Seminars in Pediatric Surgery (2012) 21, 354-363

Reoperative surgery for Hirschsprung disease

Matthew W. Ralls, MD, Arnold G. Coran, MD, Daniel H. Teitelbaum, MD

From the C.S. Mott Children’s Hospital, University of Michigan Hospital and Health Systems, Ann Arbor, Michigan.

KEYWORDS Despite most children undergoing a successful pull through for Hirschsprung disease, a small portion
Hirschsprung disease; of children are left with persistent stooling issues. Most of these stooling issues can be addressed by
Aganglionosis; nonoperative approaches. However, in a small group of remaining children, a reoperation may be
Stricture; necessary. Most children who may need a redo pull-through procedure may have a persistent area of
Enterocolitis; aganglionosis, unremitting enterocolitis, or a torsion or stricture of the pull-through segment. Each of
Pull through; these influences the approach the surgeon must take to correct the presenting problem. The chapter
Reoperation details the diagnostic approach as well as the operative techniques, which best deal with each of these
complications.
© 2012 Elsevier Inc. All rights reserved.

Congenital aganglionosis was named after Härold Long-term postoperative complications

Hirschsprung who described the clinical features in 1886;
however, there is new evidence which suggests there is
recognition of the disease process dating several centuries
before this.1 Since Swenson and Bill2 described successful
treatment of Hirschsprung disease (HD) in 1948, there have
been many modifications and advances in technique leading
to improved quality of life for those affected by HD. Pres-
ently, the most common techniques are a transanal approach
with or without laparoscopic assistance, a modified Soave
pull-through and a Duhamel’s retrorectal pull-through pro-
cedures. Many studies have attempted to tease out which of
these is superior without consensus.3-5 Each of these ap-
proaches can ultimately have very good long-term results.
However, regardless of operative technique, many patients
have long-term postoperative complications; and in some of
these children, the type and extent of the morbidity may
require reoperation to adequately address the child’s prob-
lems.
Address reprint requests and correspondence: Daniel H. Teitel-
baum, MD, Section of Pediatric Surgery, C.S. Mott Children’s Hospital
F3970, University of Michigan, Ann Arbor, MI 48109-0245.
E-mail: dttlbm@umich.edu.
following pull though
A significant number of children whom undergo a success-
ful pull through will develop postoperative complications.
The most common and serious long-term complications
after definitive treatment for HD can be divided into 3
groups: soiling/incontinence, persistent problems with the
passage of stool (eg, constipation), and recurrent Hirsch-
sprung-associated enterocolitis (HAEC).6 The contributing
factors resulting in these problems can be described as
either pathologic or anatomic. The pathologic causes in-
clude residual aganglionosis or transitional zone pathology.
The anatomic causes comprise stricture, retained dilated
segment, obstructing Duhamel pouch or Soave (agangli-
onic) muscular cuff, and finally a twisted pull through.
Lastly, immunologic and or physiological causes may con-
tribute to the development of recurrent episodes of HAEC.
Recent reviews have described a wide range in the inci-
dence of incontinence, which may range from 3% to 30%.
Although most of these episodes can be treated with med-
ical care, some have required an appendicostomy and, at
times, a permanent stoma.7 Postoperative HAEC ranges
from 10% in some groups up to 45% in our own group of
patients. However, the differences between series maybe
more related to differences in definitions between centers.

1055-8586/$ -see front matter © 2012 Elsevier Inc. All rights reserved.
http://dx.doi.org/10.1053/j.sempedsurg.2012.07.011

