to that of standard ureterocystoplasty.
Key Words: bladder; ureter; anastomosis, surgical; ileum
espite advances in the preoperative and postnatal

diagnosis and management of neurogenic bladder
disease, bladder augmentation remains a necessity
for many affected patients to protect the upper tracts. In the
last decades several groups have tried to replace mucous
producing segments of the gastrointestinal tract with suitable tissue to decrease side effects and complications.1 Ureterocystoplasty has been shown to be an effective technique
for bladder reconstruction in a select group of patients with
megaureters. Initial reports demonstrated a significant increase in bladder capacity and compliance with stable renal
function and hydronephrosis.2 Landau et al compared 8
patients undergoing ureterocystoplasty with a matched
group undergoing ileocystoplasty and reported almost identical clinical and urodynamic outcomes.3 Initially the ureter
was harvested from poorly functioning kidneys at nephrectomy. Other groups used the dilated distal ureters of the 2
functioning kidneys and combined the ureters via transureteroureterostomy for single ureteral reimplantation.
These initial positive results were challenged by Husmann et al in a multi-institutional study after several
groups found that patients frequently required repeat augmentation after ureterocystoplasty.4 Husmann et al reviewed preoperative ultrasound, voiding cystourethrogram,
and preoperative and postoperative urodynamic studies in
64 patients to evaluate who would benefit most from the
Study received approval from the institutional review board committee at University of Oklahoma.
0022-5347/07/1784-1819/0
procedure. They found that the used ureteral tissue was

often not sufficient to provide the necessary bladder capacity
and compliance needed to protect the upper tracts.
We evaluated whether repeat augmentation after ureterocystoplasty could be avoided by using two-thirds of each
megaureter. The remaining proximal ureters were anastomosed to a tubularized piece of ileum, which was reimplanted into the bladder in an antireflux manner. The clinical outcome of this common ileal ureter was evaluated using
the same strict criteria established by Hussmann et al.4

We performed a retrospective chart review of 6 patients
undergoing ureterocystoplasty from 2000 to 2001 at our
institution. The study protocol was approved by the institutional review board committee of the University of Oklahoma. Informed consent was obtained before surgery was
performed.
All patients underwent preoperative and postoperative
evaluation with US, creatinine, voiding cystourethrogram,
nuclear renal scan and VUDS. Preoperatively all patients
had some degree of spontaneous voiding and they additionally performed CIC via the urethra.
All patients underwent ureterocystoplasty using the distal two-thirds of the 2 ureters. Urinary system continuity
was reestablished by ureteral anastomosis of the remaining
proximal ureters to a tubularized and tapered piece of ileum.
This common ileal ureter was subsequently reimplanted in
1819
Vol. 178, 1819-1823, October 2007

Printed in U.S.A.
DOI:10.1016/j.juro.2007.03.158
1820
COMMON ILEAL URETER
an antireflux manner into the reconstructed bladder. Four of

the 6 patients additionally underwent creation of a catheterizable appendicovesicostomy as described by Mitrofanoff to
circumvent urethral catheterization.
US was graded following Society for Fetal Urology guidelines for hydronephrosis. The amount of ureteral dilatation
was measured in cm using longitudinal ultrasound at the
bladder level. Urodynamic studies were performed as described by Hussmann et al.4 VUDS was added to visualize
the onset of VUR and judge the voiding phase. Bladder
compliance was defined as the change in bladder volume per
unit change in bladder pressure in ml/cm H2O, as measured
at end fill bladder capacity. Normal compliance was defined
as greater than 30 ml/H2O. Mild, moderate and severe noncompliance was defined as 29 to 20, 19 to 10 and less than 10
ml/cm H2O, respectively.5 Preoperative bladder capacity
was defined as measured capacity at VUR onset. Nuclear
renal scans were done to evaluate drainage, function and
renal scar formation using diethylenetriamine pentaacetic
acid with Lasix administration at 20 minutes or mercaptoacetyltriglycine effective renal plasma flow in ml/min.
Surgical Technique
Standard access was achieved via a midline incision. The
ureters were mobilized and divided 5 to 7 cm distal to the
ureteropelvic junction. For the common ileal ureter a 7 to 15
cm segment of ileum was harvested in isoperistaltic orientation. The proximal end was anastomosed to the proximal
ureters using the Bricker6 or Wallace7 technique. The distal
5 to 7 cm of the ileal segment was tapered over a 10Fr
catheter following the standard excisional tapering technique described for megaureters and reimplanted antireflux
into the native bladder.8 The remaining 15 to 20 cm distal
segments of the 2 ureters were detubularized, reconfigured
to form a patch and augmented onto the opened bladder, as
described by Wolf et al.9
Suprapubic and ureteral catheters were left indwelling
for 2 and 3 weeks, respectively. A closed suction drain was
left for 1 week. Postoperative care was done following the
principles of routine enterocystoplasty (fig. 1).
RESULTS
The etiology for noncompliant bladder with associated
megaureters was PUV in 4 cases, anterior urethral valves in
1 and nonneurogenic neurogenic bladder in 1. All patients
with PUV underwent vesicostomy in the infant period, followed by vesicostomy closure and valve resection in the first
2 years of life. All patients were treated preoperatively with
spontaneous voiding, CIC, and anticholinergic and prophylactic antibiotic treatment. Patient age at ureterocystoplasty
was between 7 and 15 years (median 10.6). Followup was 27
to 52 months (median 45.3).
All patients performed regular CIC with spontaneous
voiding and were on nighttime catheter drainage. All patients are currently continent via the urethra and/or urinary
stoma, defined as a dry interval of more than 3 hours. We
continue to regularly follow 5 of the 6 patients at our institution. One patient was lost to followup due to incarceration.
All surgeries were performed by a single surgeon (BPK) and
no intraoperative complications occurred.
All patients presented preoperatively with grade 4 dilatation of the renal collecting system on US. Postoperatively
FIG. 1. Common ileal ureter with tapered ileal segment reimplanted

