Accepted Manuscript

High GMS score hypospadias: outcomes after one- and two-stage operations

Jonathan Huang, Lael Rayfield, Bruce Broecker, Wolfgang Cerwinka, Andrew Kirsch,
Hal Scherz, Edwin Smith, James Elmore

PII: S1477-5131(16)30404-1
DOI: 10.1016/j.jpurol.2016.11.022
Reference: JPUROL 2407

To appear in: Journal of Pediatric Urology

Received Date: 9 August 2016

Accepted Date: 24 November 2016

Please cite this article as: Huang J, Rayfield L, Broecker B, Cerwinka W, Kirsch A, Scherz H, Smith
E, Elmore J, High GMS score hypospadias: outcomes after one- and two-stage operations, Journal of
Pediatric Urology (2017), doi: 10.1016/j.jpurol.2016.11.022.

This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to
our customers we are providing this early version of the manuscript. The manuscript will undergo
copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please
note that during the production process errors may be discovered which could affect the content, and all
legal disclaimers that apply to the journal pertain.
 ACCEPTED MANUSCRIPT

High GMS score hypospadias: outcomes after one- and two-stage operations
Jonathan Huanga, Lael Rayfieldb, Bruce Broeckerc, Wolfgang Cerwinkac, Andrew
Kirschc, Hal Scherzc, Edwin Smithc, and James Elmorec,*

a
Department of Urology, Emory University School of Medicine and Children’s
Healthcare of Atlanta, Atlanta, GA, USA

PT
b
Department of Statistics, Children’s Healthcare of Atlanta, Atlanta, GA, USA
c
Department of Pediatric Urology, Emory University School of Medicine and Children’s

RI
Healthcare of Atlanta, Atlanta, GA, USA

SC
* Corresponding author. 5730 Glenridge Drive, Suite #200, Sandy Springs, GA 30328,
USA.

U
E-mail address: jmelmoremd@gmail.com
AN
Summary Introduction: Established criteria to assist surgeons in deciding between a
one- or two-stage operation for severe hypospadias are lacking. While anatomical
M

features may preclude some surgical options, the decision to approach severe
hypospadias in a one- or two-stage fashion is generally based on individual surgeon
D

preference. This decision has been described as a dilemma as outcomes range widely and
TE

there is lack of evidence supporting the superiority of one approach over the other.
Objectives: The aim of this study is to determine whether the GMS hypospadias score
may provide some guidance in choosing the surgical approach used for correction of
EP

severe hypospadias.
Study design: GMS scores were preoperatively assigned to patients having primary
C

surgery for hypospadias. Those patients having surgery for the most severe hypospadias
AC

were selected and formed the study cohort. The records of these patients were reviewed
and pertinent data collected. Complications requiring further surgery were assessed and
correlated with the GMS score and the surgical technique used for repair (one-stage vs.
two-stage).
Results: Eighty-seven boys were identified with a GMS score (range 3-12) of 10 or
higher. At a mean follow-up of 22 months the overall complication rate for the cohort
after final planned surgery was 39%. For intended one-stage procedures (n=48) an
acceptable result was achieved with one surgery for 28 patients (58%), with two surgeries

1
 ACCEPTED MANUSCRIPT

for 14 (29%), and with three to five surgeries for six (13%). For intended two-stage
procedures (n=39) an acceptable result was achieved with two surgeries for 26 patients
(67%), three surgeries for eight (21%), and four surgeries for three (8%). Two other
patients having two-stage surgery required seven surgeries to achieve an acceptable
result. Complication rates are summarized in the Table. The complication rates for GMS
10 patients were similar (27% and 33%, p=0.28) for one- and two-stage repairs,

PT
respectively. GMS 11 patients having a one-stage repair had a significantly higher
complication rate (69%) than those having a two-stage repair (29%) (p=0.04). GMS 12

RI
patients had the highest complication rate with a one-stage repair (80%) compared with a
complication rate of 37% when a two-stage repair was used (p=12).