Downloaded for FK Universitas Diponegoro (trialckundip01@gmail.com) at ClinicalKey Global Flex Package Trial from ClinicalKey.com by Elsevier on May 26, 2019.
For personal use only. No other uses without permission. Copyright ©2019. Elsevier Inc. All rights reserved.
Ralls et al Re-operative Surgery for Hirschsprung Disease
Despite conservative treatment working in most cases, such
approaches prove unsuccessful in nearly 20% of children.
Many reports indicate that patients will outgrow most of
these complications by the age of 5 years for unexplained
reasons, without the need for further treatment. Unfortu-
nately, this is not always the case; some children may
develop treatment-resistant symptoms or recurrent episodes.
In these latter cases, further intervention is required. Pa-
tients presenting with stooling problems require thorough
and thoughtful workup. These complications are described
brief. The reader should refer to earlier chapters on com-
plications following pull-through procedures for further de-
tail.
Incontinence
Soiling and incontinence are theoretically preventable by
use of proper surgical techniques; however, many patients
are affected even in the most experienced hands. Inconti-
nence has been reported in variable percentages and de-
grees. Rates range from as little as none in several series8-10
to as high as 76% in 1 series.11 Soiling may simply be due
to a loss of rectal sensation. Sensory rectal mucosa is critical
for differentiating between gas, liquid, and solid stool, and
it can be compromised after a very low anastomosis. Incon-
tinence may also be due to scaring or iatrogenic damage
to the anal sphincters. The surgeon must be cognizant of
the fact that the level of the internal anal sphincters (IAS,
the major contributor to continence) can be identified as the
level of the dentate line, and thus all anastomoses must be
performed at least 5 mm or more above this level.
However, the most common cause of soiling after a pull
through is actually because of chronic constipation with
encopresis. The treatment of incontinence ranges from a
simple bowl management program to the placement of
a permanent stoma depending on the underlying cause. A
critical challenge for the surgeon evaluating a child with
soiling is differentiating true incontinence, which is typi-
cally managed nonoperatively, from encopresis or HAEC.
The latter 2 warrant continued workup and possible opera-
tive intervention.
Persistent stooling problems
Persistent stooling problems include recurrent distension,
bloating, and constipation. Constipation is a common com-
plaint after any pull-through procedure, although typically it
does not develop for weeks to months post-operatively.
Recent reports have recognized that obstructive symptoms
are occurring in 9%12,13-40% of children after what appears
to be a technically excellent operation. A common theme in
the literature is that symptoms improve with greater follow-
up, being reported as normal in 58% of those at ⬍5 years of
follow-up and 88% of those at greater than 15 years of
follow-up.14 The fact that, many children have persistent
problems with their IAS or more proximal tissue even after
Downloaded for FK Universitas Diponegoro (trialckundip01@gmail.com) at ClinicalKey Global Flex Package Trial from ClinicalKey.com by Elsevier on May 26, 2019.
For personal use only. No other uses without permission. Copyright ©2019. Elsevier Inc. All rights reserved.
355
removal of all aganglionic tissue suggest that our under-
standing of the pathophysiology of HD is still incomplete.
There are several causes of persistent stooling after a pull
through. Anatomic causes including an anastomotic stric-
ture or retained spur after Duhamel’s procedure are com-
mon.15 Other contributing processes, such as retained or
acquired aganglionosis,16 rarely hypoganglionosis17 or tran-
sition zone pathology, which can be associated with either
focal or general motility disorders are also frequently en-
countered. Less commonly, IAS achalasia (IASA), and
“functional megacolon,” a chronically dilated distal rectum
secondary to chronic difficulty with passing bowel move-
ments, can lead to persistent stooling problems.18
In some children there is no identifiable cause for their
symptoms. Responses to the use of botulinum toxin injec-
tions for relaxation of the sphincters are widely vari-
able.19-22 Many of these children suffer from stool-holding
behavior and are best treated by also using a bowel man-
agement regimen, consisting of laxatives, enemas, and be-
havior modification, including support for the child and
family.23 The approach to address all above causes varies,
and one should be aware of all of these various possibilities
to be able to successfully manage a complicated HD case.
Hirschsprung disease-associated enterocolitis
HAEC remains the most common cause of morbidity and
death in HD.24 The exact etiology and pathogenesis are still
unclear.24,25,26 There is no clinical factor or test that is
available to predict the development of this disorder and no
single strategy for prophylaxis from this major complica-
tion. In a multicenter review of primary endorectal pull
through (ERPT), logistic regression analysis of risk factors
for the development of enterocolitis showed that stricture
development was a significant risk factor.15 Based on this, it
is suggested that in such cases, a redo pull through may, at
times, be beneficial in controlling recurrent HAEC in some
patients once they undergo a complete workup.
Principles for workup of postoperative
complications following a pull-though
HD cases should be assessed thoroughly by experienced
physicians at an institution that has the infrastructure to
evaluate, diagnose, and treat these complex problems. Me-
ticulous documentation should be maintained in any HD
case as these records will prove useful for routine follow-up
and essential for those requiring reoperation. In the case of
referral, documentation of the neonatal period, associated
anomalies, type of presentation, time and method of diag-
nosis, preoperative course, type of procedure, pathologic
material, and detailed postoperative management should be
carefully reviewed. An algorithm should be tailored for each
case according to the type and nature of the complication
(Figure 1).
356
Figure 1 Algorithm for the workup of patients presenting with constipation or recurrent enterocolitis after a pull-through procedure.
History and physical examination
A comprehensive physical examination to assess the general
condition, developmental milestones and growth curve, in-
cluding recognition of associated syndromes and anomalies
that may impact the outcome is important. The surgeon
should elicit a detailed diet and stooling history as well as
the measures taken to manage the presenting issue. Abdom-
inal inspection and palpation is necessary to get an appre-
ciation for the severity of disease. The examiner should note
the magnitude of distension as well as any signs of perito-
nitis. Often the examiner can palpate impacted stool.
Rectal examination is essential and can yield useful in-
formation. First, rectal tone should be assessed. Abnormal
tone of the anal sphincter can allude to underlying pathol-
ogy. Increased tone is suggestive of IASA. Decreased or
lack of tone may indicate functional incontinence.
On insertion and/or removal of the examiners finger, a
gush of gas and liquid stools typically implies persistent
obstructive symptoms or ongoing HAEC. Assessment of the
anastomosis for evidence of stricture or narrowing is im-
portant. The rectal vault is also evaluated. A rectum full of
stool might suggest the diagnosis of encopresis. Those chil-
dren with no further evidence of stricture may benefit from
behavioral/training strategies to correct stool-holding be-
havior. Occasionally, a retained spur after a Duhamel proce-
dure is noted, although often this may be beyond the palpating
finger, and require an endoscopy to diagnose. Findings of a
stricture higher up might suggest ischemia or twist of the pull
through; however, most these obstructions are also above the
limits of a digital examination. Finally, a firm tightness around
Downloaded for FK Universitas Diponegoro (trialckundip01@gmail.com) at ClinicalKey Global Flex Package Trial from ClinicalKey.com by Elsevier on May 26, 2019.
For personal use only. No other uses without permission. Copyright ©2019. Elsevier Inc. All rights reserved.
Seminars in Pediatric Surgery, Vol 21, No 4, November 2012
the rectum is suggestive of a previous anastomotic leak at the
time of the original pull though, and often is a strong indication
that a redo pull through will be needed.
Physical examination may suggest an inappropriately
placed anastomosis distal to the dentate line. At times, resting
anal sphincter pressures using anal manometry can prove quite
useful in distinguishing these different disorders.
Imaging
Plain film
An upright abdominal film should be done to evaluate
degree of obstruction, but may also suggest a “cut-off sign”
suggestive of active or chronic HAEC.27
Contrast enema
A radiographic contrast enema can be useful. It usually
can differentiate between anastomotic stenosis or stricture, a
twist of the pull-through segment, as well as a chronically
dilated rectosigmoid segment. Occasionally, the enema can
show irregularities of the mucosal lining suggesting chronic
inflammation after repeated episodes of HAEC. Such infor-
mation is useful in understanding the etiology of symptoms
as well as in operative planning.
Computerized tomography
Computerized tomography is usually not needed unless
the patient has an unusual presentation.
Ralls et al Re-operative Surgery for Hirschsprung Disease 357

Figure 2 Anatomical considerations in patients with postendorectal pull-through complications. (A) Full-thickness pull-through segment
with ganglia present depicted inside seromuscular cuff with no ganglia. Note the appearance of a mixture of ganglionic, aganglionic, and
scar tissue that would be removed with a myectomy. (B) One suggested approach to a myectomy.