into bladder and ureteral anastomosis using Wallace technique.
the kidneys demonstrated decompression of the collecting

system with improvement in pelvic and caliceal dilatation.
However, due to parenchymal thinning renal dilatation still
had to be considered grade 4. All patients presented preoperatively with bilateral hydroureters. Preoperative diameter
of the distal ureters measured behind the bladder was between 0.9 and 1.8 cm (mean 1.4). In 2 patients the proximal
ureters were also visualized preoperatively on US. Postoperatively the ureters completely decompressed and were no
longer visualized on US. Preoperative VUDS demonstrated
low capacity bladders with grade 5 VUR at low volumes and
poor bladder compliance between 1 and 15 ml/cm H2O (mean
7.4). Postoperatively bladder capacity increased up to 12fold with good compliance rates between 20 and 116.7 ml/cm
H2O (mean 58). VUR resolved in all patients (fig 2).
Preoperative nuclear renal scans and creatinine values
confirmed chronic renal insufficiency in 4 patients (see table). Three of those patients had worsening renal function
during followup, while 1 showed improvement. Renal function stabilized postoperatively but 2 patients ultimately required renal transplantation. Transplant function remained
stable throughout followup. In patient 3 renal function stabilized postoperatively for 12 months but he continued to be
noncompliant with the CIC regimen. Renal failure developed and the patient was ultimately lost to followup 31
months postoperatively. The table lists individual findings
COMMON ILEAL URETER
1821
FIG. 2. Voiding cystourethrogram. A and B, preoperative view shows immediate grade 5 VUR at low bladder capacity of 45 ml. C,
postoperative view reveals resolved VUR and large bladder capacity of 700 ml.
on preoperative and postoperative evaluations for each patient.

Patient 2 had pseudo-obstruction at the entrance of the
ileal ureter into the bladder, which was caused by an over
distended bladder due to CIC noncompliance postoperatively (see table). The ileal ureter was successfully incised
endoscopically without provoking reflux and drained without further problems under a strict CIC regimen. No other
complications were observed throughout followup.
DISCUSSION
Bladder augmentation with ureteral tissue from massively
dilated ureters is a compelling concept since it promises to
avoid the complications seen with the use of bowel segments.
Previous studies were favorable but they lacked important
information concerning preoperative and postoperative
bladder function, and often had short followup. The initial
excitement of the procedure soon decreased since many patients ultimately required repeat augmentation with bowel
during followup. Before we offered the current technique of
the common ileal ureter we also performed conventional
ureterocystoplasty using a single ureter or the distal 2 ureters depending on the preoperative situation of the patient.
Four of 5 patients who underwent the conventional procedure ultimately required re-augmentation with ileum during followup (data not shown). The ureteral patch used was
simply not large enough to sufficiently increase bladder capacity and compliance for continuous protection of the upper
tracts.
Other pediatric urology groups noted similar problems in
their patients. This resulted in a retrospective multicenter
study to determine the parameters that would predict longterm success without the need for repeat augmentation in
patients who underwent conventional ureterocystoplasty.10
This study at 5 large teaching institutions included 64 eligible patients operated on between 1987 and 2002. Three
key assessments were set to determine if a patient was an
Ml/Min Renal Scan