SC
Conclusions: Guidelines to help standardize the surgical approach to severe hypospadias
are needed. Staged surgery for GMS 11 and 12 patients may result in a lower

U
complication rate but may not reduce the number of surgeries required for an acceptable
AN
result. Although further study is needed, the GMS score may be helpful for establishing
such criteria.
KEYWORDS
M

Hypospadias;
GMS;
D

Proximal;
TE

Complications

Table. Complication rates for patients with GMS 10, 11, and 12 hypospadias
EP

One-stage repair Two-stage repair

GMS score p-value
N = 48 N = 39
C

10 8/30 (27%) 2/6 (33%) 0.28

AC

11 9/13 (69%) 4/14 (29%) 0.04

12 4/5 (80%) 7/19 (37%) 0.12

Introduction
Despite the large number of techniques available to repair proximal hypospadias,
complication rates remain considerable [1]. While anatomical features may preclude
some surgical options, the decision to approach severe hypospadias in a one- or two-stage
fashion is generally based on individual surgeon preference [2]. This decision has been

2
 ACCEPTED MANUSCRIPT

described as a dilemma as outcomes range widely and there is a lack of evidence

supporting the superiority of one approach over the other [3,4]. Indeed at this time, even
the criteria used to define “severe” hypospadias remain to be established [5].
Recently, the GMS hypospadias score was developed as a means to standardize
the evaluation of the hypospadias complex [6,7]. This method grades the severity of the
hypospadias complex based on the size of the glans, features of the urethral plate,

PT
position of the meatus, and the severity of chordee (Fig. 1). Unfavorable characteristics
are given higher scores (1-4) and individual component scores for G, M, and S are then

RI
summed to give the total GMS score (range 3-12). These features were chosen as they
have been shown to, or may, impact outcomes following repair [8,9]. Prior studies have

SC
shown that GMS scoring has high inter-rater reproducibility and that individual
component and total scores correlate directly with the risk of a complication following

U
hypospadias surgery [6,7].
AN
In this study we assess the outcomes of a group of patients having surgery for
severe hypospadias and categorize results by GMS score and the surgical approach used
(one-stage vs. two-stage). We anticipate that the data may be of some help for guiding the
M

approach to surgery for severe hypospadias.

Materials and methods
D

Following IRB approval, a database containing the GMS scores of patients having initial
TE

hypospadias surgery at our institution was created. This database was queried to identify
those patients having surgery for the most severe hypospadias (highest quartile: GMS 10,
11, and 12) who formed the study cohort. The electronic medical records of these
EP

patients, including hospital and clinic notes, were then reviewed and pertinent data
collected including GMS scores, type of repair, and the occurrence of any complications
C

following surgery. Complications requiring further surgery were assessed and correlated
AC

with the GMS score and the surgical approach chosen (one-stage vs. two-stage). Only
patients with at least 6 months follow-up from their final planned surgery were included
in this analysis. Also, given the purpose of this study, only urethral complications were
considered including fistula, meatal stenosis, dehiscence, and stricture.
Descriptive statistics were calculated for all variables of interest and included
means and standard deviations or counts and percentages, as appropriate. Complication
rates were compared among GMS score subgroups using chi-square tests. In instances
where the expected number of events was small (less than five), a Fisher’s exact test was

3
 ACCEPTED MANUSCRIPT

used in place of the chi-square test. Statistical analyses were conducted using SAS v. 9.4
(Cary, NC, USA) and statistical significance was assessed at the 0.05 level.
Results
Of the 478 patients with complete data, 87 boys with a median age of 8 months (IQR
6-12) were identified with a GMS score of 10 or higher and at least 6 months follow-up
(median 23 months, range 7-34) after final planned surgery. The one-stage techniques

PT
used included tubularized incised plate (71%), Duckett tube (18%), Thiersch–Duplay
(8%), and Flip-flap (3%). All two-stage repairs were performed using Byar’s flaps

RI
followed by completion urethroplasty 6-12 months later. The overall complication rate
for the cohort after their final planned surgery was 39% (34/87). Complications included

SC
urethrocutaneous fistula in 18 (21%) patients, meatal stenosis in four (5%), urethral
diverticulum in four (5%), and glans dehiscence in three (3%). Urethrocutaneous fistula