Diagnostic evaluation of the specimen in terms of distal and proximal orientation

Rectal biopsy
A rectal biopsy, either mucosal/submucosal or full thick-
ness, is an important and precise method for evaluation of
chronic problems after a pull through.28 For endorectal or
Duhamel pull through a properly placed suction rectal bi-
opsy in the posterior wall should demonstrate ganglion
cells. Not uncommonly, attaining this information in an
older child is difficult, and often a suction rectal biopsy will
not be accurate enough in terms of the location of the biopsy
as well as obtaining sufficient depth to accurately evaluate
the patient. Should the biopsy show aganglionosis, or be too
superficial, the surgeon should then turn to a full-thickness
biopsy. The authors would also suggest an intermediate
approach. By this, the procedure is performed surgically,
but with the biopsy only going through the submucosa. This
has the advantage of reducing scaring if additional surgery
is needed. As Figure 2A suggests, such a biopsy after a
pull-through procedure is not straightforward. First, identi-
fication and distinction of the dentate line, the anastomosis,
and the distance between them may be difficult. With
growth, the distance from the dentate line to the anastomosis
may become greater than expected; and, at times, this may
lengthen what was a relatively short aganglionic segment at
the time of the initial operation. Second, in patients under-
going a previous ERPT, a full-thickness biopsy will incor-
porate both the ganglionic pull-through wall and the outer
aganglionic lining. For all pull-through procedures, a large
amount of scare tissue may be encountered. Careful labeling
is essential for correct pathologic interpretation. It is rec-
ommended to take serial biopsies at 2, 4, and 5-6 cm from
the dentate line.28,29
Evaluation of rectal biopsies may be enhanced by stain-
ing for acetylcholinesterase staining. However, more easily
performed staining methods, such as calretinin, are becom-
ing increasingly popular, and quite accurate. As with the
initial presentation of HD, rectal biopsy is needed to eval-
uate the neorectum’s enteric nervous system if there is a
retained or acquired aganglionosis. The specimen must be
examined for the presence of hypertrophied nerves at this
level, suggesting transition zone pathology, which may not
function normally. Moreover, in rare instances, intestinal
neuronal dysplasia (IND) may also be present at this
site.30,31 As many key decisions regarding a reoperation are
dependent on the pathologic interpretation, the authors find
reviewing these specimens in person with the pathologist
quite useful.
Anorectal manometry
Use of anorectal manometry is controversial in the post-
operative evaluation of HD patients but can contribute some
valuable information. Manometry can be very useful in the
diagnostic workup of IASA, which may coexist with HD.32
One problem is that most children, even after a successful
pull through, do not regain the rectoanal inhibitory reflex.
Moore et al28 reported that manometric assessment in 16
patients with stooling problems was not significantly differ-
ent from 28 patients who had normal stooling. They noted