(% rt/lt)
ideal candidate for ureterocystoplasty based on the preoperative presence or absence of VUR, the diameter of the nonrefluxing ureter and bladder compliance. Obtained data suggested that augmenting a noncompliant bladder with a
refluxing megaureter would be less successful than using
obstructed megaureters. Overall repeat augmentation was
performed and pending in 29 and 18 of the 64 patients,
respectively. The investigators theorized that the observed
difference was due to the higher content of less distensible
collagen III in refluxing ureters compared with obstructed
megaureters. Ultimately 2 patient groups were defined with
the highest chances of long-term success for ureterocystoplasty without the need for later repeat augmentation, including group 1no VUR and a ureteral width of 1.5 cm,
and group 2VUR and a compliant or mildly noncompliant
bladder with greater than 20 ml/cm H2O on VUDS. While
these strict criteria provided the best long-term outcome, it
also excluded a large population of patients from ureterocystoplasty.
Still, the idea of augmenting the bladder with urothelial
lined tissue to avoid the complications of gastrointestinal
segments remains attractive. The current proposed technique was designed to make ureterocystoplasty possible for
all patients with megaureters who would be otherwise excluded by the Husmann criteria.4 We theorized that repeat
augmentations with intestine are mostly necessary because
the amount of used ureteral tissue is simply not enough.
However, using two-thirds of the 2 distal ureters the patch
would be sufficient to create a compliant reservoir even in
the unfavorable groups with high collagen III content. Urine
flow was then reconstituted by connecting the remaining
ureteral ends to a segment of tapered ileum reimplanted
into the reconstructed bladder. Using the standard Bricker6
or Wallace7 technique of ileoureteral anastomosis the 2
proximal ureteral segments were anastomosed to the proximal part of the ileal segment. Tapering the distal part of the
common ileal ureter allowed antireflux reimplantation into
Bladder Compliance
(ml/cm H2O)
Bladder Capacity
(ml)
Creatinine
(mg/dl)
Pt No.
Diagnosis
Preop
Postop
Preop
Postop
Preop
Postop
Preop
Postop
Renal
Transplantation
at Followup
1
2
3
4
5
PUV
PUV
PUV
PUV
Anterior urethral
valves
Nonneurogenic
bladder
178 (50/47)
86 (50/50)
101 (6/94)
111 (50/50)*
43 (38/62)
89.5 (53/47)
98 (68/32)
Not available
115 (36/64)*
40 (37/63)
15
10
6.25
4.7
1
29.4
33.3
Not available
90.2
20
150
100
300
150
40
500
600
Not available
450
400
1.2
3.5
3.5
0.9
1.0
1.6
1.6
6.0
0.9
1.5
Yes
Yes
No
No
No
271 (41/59)
104
116.7
45
700
2.2
1.1
No
(39/61)
* Mercaptoacetyltriglycine/diethylenetriamine pentaacetic acid scan.
1822
COMMON ILEAL URETER
the reconstructed bladder in the standard fashion of appendicovesicostomy (fig. 1). Therefore, the short ileal segment
does not contribute to increase the bladder capacity but it
merely serves as a conduit between kidneys and bladder.
Mucous production and urine-bowel contact is minimal,
avoiding the well documented complications seen with bowel
augmentations. Concerning the development of possible
later malignancy, the situation is comparable to replacement of a diseased ureter with ileum. Routine followup in
the future would include urine tests for hematuria and imaging will be performed if necessary. However, the creation
of the common ileal ureter certainly increases the extent of
bladder reconstruction and raises questions of its justification.
Based on the strict criteria of the multicenter study10 all
conventional ureterocystoplasties would have failed in the
current patients. All patients had preoperative grade 5
VUR, small bladder capacity and a poorly compliant bladder, making them candidates for probable repeat augmentation with bowel in the future. Using the common ileal
ureter technique the ureteral patch became large enough to
create a long-term compliant reservoir. The mean preoperative poor compliance rates of 7.4 ml/cm H2O (range 1 to 15)
increased to a mean compliance of 58 ml/cm H2O postoperatively (see table). Additionally, all patients drastically enlarged the true bladder capacity up to 12-fold compared to
preoperative values. Preoperative voiding studies demonstrated that only a small amount of urine remained in the
bladder with most trapped in the upper tracts (fig. 2). We
determined preoperative bladder capacity by the amount of
bladder filling before the demonstration of reflux into the
upper tracts. VUDS confirmed that the bladders no longer
increased in volume as soon as reflux occurred (fig. 2). Postoperatively reflux resolved in all patients, while the bladders
now stored urine at safe pressures at 400 to 700 ml (mean
530.2). Most patients underwent yearly VUDS, which confirmed stable bladder compliance and capacity rates during
several years. Only 1 patient did not undergo postoperative
VUDS due to incarceration but that patient is reportedly
stable from a bladder point of view. Postoperatively US
revealed continuous grade 4 hydronephrosis due to chronic
pelvicaliceal dilatation. However, the collecting system appeared decompressed in all patients. Additionally, the preoperatively massive dilated ureters decompressed completely in all patients and were no longer visualized
postoperatively. The procedure was well tolerated with a
mean hospital stay of 3 to 5 days and a recovery time similar
to that of standard enterocystoplasty. No intraoperative complications occurred, and all patients and families were
highly satisfied with the procedure.
One patient required transurethral incision of the reimplanted ileal ureter 1 year postoperatively. He was initially
not compliant with the CIC regimen, causing high post-void
residual urine volume with subsequent elongation of the
conduit, which led to channel kinking and obstruction. After
the incision the ileal ureter drained sufficiently without
causing new onset VUR. He subsequently stayed on a strict
CIC regimen and was stable 6 years postoperatively.
Patient age at operation was between 10 and 15 years.
Underlying obstruction in patients with PUV was treated
during infancy. However, due to long-term disease 3 patients had already experienced chronic renal failure, as confirmed by preoperative nuclear renal scans and creatinine
values (see table). Ultimately 2 of the 3 patients required