U
and meatal stenosis with distal urethral stricture occurred together in five patients (6%).
AN
Preoperative testosterone was given to 15 patients (17%) and the complication rate for
this group of 46% (7/15) did not differ statistically from the complication rate overall.
For intended one-stage procedures (n=48) an acceptable result was achieved with
M

one surgery for 28 (58%), with two surgeries for 14 (29%), and with three to five
surgeries for six (13%). For intended two-stage procedures (n=39) an acceptable result
D

was achieved with two surgeries for 29 (74%), three surgeries for six (15%), and four
TE

surgeries for two (5%). Two other patients having two-stage surgery required seven
surgeries to achieve an acceptable result. Patients undergoing two-stage repairs had
higher average M-scores than patients having one-stage repairs (3.8 vs. 3.5) but this
EP

difference did not reach statistical significance (p=0.31).

Details regarding GMS scores, the surgical approach used, and the incidence of
C

complications are given in the Table in the summary. The incidence of complications for
AC

GMS 10 patients was similar for one-stage (27%) and two-stage (33%) surgeries
(p=0.28). For GMS 11 patients the incidence of complications was higher for patients
having one-stage surgery (69%) compared with those having two-stage surgery (29%)
(p=0.04). There were only five patients with GMS 12 hypospadias who had repair
attempted in one stage. While the numbers prevented statistical significance, these
patients had the highest complication rate (80%), compared with a complication rate of
37% when a two-stage approach was used.
Discussion

4
 ACCEPTED MANUSCRIPT

Hypospadias is a complex of anomalies characterized by an abnormally proximal meatus,

deficient ventral tissues, and penile curvature. The clinical presentation of hypospadias
ranges from mild to severe and the wide array of possible anatomical permutations makes
every patient unique. The surgical techniques used to repair hypospadias are almost as
diverse with hundreds of described methods. Given the large number of patient and
operative factors that may influence outcomes, it is not surprising that surgical results

PT
differ significantly. This is especially true for proximal hypospadias where, regardless of
whether a single or multistage surgery is used, complication rates range from 8% to 70%

RI
[4]. In an effort to standardize how the hypospadias complex is evaluated and described,
the GMS score was developed. This scoring method takes into account several features of

SC
the hypospadias complex including, but not limited to, meatal position. With the ability to
more uniformly quantify hypospadias severity, we investigated whether the GMS score

U
could help guide the decision-making process in management of severe hypospadias.
AN
Severe hypospadias remains a challenging entity [1]. At our institution, a one-
stage procedure is favored if deemed possible by the operating surgeon while a two-stage
procedure is reserved for extenuating circumstances such as a need to divide the urethral
M

plate and/or graft the ventral corpora. However, one-stage techniques are still available in
these instances and therefore the surgeon is left to make this decision based on personal
D

judgment and/or preference. This decision may be a difficult one as each technique has
TE

advantages and disadvantages. The overall complication rate of 39% is comparable with
reports from other centers [1].
Our data suggest that for GMS 10 hypospadias, complication rates are similar
EP

with either a one- or two-stage repair (~30%). As primary repair for these patients
required fewer surgeries per patient on average, this option when feasible, would be
C

preferable. Conversely, for GMS 11 hypospadias outcomes after primary repair were
AC

significantly worse than following staged surgery (69% vs. 29%). Again while outcomes
were better, a staged approach still required more surgeries, on average, to achieve an
acceptable result. Primary surgery for GMS 12 hypospadias had the highest complication
rate (80%) and compared with a complication rate of 37% for staged surgery. However,
as a result of the low number of patients with GMS 12 hypospadias that could be repaired
in one stage, statistical significance was not reached between the two techniques.
Nonetheless our data suggest that one-stage surgery for GMS 10 hypospadias and two-
stage surgery for GMS 11 and 12 hypospadias may be preferable and a reasonable

5
 ACCEPTED MANUSCRIPT

hypothesis to test with future study. Indeed, further study with larger patient numbers
would be useful for developing more definitive criteria. However, currently, the decision
to use a one- or two-stage technique is based only on individual surgeon
preference/judgment, which could be an obstacle to more uniform and reproducible
outcomes data.
We recognize that there are several limitations to our study. First, it is

PT
retrospective in nature and therefore has all the usual limitations of such studies. In
addition to a relatively small patient cohort (n=87), different one-stage surgical