Downloaded for FK Universitas Diponegoro (trialckundip01@gmail.com) at ClinicalKey Global Flex Package Trial from ClinicalKey.com by Elsevier on May 26, 2019.
For personal use only. No other uses without permission. Copyright ©2019. Elsevier Inc. All rights reserved.
358
that of these 44 patients, a normal inhibitory relaxation was
observed in only 6 children, of which 3 had IASA and 3 had
normal stooling function, and only 2 of 8 children with a
high resting anorectal tone had stooling problems. Addition-
ally, there is wide variability in the normal values for basal
resting sphincter tone in published series. Other problems
with manometry are that measurements can be subject to
artifacts associated with movements or crying.33 To im-
prove cooperation, it is often helpful to perform manometry
with mild sedation or anesthesia in these young infants.
Children with abnormally low IAS, may have suffered in-
jury to their sphincter complex, and are not candidates for a
redo pull-through procedure; however, such children can
greatly benefit from a bowel management program. Chil-
dren with excessive IAS tone may have recurrent HAEC
and may benefit from a posterior myomyectomy,21 but if
this fails, can dramatically improve with a redo pull-through
procedure. Although not totally accurate and reproducible,
ano-rectal manometry will differentiate incontinence from
encopresis in the patient with constant soiling.
Colonic transit time: Transit time and other more sophis-
ticated colonic motility studies may be helpful in detecting
focal or generalized motility disorders. The reported inci-
dence of IND associated with HD is widely variable,34,35
however, it is almost never indicated to resect bowel dem-
onstrating the findings of IND if the bowel is normal oth-
erwise.
Flexible sigmoidoscopy
Flexible endoscopy may help determine the anatomy of
the pull through beyond that palpable from rectal examina-
tion. It may also demonstrate a torsion or ischemic segment
of the pull-through segment more accurately than contrast
studies. This may be particularly true in those with dysfunc-
tion associated with postoperative anastomotic leak. Endos-
copy can aid in the evaluation of a stricture or scaring
around rectum causing difficulty with evacuation. Finally,
biopsies may help demonstrate the presence of active
HAEC.
Reoperative surgery for HD
There are several operative procedures available for redo
pull-throughs in HD. All of these procedures are much more
difficult to perfom than the original pull-through. This sec-
tion surgical procedures used for reoperations will focus on
the various. It is important to consider that most patients
with protracted defecation disorders will have favorable
outcomes without the need for an additional operation when
following an organized algorithm36 as depicted in Figure 1.
Repeated dilatation
Strictures and IASA are the main indications for repeated
dilatation. Strictures fall into 2 basic types. Simple strictures
Downloaded for FK Universitas Diponegoro (trialckundip01@gmail.com) at ClinicalKey Global Flex Package Trial from ClinicalKey.com by Elsevier on May 26, 2019.
For personal use only. No other uses without permission. Copyright ©2019. Elsevier Inc. All rights reserved.
Seminars in Pediatric Surgery, Vol 21, No 4, November 2012
are moderate narrowing at the anastomotic site. These typ-
ically present within a few weeks after the pull through as
persistent stooling problems or HAEC. In general, these can
be readily addressed with serial outpatient dilations, and
usually do not require subsequent surgery. Serial dilations
can be performed by the surgical staff in the office and
subsequently caregivers who can be adequately trained. The
size of dilations should parallel the goals used after a pull
through for anorectal atresia. Thus, a 10-12-month-old in-
fant should eventually be dilated up to a number 14 Hegar.
Until the stricture has softened and been dilated to this
extent, the family will also need to perform rectal washouts
on the child on at least a twice a day basis.
Less frequently, strictures can be associated with a post-
operative complication, such as an anastomotic leak or those
due to an ischemic segment of bowel. These will tend to be
far more persistent, and will typically not readily respond to
dilations. Minor stricturoplasty may be considered; how-
ever, most children ultimately require surgical intervention.
Of note, when a stricture is located high in the rectal canal,
one can generally assume that these are because of an
ischemic segment. These usually do not respond to simple
or radiologic dilations or stent placement.
Spasm of anal sphincter
Spasm of the anal sphincters after a successful pull through
is not uncommon and could be considered a mild form of
IASA. The diagnosis is established with anorectal manom-
etry showing high resting pressures and a normal rectal
biopsy.37 The typical presentation of such patients includes
recurrent HAEC, and occasionally significant constipation.
These children should be initially approached with bowel
management regimens, including laxatives. Over 1-2 years
after a pull through, most of these symptoms should resolve.
If this regimen is unsuccessful, 2 approaches may be con-
sidered. An initial approach with temporary sphincter-relax-
ing measures, such as botulinum toxin injection,19-22,38-41 or
nitroglycerin paste42-44 may be successful in many children.
Botulinum toxin inhibits the local release of acetylcholine
from presynaptic cholinergic nerves, which in turn causes a
loss of excitatory sympathetic input that leads to increased
IAS tone. It is thought that IASA plays a role in obstructive
symptoms in postoperative HD patients. Botulinum toxin
had been used with the resultant marked improvement of
symptoms.41 Typically, this is done in conjunction with an
aggressive dilation of the anal sphincters under the same
anesthesia. This approach has considerable appeal owing to
the likelihood that the symptoms will improve significantly
over time in most of these children.38 In general, 3-5 mg/kg
of botulinum toxin should be used, injecting this between
the IAS and external anal sphincter (EAS) using a spinal
needle in 5 divided doses around the anal opening, taking
care to avoid the urethra anteriorly. Results have shown that
this technique may produce good results in some patients
with difficulty stooling or persistent enterocolitis and de-
crease hospitalizations.19 The improvement may be long
Ralls et al Re-operative Surgery for Hirschsprung Disease
term, but it usually only lasts three to six months. Should
these approaches fail to show improvement, an anal sphinc-
terotomy, also referred to as ano-rectal a myectomy, should
be strongly considered.45,46
Posterior myotomy/myectomy procedures
Although these procedures were first described for the treat-
ment of short-segment HD, they may have far greater ben-
efit in dealing with children with persistent stooling prob-
lems after a pull-through procedure.
The first of 3 indications for a myectomy is the child with
persistent enterocolitis who fails to respond to medical treat-
ment. Before the performance of the myectomy, a biopsy
should be done to rule out an extensive aganglionic seg-
ment. In general, this should include multiple levels to
determine the extent of any post pull-through angangliono-
sis.
A second indication for myectomy is for severe consti-
pative symptoms in a child who has normal ganglion cells
on repeat biopsy. Typically, these patients will have a di-
lated rectum and sigmoid colon, which will resolve after the
myectomy; however, in some older patients this rectal di-
lation may not resolve. In these later cases, a redo pull-
through with resection of this dilated bowel segment will be
required. Older patients, this rectal dilation may not resolve,
even after a correctly performed myectomy. In these later
cases, these patients will require a redo pull through, with
resection of this dilated bowel segment.
The third group of patients, who may benefit from a
myectomy after a previous pull through, are those children
found to have a retained or acquired aganglionic bowel. If
the length of aganglionic segment is too long, a simple
myectomy will not be sufficient and a redo pull-through will
be required. In general, if the length of aganglionosis is ⬎3
cm, a redo pull through should be performed. This is be-
cause the transition zone immediately proximal to the agan-
glionic segment likely has poor function, and one can rarely