renal transplantation during followup due to the initial renal damage. Also, the common ileal ureter procedure provided them with a safe bladder environment, although it
could not reverse the already occurred upper tract damage.
However, the now compliant bladders allowed all affected
patients to obtain a renal transplant without compromising
transplant function. All patients perform regular CIC and
are on continuous nighttime bladder drainage. Some also
have spontaneous voids. With this regimen all patients are
currently dry for more than 3 hours. None required repeat
augmentation.
CONCLUSIONS
The common ileal ureter technique provides successful ureterocystoplasty in patients excluded from standard ureterocystoplasty by the Husmann criteria.4 The bladders remained compliant with good capacity, providing upper tract
protection without the need for future repeat augmentation.
Using standard urological techniques of ureter-bowel anastomosis and conduit reimplantation the complication rates
remain low and recovery time is similar to that of standard
enterocystoplasty. As with all augmentation techniques, patients must be compliant with the CIC regimen.

CIC
PUV
US
VUDS
VUR

posterior urethral valves
ultrasound
videourodynamic testing
REFERENCES
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
Gerharz EW, Turner WH, Kaelble T and Woodhouse CRJ:

Metabolic and functional consequences of urinary reconstruction with bowel. BJU Int 2003; 91: 143.
Churchill BM, Aliabadi H, Landau EH, McLorie GA, Steckler
RE, McKenna PH et al: Ureteral bladder augmentation.
J Urol 1993; 150: 716.
Landau EH, Jayanthi VK, Khoury AE, Churchill BM, Gilmour
RF, Steckler RE et al: Bladder augmentation: ureterocystoplasty versus ileocystoplasty. J Urol 1994; 152: 716.
Husmann DA, Snodgrass WT, Koyle MA, Furness PD 3rd,
Kropp BP, Cheng EY et al: Ureterocystoplasty: indications
for a successful augmentation. J Urol 2004; 171: 376.
Puri A, Grover VP, Agarwala S, Mitra DK and Bhatnagar V:
Initial surgical treatment as a determinant of bladder dysfunction in posterior urethral valves. Pediatr Surg Int 2002;
18: 438.
Bricker EM: Bladder substitution after pelvic evisceration.
Surg Clin North Am 1950; 30: 1511.
Wallace DM: Uretero-ileostomy. Br J Urol 1970; 42: 529.
Johnston JH: Reconstructive surgery of mega-ureter in childhood. Br J Urol 1967; 39: 17.
Wolf JS Jr and Turzan CW: Augmentation ureterocystoplasty.
J Urol 1993; 149: 1095.
Lee BR, Silver RI, Partin AW, Epstein JI and Gearhart JP: A
quantitative histologic analysis of collagen subtypes: the
primary obstructed and refluxing megaureter of childhood.
Urology 1998; 51: 820.
COMMON ILEAL URETER

EDITORIAL COMMENT
The authors present their initial experience with an innovative
surgical technique for bladder reconstruction in 6 patients. The
distal two-thirds of the ureters are used as material for augmentation cystoplasty. The proximal ureters are anastomosed
to a small segment of ileum which is then tapered, tubularized
and reimplanted into the bladder. This technique allows the
surgeon to take advantage of using the redundant ureters,
which would otherwise have been discarded, and to limit the
disadvantages of using bowel. These positive results reaffirm
how the ureter can be used to increase bladder capacity and
reduce storage pressure. More importantly this technique appears to prevent the need for re-augmentation, which has
discouraged the widespread use of ureter for bladder augmentation. Although a segment of bowel is still used in this tech-
1823
nique, it is much smaller than the bowel segment that would

ordinarily be used for augmentation cystoplasty. We should
remain vigilant regarding the long-term risks of interposed
bowel including metabolic alterations, stones and most importantly malignancy. Although the current series is small, the
median followup of the 6 patients is 45 months, which is long
enough to advocate using this innovative technique. We await
reports of larger experiences and anticipate confirmation of its
success.
Lane S. Palmer
Schneider Childrens Hospital
North ShoreLong Island Jewish Health System
New Hyde Park, New York
Computational Model for Predicting the Chance