RI
techniques were used including long tubularized incised plate procedures, Thiersch-
Duplay repairs, and Duckett tubes. Therefore it is possible that outcomes for one-stage

SC
surgeries were more a reflection of the technique used rather it being performed in one
stage. However, complications were evenly distributed between the various repair types

U
and assessing outcomes of specific one-stage techniques was beyond the scope of this
AN
study. Ancillary procedures also made our study cohort more heterogeneous than ideal.
For example some patients had dorsal plication, corporeal grafting, or a combination of
these while other patients did not. As comparing the success rates of the various methods
M

to correct chordee was not the focus of this study, recurrent chordee was not included in
our analysis. Some patients may have had reoperation for recurrent chordee making our
D

outcomes appear better than they would if this complication had been included, and this
TE

is a point that deserves emphasis. What constitutes an “acceptable” outcome is also

subjective and may have differed somewhat from surgeon to surgeon. Future studies
would benefit from having standard outcome measures such as HOSE criteria [8]. Lastly,
EP

a primary criticism of the GMS hypospadias scoring is its subjective nature. While inter-
rater reliability has been shown to be good at our institution using a computer enhanced
C

visual learning (CEVL) platform, further validation of the scoring method is still needed.
AC

Conclusions
The GMS score may be useful in guiding the decision-making process in the
management of severe hypospadias. Although this is a retrospective study and patient
numbers are relatively small, it appears that GMS 10 hypospadias can be repaired in one
stage with a relatively low complication rate while GMS 11 and 12 hypospadias have a
higher complication rate following primary repair. A staged approach for these patients
may result in fewer complications but may not reduce the total number of surgeries
needed to achieve an acceptable outcome. Further study with prospectively acquired data

6
 ACCEPTED MANUSCRIPT

is needed.
Conflict of interest
None.
Funding
None.
References

PT
[1] Pippi Salle JL, Sayed S, Salle A, Bagli D, Farhat W, Koyle M, Lorenzo AJ. Proximal
hypospadias: A persistent challenge. Single institution outcome analysis of three surgical

RI
techniques over a 10-year period. J Pediatr Urol 2016;12(1):28.
[2] Dason S, Wong N, Braga LH. The contemporary role of 1 vs. 2-stage repair for

SC
proximal hypospadias. Transl Androl Urol 2014;3(4):347.
[3] Badawy H, Fahmy A. Single- vs. multi-stage repair of proximal hypospadias: The

U
dilemma continues. Arab J Urol 2013;11:174.
AN
[4] Canning DA. Re: single- vs multi-stage repair of proximal hypospadias: the dilemma
continues. J Urol 2014;191(6):1873.
[5] Snodgrass W, Macedo A, Hoebeke P, Mouriquand PD. Hypospadias dilemmas: a
M

round table. J Pediatr Urol 2011;7:145-57.

[6] Merriman LS, Arlen AM, Broecker BH, Smith EA, Kirsch AJ, Elmore JM. The GMS
D

hypospadias score: assessment of inter-observer reliability and correlation with post-

TE

operative complications. J Pediatr Urol 2013;9(6):707-12.

[7] Arlen AM, Kirsch AJ, Leong T, Broecker BH, Smith EA, Elmore JM. Further
analysis of the Glans-Urethral Meatus-Shaft (GMS) hypospadias score: Correlation with
EP

postoperative complications. J Pediatr Urol 2015;11(2):71.

[8] Liu MM, Holland AJ, Cass DT. Assessment of postoperative outcomes of
C

hypospadias repair with validated questionnaires. J Pediatr Surg 2015;50(12):2071.

AC

[9] Elmore JM, Maizels M. CEVL e-learning teaches GUMS method to “score”
hypospadias preoperatively and predict postoperative outcomes. J Pediatr Urol
2015;11(5):234.

Figure 1. GMS criteria and representative photos. Note: G, M, and S scores summed to
give GMS score (range 3-12).

7
Figure: GMS Criteria and Representative Photos. Note: G, M, and S scores summed to give GMS
score (Range 3 to 12) ACCEPTED MANUSCRIPT

PT
RI
U SC
AN
M
D
TE
C EP
AC