get a myectomy length of ⬎5 or 6 cm. It has been our
experience that a redo pull through may work better in
almost all of these patients.47
Several approaches may be used for a myectomy. How-
ever, after a pull-through procedure, the placement of mul-
tiple sutures, each placed progressively higher into the ano-
rectal canal, and a full-thickness incision (Figure 2B) is the
most direct way to perform this procedure.
Repeat pull-through procedures
The decision to perform a repeat pull-through procedure
represents a challenge for the pediatric surgeons. It carries
risks far greater than the original surgery. It should only be
done by a highly experienced surgeon after following the
aforementioned algorithm of management (Figure 1). How-
ever, one should never hesitate to perform a redo pull
through after careful planning by detailed preoperative bi-
Downloaded for FK Universitas Diponegoro (trialckundip01@gmail.com) at ClinicalKey Global Flex Package Trial from ClinicalKey.com by Elsevier on May 26, 2019.
For personal use only. No other uses without permission. Copyright ©2019. Elsevier Inc. All rights reserved.
359
opsies, contrast enemas, and possibly anal manometry, as
detailed earlier in the text. Each patient should be ap-
proached on an individual basis, and an operative interven-
tion should be planned according to their presentation, un-
derlying pathology and previous surgery.
The common indications for a redo pull through are a
residual aganglionic segment exceeding 3 cm, a transitional
zone pull through with dilated neorectum, persistent long
and/or high stricture after a trial of repeated dilatations, a
twisted pull through, those with obstructive symptoms who
have failed to respond to conservative management or my-
ectomy. One unique situation that may occasionally occur is
when a surgeon is notified of a misread frozen section
within a few days after the performance of the initial pull
through. If possible, it is advisable to proceed with an
immediate repeat pull through. If this approach is taken, few
adhesions will be present, and one can typically revise the
pull through via the same transanal approach. One should
not be dissuaded from proceeding to redo the surgery even
if the patient stooled after the initial pull-through because
obstructive symptoms will eventually recur.
The approach to performing a repeat pull through is
variable depending on what procedure(s) has/have been
previously performed, other interventions performed on the
child and most importantly the contributing factor leading to
the need for a redo pull through. Laparoscopy can often be
useful in exploration and obtaining leveling biopsies.
Should the surgeon have expertise in minimally invasive
surgery then much of the dissection and intra-abdominal
mobilization can be performed with this approach. In fact, a
laparoscopic approach will often allow one to dissect con-
siderably deeper into the pelvis. If laparoscopy is not used,
a lower midline incision is ideal, although if the child has
had a previous left hockey-stick-type incision, this can be
extended, and is usually adequate. Although a transverse
lower abdominal incision is a more cosmetically accepted
incision, it may not allow adequate proximal mobilization
of the colon, which is often needed for a redo pull through.
A thorough lysis of adhesions is frequently required and
should be performed.
Establishing the appropriate level for the pull-through
segment is of the highest priority. A pediatric pathologist
who has expertise with reading frozen sections for the
presence of ganglion cells and hypertrophied nerves must be
available. A circumferential full-thickness ring of bowel at
the level of purposed transection and anastomosis is re-
quired for pathologic confirmation of a normal innervation
of the bowel.48
Once the level is established, a complete mobilization of
the distal colon is performed. Two critical challenges at this
point are to ascertain the vascularity of the bowel and to
ensure adequate length for a tension free anastomosis. There
is the potential for a devascularization of the distal pull-
through segment if a marginal artery is transected. This may
happen when the period between the original pull through
and redo pull through is relatively close, thus preventing an
360
Figure 3 Flow chart of operative planning with respect to the underlying etiology of the post pull-through complication.
adequate secondary neorevascularization. This can also
happen when the child has had a temporizing colostomy
after a failed initial pulthrough. Great care must be taken to
preserve the marginal artery when taking the ostomy down.
The mobilized pull-through segment should reach about 2
cm below the level of the pubic symphysis without tension
to ensure adequate length.
A purely transanal dissection is an option in some chil-
dren. However, in those children with a stricture or previous
leak, this dissection may be difficult, and an open laparot-
omy (as described earlier in the text) is the preferred route.
The decision tree in determining the approach to redo
pull-through procedures is shown in Figure 3. The decision
is one of judgment and may not be readily apparent until the
time of operation. Our general approach has always been to
attempt to perform an ERPT in those patients who have had
either a primary Swenson or ERPT. When this fails, because
of an inability to adequately dissect submucosal plane, a
Duhamel may be used. For those patients who underwent a
primary Duhamel, a Swenson pull through should be con-
sidered. A Swenson may be particularly challenging if one
attempts to evert the rectum, as the last 3-4 cm of rectum are
often densely scared and stuck to the pelvic walls. It typi-
cally is necessary to perform the anastomosis within the
anal canal without eversion.
Understanding the etiology of the failed pull through will
direct the surgeon to the most optimal approach, as indicated in
Figure 3. In this regard, a transanal endorectal approach can be
taken for many children who have a sphincteric spasm or have
a tight endorectal cuff as well as in those patients with a
relatively short segments of retained or recurrent aganglionosis
(⬍5 cm). At times, this will require laparoscopy to assist with
the mobilization of the colon, and to perform confirmatory
Downloaded for FK Universitas Diponegoro (trialckundip01@gmail.com) at ClinicalKey Global Flex Package Trial from ClinicalKey.com by Elsevier on May 26, 2019.
For personal use only. No other uses without permission. Copyright ©2019. Elsevier Inc. All rights reserved.
Seminars in Pediatric Surgery, Vol 21, No 4, November 2012
seromuscular biopsies to ensure a correct level is being pulled
through. In cases where there are long segments of agangli-
onosis, a twist of the pull through or a stricture due to ischemia
or leak, a transabdominal approach is preferred for much of the
dissection. Laparoscopy may be possible in some of these
cases; however, very often, a clear identification of the anat-
omy, extensive mobilization, and dissection through dense scar
requires an open operation. Patients with a twisted segment can
be managed through an endorectal approach or, if the surgeon
prefers, a Duhamel or Swenson procedure.
For patients with a stricture, either a Duhamel or a
Swenson procedure will be required. It is critical that the
surgeon is able to dissect completely below the area of the
stricture. An interesting approach to handle these strictures
was described by Langer.49 He used a stapling device to
come across the stricture while simultaneously creating a
redo-Duhamel anastomosis.
When performing a redo-Duhamel procedure, one must
consider the excess of tissue layers, which must be stapled
together. Unlike a primary Duhamel, the surgical stapler is
joining an additional full-thickness colon wall. If an ERPT
was the original procedure, the anastomosis will involve a
new segment of ganglionic bowel, the original aganglionic
muscularis propria, the primary pull-through segment, as
well as various amounts of scar tissue. A green load extra
long (4.5 mm) stapler should be used if stapling the anas-
tomosis is being contemplated. The surgeon should always
be prepared to perform a hand sewn anastomosis. Alterna-
tively, and in extreme cases, one may revert back to an
original Duhamel procedure, whereby the anastomosis is
created by the placement of 2 extralong Kocher clamps
placed in an inverted “V” configuration, and kept in place
Ralls et al Re-operative Surgery for Hirschsprung Disease 361