of Early Resolution in Children With Vesicoureteral Reflux
Matthew J. Knudson, J. Christopher Austin, Moshe Wald, Antoine A. Makhlouf,
Craig S. Niederberger and Christopher S. Cooper*
From the Division of Pediatric Urology, Department of Urology, University of Iowa (MJK, JCA, MW, CSC), Iowa City, Iowa, and
Departments of Urology, University of Minnesota (AAM), Minneapolis, Minnesota, and University of Illinois at Chicago (CSN),
Chicago, Illinois
Purpose: Minimally invasive treatment options and concern regarding long-term antibiotics have increased emphasis on
predicting the chance of early vesicoureteral reflux resolution. Computational models, such as artificial neural networks, have
been used to assist decision making in the clinical setting using complex numeric constructs to solve multivariable problems.
We investigated various computational models to enhance the prediction of vesicoureteral reflux resolution.
Materials and Methods: We reviewed the records of 205 children with vesicoureteral reflux, including 163 females and 42
males. In addition to reflux grade, several clinical variables were recorded from the diagnostic visit. Outcome was noted as
resolved or unresolved at 1 and 2 years after diagnosis. Two separate data sets were prepared for the 1 and 2-year outcomes,
sharing the same input features. The data sets were randomized into a modeling set of 155 and a cross-validation set of 50.
The model was constructed with several constructs using neUROn, a set of C programs that we developed, to best fit
the data.
Results: A linear support vector machine was found to have the highest accuracy with a test set ROC curve area of 0.819 and
0.86 for the 1 and 2-year models, respectively. The model was deployed in JavaScript for ready availability on the Internet,
allowing all input variables to be entered and calculating the odds of 1 and 2-year resolution.
Conclusions: This computational model allowed the use of multiple variables to improve the individualized prediction of
early reflux resolution. This is a potentially useful clinical tool regarding treatment decisions for vesicoureteral reflux.
Key Words: bladder, ureter, vesico-ureteral reflux, computer simulation, computational biology
revious research demonstrates that predicting the

children in whom reflux will spontaneously resolve
and the time to resolution is impacted by several
clinical variables.1 8 Currently the best treatment for vesicoureteral reflux remains unknown. It is largely determined
by parent preference and physician assessment of the
chance of reflux resolution in a child. With the introduction
of minimally invasive antireflux surgical techniques coupled
with concern regarding long-term antibiotic prophylaxis the
importance of predicting which children will achieve resolution within 1 or 2 years of diagnosis has increased. Improved
prediction of when reflux will likely resolve spontaneously
permits more informed management decisions.
Computational models such as artificial neural networks
and SVMs are designed to use complex numerical constructs
to solve multivariable problems and they have been used in
several areas of medicine. These models allow individual
variables to be inputted and based on the previous training
of the model a prediction of the outcome is generated. We
assessed the usefulness of these models in vesicoureteral
reflux by constructing a model that predicts 1 and 2-year
resolution.
* Correspondence: Department of Urology, University of Iowa, 200

Hawkins Dr., 3 RCP, Iowa City, Iowa 52242-1089 (telephone: 319-3848299; FAX: 319-356-3900; e-mail: christopher-cooper@uiowa.edu).
0022-5347/07/1784-1824/0

We reviewed the medical records and initial cystograms of
children diagnosed with primary reflux between 1988 and
2004 at our institution. Eight clinical variables were recorded from the patient diagnostic visit, including age, gender, presenting symptom, reflux grade, laterality, whether
reflux occurred during filling or voiding, initial bladder volume at onset of reflux and complete ureteral duplication. A
total of 205 children, including 42 males and 163 females,
had complete data available and were included in the study.
A total of 18 children were diagnosed with nuclear cystogram, while 187 were diagnosed with voiding cystourethrogram. Also noted was voiding dysfunction, defined as increased incontinence episodes more than that expected for
age, or the prescription of anticholinergic medicine.
Reflux was graded in accordance with the International
Reflux Study.9 By highest grade of reflux there was grades 1
to 5 reflux in 29, 93, 62, 16 and 5 patients, respectively.
Presenting symptoms included febrile UTI, nonfebrile UTI,
prenatal hydronephrosis, screening or other, ie dysuria and
febrile seizures. Duplication was categorized as unilateral in
15 patients and bilateral in 2. Initial bladder volume at
onset of reflux was normalized by age predicted bladder
capacity (PBC) using the equation, PBC (2 age) 30 ml.
Recorded outcomes included spontaneous resolution or
unresolved reflux if the patient underwent surgery or had
persistent reflux on followup cystogram 1 and 2 years after
1824
Vol. 178, 1824-1827, October 2007

Printed in U.S.A.
DOI:10.1016/j.juro.2007.05.093
COMPUTATIONAL MODEL FOR EARLY VESICOURETERAL REFLUX RESOLUTION

diagnosis. Patients underwent followup imaging a median
SD of 1.1 0.9 years after diagnosis. To compensate for
variability in return visits any imaging study within 0.4
years of the specified time frame was included in analysis.
Exclusion criteria included ureterocele, ectopic ureter,
megaureter and neurological or structural bladder abnormalities.
Input features other than age and bladder volume at
onset of reflux were converted into a numerical code and 2
models were created for outcome 1 and 2 years after diagnosis. Input features were identical in the 2 data sets, including gender, presenting symptom, age, laterality, bladder
volume at onset, right/left side grade of reflux, when reflux
occurred on the right/left side, duplicated ureters and voiding dysfunction. The 2 data sets were randomized into a
modeling (training) set of 155 patients and a cross-validation (test) set of 50 with similar outcome frequencies preserved in each set. We used neUROn, a set of C
programs that we developed, to model the data set using
linear and quadratic discriminant function analysis, logistic
regression and neural computation with several investigated architectures. We also modeled the data set with
SVMs using LibSVM, a public domain software library. ROC
curve area was used to assess model accuracy. It was calculated using the statistical method described by Wickens10
and by the traditional trapezoidal method. On a scale of 0 to
1 an ROC value of 1.0 would be optimal and indicate a model
with 100% accuracy.
RESULTS
Table 1 lists patient clinical data. Male patients were diagnosed at a younger median age than females (0.2 1.9 vs
2.5 2.2 years, p 0.001). Reflux resolved spontaneously at
the 1-year followup in 64 (31%) patients, while 76 (37%)
achieved resolution by 2 years. A total of 38 patients (19%)
underwent corrective reflux operations within 2 years of
diagnosis. The reasons for surgery were breakthrough UTI
in 17 cases, decreasing/worsening renal function in 18, severe reflux in 1, parental preference in 1 and a solitary
kidney in 1.
For the 1 and 2-year outcome models the method that
best fit the data was an LSVM with a test set ROC curve
area of 0.819 and 0.86, respectively. Table 2 shows the test
1825
TABLE 2. Test set ROC areas from methods tested with 2-year
outcome data
Model
Test Set ROC Area
LSVM
Radial SVM
Logistic regression
Neural network (3 hidden nodes)
Linear discriminant function analysis
Quadratic data flow analysis
0.86
0.838
0.813
0.795
0.693
0.557
set ROC for all methods tested for 2-year outcome data.
Figure 1 shows how well the predictive variables fit the
2-year model.
One and 2-year models can be deployed in JavaScript,
allowing them to be placed on the Internet. Thus, the model
can be accessed and downloaded, and used to predict resolution based on the clinical input variables discussed.
Figure 2 shows the interface that users would be able to
access. We term this interface a prognostic calculator since
the 9 clinical variables can be input and an individualized
output prediction is generated. The prediction is coded as an
OR for or against resolution by 1 and 2 years. A result of 1
indicates that the odds are in favor of resolution and 0 is
against resolution.
DISCUSSION
The management of vesicoureteral reflux remains a challenge because, while most children spontaneously achieve
reflux resolution without serious complications, some children have significant problems.1 Minimally invasive surgery, open surgery, prophylactic antibiotics and watchful
waiting are the most common treatment options. Minimally
invasive surgery has been proposed to decrease the morbidity and risk of complications1113 but there is a distinct
TABLE 1. Clinical data on 205 patients with primary reflux

No. Pts (%)
Sex:
M
F
Median age at diagnosis SD
Presentation:
Febrile UTI
Nonfebrile UTI
Screening
Other
Bilat
Voiding dysfunction
Ureteral duplication:
Unilat
Bilat
Reflux onset during voiding only
Resolved after diagnosis (yrs):
1
2
42 (20)
163 (80)
1.6 2.2
107 (52)
57 (28)
31 (15)
6 (3)
4 (2)
96 (47)
24 (12)
15 (7)
2 (1)
29 (14)
64 (31)
76 (37)
FIG. 1. Result of LSVM model testing for predicting 2-year vesicoureteral reflux resolution (ROC 0.86).
1826

prostate cancer. Poulakis et al found that prostate cancer
stage could be better predicted using a neural network compared to logistic regression.14 Porter et al looked at 6 clinical
variables to predict positive biopsy and developed a neural
network with an ROC curve area of 0.746.15 Other groups
have used neural networks in bladder cancer16 and erectile
dysfunction17 with similar accuracy. An even higher ROC
achieved by our computational model compared to that in
other published reports demonstrates that neural networks
can be used to aid the prediction of reflux resolution in a
similar way.
The retrospective nature of our study creates several
limitations. Since the decision regarding surgical intervention was based on physician and parent choice, 38 patients
underwent surgery within 2 years of diagnosis. While the
majority of these patients had strong indications for operative intervention, there exists the potential for selection bias
with any retrospective review. Also, our patients did not
routinely undergo urodynamic screening for voiding dysfunction. Although we assessed voiding dysfunction during
the medical record review, it remains possible that some
patients with abnormal bladder function were not recognized and, therefore, they may have had a component of
secondary reflux. Finally, although the model was constructed from a large number of patients from a single
institution, additional analysis from multiple institutions
would strengthen and further validate this computational
model.
FIG. 2. Interface based on 2-year resolution model. A, data input