until the crushed bowel forms an anastomosis, generally

7-10 days.
An alternative approach is a repeat pull through proce-
dure using an end-to-end circular-stapled anastomosis. The
proximal colon is divided with a gastrointesinal anastomosis
(GIA) stapler at the appropriate level. A dissection of the
rectum is performed analogous to a Swenson procedure,
realizing that the last 2-3 cm of rectum is typically quite
scarred, and any type of mobilization outside of the rectum
is difficult. Dissection is carried down into the pelvis to
within 1-1.5 cm above the dentate line. Great care is taken
to avoid injury of the pelvic structures. Once adequately
mobilized, a Kelly clamp is then introduced from the anus
to the neorectum, and the colonic stump is everted out of the
anus. A tissue approximating or EndoGIA stapler is used to
resect the bulk of the neorectum outside the anus in its
everted state, ideally within 1-2 cm above the dentate line
(Figure 4). The neorectum is then reinserted into the pelvis.
The end result is approximately a 2.5-cm length of residual
rectum. The staple line of the proximal ganglionated colon
is then removed, and a purse-string suture is used to secure
the anvil of the EEA stapler in place. A 21 mm (or larger)
EEA stapler is then inserted into the rectal remnant. An
EEA stapled anastomosis is then performed (Figure 4). The
anastomosis is assessed by verifying the presence of two
complete donuts that contained both mucosa and wall on
both sides. Finally, an air bubble test is performed after
filling the pelvis with saline.
Regardless of the technique used, the surgeon must take
great care to assess the completed procedure. Technical
problems with the pull through, including a twisting, com-
promise of the blood supply, or a kink of the bowel in the
pelvis, must be avoided. Long-segment HD can be partic- Figure 4 Surgical approach using an end-to-end stapled anas-
ularly troublesome, as the use of more proximal segments of tomosis for a difficult redo pull through. (A) The previous pull
bowel can be difficult. In fact, if the child is left with only through is everted out via the anal canal. The dotted line indicates
the cecum, a safer approach would be to do a completion the target line of transection. (B) Transection of the pull through
cecectomy and an ileal pull through. Cecal pull throughs with either a tissue approximating-stapling device (as shown) or a
have trouble owing to the risk of twisting the bowel as well gastrointestinal anastomosis (GIA). (C) Circular end-to-end anas-
as a mismatch of size when the anastomosis is performed. tomosis (EEA) stapling device used to create anastomosis between
neorectum and the ganglionated colon.
A decision to perform a protective proximal ostomy
should be made at the end of the redo pull through. If the
anastomosis is performed with a high degree of tension, or
decide that the child may not benefit from a repeat pull
any question remains about the anastomosis, a proximal
through. In these cases, one could consider a permanent
colostomy or ileostomy should be placed. Great care should
stoma. Assistance from a stomal therapist may help ensure
be taken when creating this diversion, as the blood supply in
that this stoma is placed in the most appropriate location for
a redo pull through could potentially be compromised. The
long-term use.
marginal artery could be lost either during the creation of
the stoma or when the ostomy is taken down. In such cases,
it is far safer to perform a temporary loop ileostomy. Stooling outcomes

As most patients do well after an initial pull through, most

Special considerations and outcomes
Permanent stoma
There will be a small group of patients who may fail a
repeat pull through, or the child’s parents and physician may
centers have only a handful of reoperative patients. Thus,
there is a paucity of data on the stooling outcomes of those
undergoing reoperative surgery. Studies vary greatly in the
reporting of continence rates after a redo pull through.
Continence without soiling ranges from as low as 31% in 1
series50 to 92% in another.47 A recent meta-analysis of 29