page. B, resolution prediction page.
CONCLUSIONS
possibility that many children undergo unnecessary treatment. A more individualized prediction of which children
would achieve reflux resolution early may help with management decisions. Reflux grade remains the major clinical
variable used to predict reflux resolution1 but other variables also impact resolution.2 8 To our knowledge for the
first time our computational model assimilates all of these
factors in a user friendly method with a high degree of
accuracy that can aid clinical decision making.
An SVM may be considered a statistical model with a
complex decision surface. We included this method in modeling our data set and present a model with high goodness of
fit for the data set. For the 1 and 2-year models the LSVM
resulted in the best data fit. The LSVM model generated an
optimal, nonlinear separating surface between the 2 outcomes by focusing on the training cases that were most
difficult to classify. At this time to our knowledge feature
extraction of SVM models to determine the significance of
each individual input parameter to model accuracy is not
available. However, the LSVM model is preferable, given its
higher accuracy compared to other methods (table 2). We
previously described hazard regression to analyze several of
these variables, including grade, age, bladder volume at
onset of reflux, gender, laterality and presenting symptoms.7
The incorporation of all of these variables to generate an
accurate outcome prediction is the main advantage of computer modeling and it provides a powerful tool for clinical
decision making in different reflux scenarios.
Neural networks have been used in other areas of urology
to help predict outcomes using several clinical variables. A
major application of computational models in urology is in
Our computational model accurately predicts the odds of

reflux resolution at 1 and 2 years on an individualized basis
through the use of multiple variables. Model reliability,
ready availability and ease of use offer significant potential
clinical usefulness for reflux management.

LSVM linear SVM
SVM support vector machine
UTI urinary tract infection
REFERENCES
1.
2.
3.
4.
5.
Mozley PD, Heyman S, Duckett JW, Turek P, Snyder H,
Maislin G et al: Direct vesicoureteral scintigraphy: quantifying early outcome predictors in children with primary
reflux. J Nucl Med 1994; 35: 1602.
Papachristou F, Printza N, Doumas A and Koliakos G: Urinary
bladder volume and pressure at reflux as prognostic factor
of vesicoureteral reflux outcome. Pediatr Radiol 2004; 34:
556.
Greenfield SP, Ng M and Wan J: Resolution rates of low grade
vesicoureteral reflux stratified by patient age at presentation. J Urol 1997; 157: 1414.
Schwab CW Jr, Wu HY, Selman H, Smith GH, Synder HM 3rd
and Canning DA: Spontaneous resolution of vesicoureteral
reflux: a 15-year perspective. J Urol 2002; 168: 2594.

6.
Brophy MN, Austin PE, Yan Y and Coplen DE: Vesicoureteral

reflux and clinical outcomes in infants with prenatally detected hydronephrosis. J Urol 2002; 168: 1716.
7. Knudson MJ, Austin JC, McMillian Z, Hawtrey CE and Cooper
CS: Predictive factors of early spontaneous resolution in
children with primary vesicoureteral reflux. J Urol 2007;
178: 1684.
8. McMillan ZM, Austin JC, Hawtrey CE, Knudson MJ and
Cooper CS: Bladder volume at onset of reflux on initial
cystogram predicts spontaneous resolution. J Urol 2006;
176: 1838.
9. Lebowitz RL, Olbing H, Parkkulainen KV, Smellie JM and
Tamminen-Mobius TE: International system of radiographic
grading of vesicoureteric reflux. International Reflux Study in
Children. Pediatr Radiol 1985; 15: 105.
10. Wickens TD: Elementary Signal Detection Theory. New York:
Oxford University Press 2002.
11. Capozza N, Lais A, Nappo S and Caione P: The role of endoscopic treatment of vesicoureteral reflux: a 17-year experience. J Urol 2004; 172: 1626.
12. Lackgren G, Wahlin N, Skolenberg E and Stenberg A: Longterm followup of children treated with dextranomer/hyal-
1827
uronic acid copolymer for vesicoureteral reflux. J Urol 2001;

166: 1887.
13. Peters C: Endoscopic treatment of vesicoureteral reflux: have
we found a new gold standard. Nat Clin Pract Urol 2006; 3:
360.
14. Poulakis V, Witzsch U, De Vries R, Emmerlich V, Meves M and
Altmannsberger M: Preoperative neural network using
combined magnetic resonance imaging variables, prostate
specific antigen and Gleason score to predict prostate cancer stage. J Urol 2004; 172: 1306.
15. Porter CR, Crawford D, Burtsch G, Presti J, Gamito and:
Predicting prostate biopsy outcomes, an international prospective multicenter model involving 4,788 men. J Clin
Oncol 2004; 22: 411.
16. Catto J, Abbod M, Linkons D and Handy F: Bladder cancer
neuro-fuzzy modeling, an accurate and interpretable
method for predicting bladder cancer progression. J Urol
2006; 175: 974.
17. Wald M, Seffel A, Ross L, Mohamed M and Niederberger C:
Computational model for detection of erectile dysfunction.
J Urol 2005; 173: 167.

542515

Hochgeladen von

Dokumentinformationen

Copyright

Verfügbare Formate

Dieses Dokument teilen

Dokument teilen oder einbetten

Freigabeoptionen

Stufen Sie dieses Dokument als nützlich ein?

Sind diese Inhalte unangemessen?

Copyright:

Verfügbare Formate

542515

Hochgeladen von

Copyright:

Verfügbare Formate

The Journal of Urology

Volume 178, Issue 4, Supplement 1, Pages.1563-1827 (October 2007)

This Issue in Pediatric Urology

Determining the Proper Stent Length to Use in Children: Age Plus 10

Prenatal Magnetic Resonance Imaging Demonstrates Linear Growth of the Human

Minimally Invasive Open Renal Surgery

Transperitoneal Laparoscopic Pyeloplasty for Primary Repair of Ureteropelvic

Clinical Prize Finalists

Relationship of Varicocele Grade and Testicular Hypotrophy to Semen Parameters

Surgical Treatment of Unilaterally Undescended Testes: Testicular Growth After

Endoscopic Treatment of Vesicoureteral Reflux Associated With Ureterocele

Nerve Sparing Ventral Clitoroplasty: Analysis of Clitoral Sensitivity and Viability

Challenges for the Pediatric Urologic Surgeon in the 21st Century

Esthetic Outcomes of Genitoplasty in Males Born With Bladder Exstrophy and

Urachal Anomalies: A Longitudinal Study of Urachal Remnants in Children and

Results of Bladder Neck Reconstruction After Newborn Complete Primary Repair

A Comparison of the Monti and Casale (Spiral Monti) Procedures

Endoscopic Injection of Bulking Agents in Children With Incontinence:

First Stage Approximation of the Exstrophic Bladder in Patients With Cloacal

Basic Science Research Finalists

In Utero Exposure to Benzophenone-2 Causes Hypospadias Through an Estrogen

Urothelial Inhibition of Transforming Growth Factor- in a Bladder Tissue

Association of Transforming Growth Factor-1 Gene Polymorphism With

Pediatric Urology Medal

Traumatic Obliterative Urethral Strictures in Pediatric Patients: Failure of the Cut

Long-Term Patient Reported Outcome of Urinary Symptoms After Hypospadias

Split Prepuce In Situ Onlay Hypospadias Repair: 17 Years of Experience

The Posterior Sagittal Approach for Recurrent Genitourinary Pathology

3-Dimensional Magnetic Resonance Genitography: A Different Look at Cloacal

Intravesical Electrotherapy for Neurogenic Bladder Dysfunction: A 22-Year

Predictive Factors of Early Spontaneous Resolution in Children With Primary

Ipsilateral Ureteroureterostomy in the Surgical Management of the Severely

Outcome Analysis of Severe Chordee Correction Using Tunica Vaginalis as a Flap

Small Intestinal Submucosa for Corporeal Body Grafting in Severe Hypospadias

Multivariate Comparison of the Efficacy of Intraureteral Versus Subtrigonal

Histopathological Changes Associated With Dextranomer/Hyaluronic Acid

New Contralateral Vesicoureteral Reflux After Endoscopic Correction of

Endoscopic Treatment of Moderate and High Grade Vesicoureteral Reflux in

Scrotal Exploration for Unilateral Nonpalpable Testis

Incidence of Testicular Ascent in Boys With Retractile Testes

Further Evidence for Spontaneous Descent of Acquired Undescended Testes

Ultrasonography Underestimates the Volume of Normal Parenchyma in Benign

Doping in Sports: The Pediatric Perspective

Clinical Urological Outcomes Following Primary Tethered Cord Release in

Treatment of Nocturnal Enuresis in Children With Attention Deficit Hyperactivity

Spinal Cord Magnetic Resonance Imaging for Investigation of Nonneurogenic

Development of an Objective Score to Quantify the Pediatric Cystometrogram

Office Management of Pediatric Primary Nocturnal Enuresis: A Comparison of

Incontinence Symptom Index-Pediatric: Development and Initial Validation of a

The Durability of Intravesical Oxybutynin Solutions Over Time

Tumor in Bladder Reservoir After Gastrocystoplasty

The Use of Botulinum Toxin A Injection for the Management of External

Which is BetterRetroperitoneoscopic or Laparoscopic Dismembered Pyeloplasty

Corporeal Sparing Dismembered Clitoroplasty: An Alternative Technique for

Basic Science Research

Nerve Sparing Ventral Clitoroplasty Preserves Dorsal Nerves in Congenital

The Effect of Oxybutynin on Structural Changes of the Obstructed Guinea Pig

Artificial Urinary Sphincters Placed After Posterior Urethral Distraction Injuries in

Long-Term Continence Outcomes After Immediate Repair of Pediatric Bladder

The Common Ileal Ureter: A New Technique for Compliant

Computational Model for Predicting the Chance of Early Resolution in Children

Dx/HA is stable and safe as long as 22 months after

PREDICTIVE FACTORS OF EARLY

TUMOR IN BLADDER RESERVOIR