Downloaded for FK Universitas Diponegoro (trialckundip01@gmail.com) at ClinicalKey Global Flex Package Trial from ClinicalKey.com by Elsevier on May 26, 2019.
For personal use only. No other uses without permission. Copyright ©2019. Elsevier Inc. All rights reserved.
362
articles on the subject resulted in the accumulation of data
on 555 reoperative pull-through patients. In a previous re-
view from Mott Children’s Hospital, 32 children underwent
a redo-pullthrough. Excluding those who were neurologi-
cally impaired, 94% had reasonable stooling outcomes.
Since this report, an additional 20 children have undergone
a redo-pullthrough, with similar outcomes. Stooling out-
comes were reported in 134 patients, of whom 74% had
normal bowel habits after a redo pull through.51
Conclusions
Reoperations for Hirschsprung disease require a great deal
of planning and a very thoughtful workup. In cases where a
reoperation is needed, however, reasonably good results
may be found in most cases. In general, the type of proce-
dure performed for the redo pull through is based on the
original procedure and the reason the pull through failed.
Therefore, the pediatric surgeon performing such proce-
dures must be experienced with all the different approaches
because every case is unique and often requires many of the
available techniques.
References
1. Raveenthiran V. Knowledge of ancient Hindu surgeons on Hirsch-
sprung disease: Evidence from Sushruta Samhita of circa 1200-600
BC. J Pediatr Surg 2011;46:2204-8.
2. Swenson O, Bill AH Jr. Resection of rectum and rectosigmoid with
preservation of the sphincter for benign spastic lesions producing
megacolon; an experimental study. Surgery 1948;24:212-20.
3. Minford JL, Ram A, Turnock RR, et al. Comparison of functional
outcomes of Duhamel and transanal endorectal coloanal anastomosis
for Hirschsprung’s disease. J Pediatr Surg 2004;39:161-5.
4. Saleh W, Rasheed K, Mohaidly MA, et al. Management of Hirsch-
sprung’s disease: A comparison of Soave’s and Duhamel’s pull-
through methods. Pediatr Surg Int 2004;20:590-3.
5. Craigie RJ, Conway SJ, Cooper L, et al. Primary pull-through for
Hirschsprung’s disease: Comparison of open and laparoscopic-assisted
procedures. J Laparoendosc Adv Surg Tech A 2007;17:809-12.
6. Murphy F, Puri P. New insights into the pathogenesis of Hirsch-
sprung’s associated enterocolitis. Pediatr Surg Int 2005;21:773-9.
7. Ruttenstock E, Puri P. Systematic review and meta-analysis of entero-
colitis after one-stage transanal pull-through procedure for Hirsch-
sprung’s disease. Pediatr Surg Int 2010;26:1101-5.
8. Zhang SC, Bai YZ, Wang W, et al. Stooling patterns and colonic
motility after transanal one-stage pull-through operation for Hirsch-
sprung’s disease in children. J Pediatr Surg 2005;40:1766-72.
9. Teitelbaum DH, Drongowski RA, Chamberlain JN, et al. Long-term
stooling patterns in infants undergoing primary endorectal pull-
through for Hirschsprung’s disease. J Pediatr Surg 1997;32:1049-52;
discussion 52-3.
10. Di Lorenzo C, Solzi GF, Flores AF, et al. Colonic motility after
surgery for Hirschsprung’s disease. Am J Gastroenterol 2000;95:
1759-64.
11. Athanasakos E, Starling J, Ross F, et al. An example of psychological
adjustment in chronic illness: Hirschsprung’s disease. Pediatr Surg Int
2006;22:319-25.
Downloaded for FK Universitas Diponegoro (trialckundip01@gmail.com) at ClinicalKey Global Flex Package Trial from ClinicalKey.com by Elsevier on May 26, 2019.
For personal use only. No other uses without permission. Copyright ©2019. Elsevier Inc. All rights reserved.
Seminars in Pediatric Surgery, Vol 21, No 4, November 2012
12. Gad El-Hak NA, El-Hemaly MM, Negm EH, et al. Functional out-
come after Swenson’s operation for Hirshsprung’s disease. Saudi. J
Gastroenterol 2010;16:30-4.
13. Zhang SC, Bai YZ, Wang W, et al. Long-term outcome, colonic
motility, and sphincter performance after Swenson’s procedure for
Hirschsprung’s disease: A single-center 2-decade experience with 346
cases. Am J Surg 2007;194:40-7.
14. Yanchar NL, Soucy P. Long-term outcome after Hirschsprung’s dis-
ease: Patients’ perspectives. J Pediatr Surg 1999;34:1152-60.
15. Teitelbaum DH, Cilley RE, Sherman NJ, et al. A decade of expe-
rience with the primary pull-through for Hirschsprung disease in
the newborn period: A multicenter analysis of outcomes. Ann Surg
2000;232:372-80.
16. van Leeuwen K, Teitelbaum DH, Elhalaby EA, et al. Long-term
follow-up of redo pull-through procedures for Hirschsprung’s disease:
Efficacy of the endorectal pull-through. J Pediatr Surg 2000;35:829-
33; discussion 33-4.
17. Wei J, Zhang W, Feng JX, et al. Transumbilical laparoscopic pull-
through for children with hypoganglionosis [in Chinese]. Zhonghua
Wei Chang Wai Ke Zhi 2011;14:762-3.
18. Dingemann J, Puri P. Isolated hypoganglionosis: Systematic review of
a rare intestinal innervation defect. Pediatr Surg Int 2010;26:1111-5.
19. Patrus B, Nasr A, Langer JC, et al. Intrasphincteric botulinum toxin
decreases the rate of hospitalization for postoperative obstructive
symptoms in children with Hirschsprung disease. J Pediatr Surg 2011;
46:184-7.
20. Koivusalo AI, Pakarinen MP, Rintala RJ. Botox injection treatment for
anal outlet obstruction in patients with internal anal sphincter achalasia
and Hirschsprung’s disease. Pediatr Surg Int 2009;25:873-6.
21. Wildhaber BE, Pakarinen M, Rintala RJ, et al. Posterior myotomy/
myectomy for persistent stooling problems in Hirschsprung’s disease.
J Pediatr Surg 2004;39:920-6.
22. Jiang da P, Li ZZ, Zhang YB, et al. Treatment of anal achalasia after
transanal endorectal pull-through for Hirschsprung’s disease with top-
ical botulinum toxin. J Trop Pediatr 2008;54:211.
23. Dasgupta R, Langer JC. Evaluation and management of persistent
problems after surgery for Hirschsprung disease in a child. J Pediatr
Gastroenterol Nutr 2008;46:13-9.
24. Reding R, de Ville de Goyet J, Gosseye S, et al. Hirschsprung’s
disease: A 20-year experience. J Pediatr Surg 1997;32:1221-5.
25. Vieten D, Spicer R. Enterocolitis complicating Hirschsprung’s disease.
Semin Pediatr Surg 2004;13:263-72.
26. Teitelbaum DH, Caniano DA, Qualman SJ. The pathophysiology of
Hirschsprung’s-associated enterocolitis: Importance of histologic cor-
relates. J Pediatr Surg 1989;24:1271-7.
27. Blane CE, Elhalaby E, Coran AG. Enterocolitis following endorectal
pull-through procedure in children with Hirschsprung’s disease. Pedi-
atr Radiol 1994;24:164-6.
28. Moore SW, Millar AJ, Cywes S. Long-term clinical, manometric, and
histological evaluation of obstructive symptoms in the postoperative
Hirschsprung’s patient. J Pediatr Surg 1994;29:106-11.
29. Meier-Ruge WA, Bruder E, Holschneider AM, et al. Diagnosis and
therapy of ultrashort Hirschsprung’s disease. Eur J Pediatr Surg 2004;
14:392-7.
30. Kobayashi H, Yamataka A, Lane GJ, et al. Inflammatory changes
secondary to postoperative complications of Hirschsprung’s disease as
a cause of histopathologic changes typical of intestinal neuronal dys-
plasia. J Pediatr Surg 2004;39:152-6.
31. Kobayashi H, Hirakawa H, Surana R, et al. Intestinal neuronal dys-
plasia is a possible cause of persistent bowel symptoms after pull-
through operation for Hirschsprung’s disease. J Pediatr Surg 1995;30:
253-7.
32. Martucciello G, Pini Prato A, Puri P, et al. Controversies concern-
ing diagnostic guidelines for anomalies of the enteric nervous
system: A report from the fourth International symposium on
Hirschsprung’s disease and related neurocristopathies. J Pediatr
Surg 2005;40:1527-31.
Ralls et al Re-operative Surgery for Hirschsprung Disease
33. Pfefferkorn MD, Croffie JM, Corkins MR, et al. Impact of sedation
and anesthesia on the rectoanal inhibitory reflex in children. J Pediatr
Gastroenterol Nutr 2004;38:324-7.
34. Meier-Ruge WA, Ammann K, Bruder E, et al. Updated results on intes-
tinal neuronal dysplasia (IND B). Eur J Pediatr Surg 2004;14:384-91.
35. Smith B. The neuropathology of pseudo-obstruction of the intestine.
Scand J Gastroenterol Suppl 1982;71:103-9.
36. Chumpitazi BP, Nurko S. Defecation disorders in children after surgery
for Hirschsprung disease. J Pediatr Gastroenterol Nutr 2011;53:75-9.
37. Dasgupta R, Langer JC. Hirschsprung disease. Curr Probl Surg 2004;
41:942-88.
38. Ciamarra P, Nurko S, Barksdale E, et al. Internal anal sphincter
achalasia in children: Clinical characteristics and treatment with Clos-
tridium botulinum toxin. J Pediatr Gastroenterol Nutr 2003;37:315-9.
39. Messineo A, Codrich D, Monai M, et al. The treatment of internal anal
sphincter achalasia with botulinum toxin. Pediatr Surg Int 2001;17:
521-3.
40. Chumpitazi BP, Fishman SJ, Nurko S. Long-term clinical outcome
after botulinum toxin injection in children with nonrelaxing internal
anal sphincter. Am J Gastroenterol 2009;104:976-83.
41. Jiang da P, Xu CQ, Wu B, et al. Effects of botulinum toxin injection
on anal achalasia after pull-through operations for Hirschsprung’s
disease: A 1-year follow-up study. Int J Colorectal Dis 2009;24:597-8.
42. Millar AJ, Steinberg RM, Raad J, et al. Anal achalasia after pull-
through operations for Hirschsprung’s disease—Preliminary expe-
rience with topical nitric oxide. Eur J Pediatr Surg 2002;12:207-11.
Downloaded for FK Universitas Diponegoro (trialckundip01@gmail.com) at ClinicalKey Global Flex Package Trial from ClinicalKey.com by Elsevier on May 26, 2019.
For personal use only. No other uses without permission. Copyright ©2019. Elsevier Inc. All rights reserved.
363
43. Lindsey I, Jones OM, Cunningham C, et al. Botulinum toxin as
second-line therapy for chronic anal fissure failing 0.2 percent Glyc-
eryl trinitrate. Dis Colon Rectum 2003;46:361-6.
44. Demirbag S, Tander B, Atabek C, et al. Long-term results of topical
glyceryl trinitrate ointment in children with anal fissure. Ann Trop
Paediatr 2005;25:135-7.
45. De Caluwé D, Yoneda A, Akl U, et al. Internal anal sphincter acha-
lasia: Outcome after internal sphincter myectomy. J Pediatr Surg
2001;36:736-8.
46. Heikkinen M, Lindahl H, Rintala RJ. Long-term outcome after internal
sphincter myectomy for internal sphincter achalasia. Pediatr Surg Int
2005;21:84-7.
47. Teitelbaum DH, Coran AG. Reoperative surgery for Hirschsprung’s
disease. Semin Pediatr Surg 2003;12:124-31.
48. Das K, Kini U, Babu MK, et al. The distal level of normally innervated
bowel in long segment colonic Hirschsprung’s disease. Pediatr Surg
Int 2010;26:593-9.
49. Langer JC. Repeat pull-through surgery for complicated Hirsch-
sprung’s disease: Indications, techniques, and results. J Pediatr Surg
1999;34:1136-41.
50. Peña A, Elicevik M, Levitt MA. Reoperations in Hirschsprung disease.
J Pediatr Surg 2007;42:1008-13; discussion 13-4.
51. Friedmacher F, Puri P. Residual aganglionosis after pull-through op-
eration for Hirschsprung’s disease: A systematic review and meta-
analysis. Pediatr Surg Int 2011;27:1053-7.