Pe d i a t r i c I m a g i n g • O r i g i n a l R e s e a r c h

Ntoulia et al.
Radiologic and Other Findings of Failed Intussusception Reduc-
tion in Children

Pediatric Imaging
Original Research

Failed Intussusception
Downloaded from www.ajronline.org by 180.251.162.88 on 02/13/18 from IP address 180.251.162.88. Copyright ARRS. For personal use only; all rights reserved

Reduction in Children: Correlation

Between Radiologic, Surgical, and
Pathologic Findings
Aikaterini Ntoulia1 OBJECTIVE. The objective of this study was to identify causes of irreducible intussusception
Sasha J. Tharakan after contrast enema and to correlate imaging findings with surgical and histopathologic findings.
Janet R. Reid MATERIALS AND METHODS. Between 2005 and 2013, a total of 543 children un-
Soroosh Mahboubi derwent reduction of intussusception with the use of an enema technique (hereafter referred to
as “enema reduction”). The medical records of 72 children (56 boys [mean age, 24.8 months;
Ntoulia A, Tharakan SJ, Reid JR, Mahboubi S range, 3.8 months to 10.9 years] and 16 girls [mean age, 14.2 months; range, 1.5 months to
6.9 years) who underwent unsuccessful reduction and were treated surgically were retrospec-
tively analyzed. The data collected included information on the cause of intussusception, the
risk factors noted on ultrasound, operative management, outcome, and the length of the hospi-
tal stay. The imaging findings for these patients were compared with findings for statistically
similar age-matched control subjects.
RESULTS. Ultrasound detected 56 of 57 cases of intussusception, but it failed to detect the
lead point in three cases and failed to detect ischemic necrosis in seven cases. Positive predic-
tors of failed enema reduction were the presence of a distal mass and observation of the dissect-
ing sign. Of the 72 patients who underwent surgical treatment of intussusception, 26 (36.1%)
underwent laparoscopy, 38 (52.8%) underwent laparotomy, and eight (11.1%) underwent conver-
sion from laparoscopy to laparotomy. Surgical reduction was performed in 61.1% of cases, small
bowel resection in 19.4%, ileocecectomy in 12.5%, and self-reduction in 69%. Pathologic lead
points (noted in 25% of cases) included lymphoid hyperplasia (n = 7), Meckel diverticulum (n =
3), Burkitt lymphoma (n = 3), enteric duplication cyst (n = 2), juvenile polyp (n = 2), and adenovi-
rus appendicitis (n = 1). The length of the hospital stay was significantly longer after laparotomy.
CONCLUSION. The distalmost location of the intussusception mass and presence of
the dissecting sign on images obtained during contrast enema have a higher positive predic-
tive value for failed reduction. Screening ultrasound decreases the number of unnecessary
contrast enemas performed; however, classic pathologic lead points, such as Burkitt lympho-
Keywords: air enema, failed reduction, fluoroscopic ma and Meckel diverticulum, may be difficult to diagnose with the use of ultrasound. Lapa-
reduction, fluoroscopy, intussusception, laparoscopy, rotomy and laparoscopy are equally safe and efficacious in reducing intussusception, with the
laparotomy, ultrasound
length of the hospital stay after laparoscopy significantly shorter than that noted after lapa-
DOI:10.2214/AJR.15.15659 rotomy. Most failed enema reductions are idiopathic, and pathologic lead points are noted in
25% of cases.
Received September 16, 2015; accepted after revision
February 26, 2016.
ntussusception is one of the most The classic clinical symptoms of intermit-
J. R. Reid receives royalties from Oxford University Press.

Based on a presentation at the Society for Pediatric

Radiology 2015 annual meeting, Seattle, WA.
1
All authors: The Children’s Hospital of Philadelphia,
324 S 34th St, Philadelphia, PA 19104. Address
correspondence to J. R. Reid (reidj@email.chop.edu).
AJR 2016; 207:424–433
0361–803X/16/2072–424
© American Roentgen Ray Society
I frequent causes of intestinal ob-
struction in early childhood and
is an abdominal emergency. It
occurs when a bowel segment invaginates
into its adjacent distal segment [1, 2]. This
telescoping of adjoining intestinal segments
is propelled forward by bowel peristalsis
leading to bowel obstruction. If intussuscep-
tion is not promptly treated, increased intra-
luminal pressure results in vascular compro-
mise, bowel wall ischemia, and perforation
with high associated morbidity rates [3–7].
tent abdominal pain are a triad of palpable ab-
dominal mass, emesis, and passage of stools
mixed with blood and mucus (known as “red
currant jelly stools”) that strongly raises the
clinical suspicion of intussusception. Howev-
er, many children with intussusception may
present with a combination of the aforemen-
tioned symptoms, which can be nonspecific
and can also overlap with other conditions.
Therefore, imaging plays an important role
in the diagnosis of intussusception. Screening
ultrasound has been established as the imag-

424 AJR:207, August 2016

 Radiologic and Other Findings of Failed Intussusception Reduction in Children

veloped and developing countries [4, 11–15].

4853 Children
underwent diagnostic investigation
In cases of failed enema reduction, surgical
for clinical suspicion of intussusception treatment is required.
A few large sample studies have explored
the relationship between the imaging findings
543 Children for children with irreducible intussusception
Downloaded from www.ajronline.org by 180.251.162.88 on 02/13/18 from IP address 180.251.162.88. Copyright ARRS. For personal use only; all rights reserved

underwent contrast enema after contrast enema reduction and the out-
for intussusception reduction comes noted after surgical intervention. The
present study includes what is, to our knowl-
edge, the largest pediatric population studied
116 Children had no evidence 427 Children had evidence to date. The goals of our study were to iden-
of intussusception at the time of intussusception at the time
contrast enema was performed contrast enema was performed tify the cause of failed contrast enema reduc-
tion of intussusception in children; determine
whether screening ultrasound can detect risk
In 353 children, contrast In 74 children, contrast factors predisposing to failure of contrast en-
enema successfully enema failed to ema reduction and associated complications;
reduced intussusception reduce intussusception and evaluate the outcomes of the operative
management of intussusception in terms of
surgical approach, surgical and histopatholog-
72 Children had surgical 2 Children had interval ic findings, and length of stay in the hospital.
management of self-reduction of
intussusception intussusception
Materials and Methods
The present retrospective study was approved
Fig. 1—Flowchart shows study population. by the institutional review board of The Children’s
Hospital of Philadelphia and was HIPAA compli-
ing modality of choice to confirm or exclude technique (hereafter referred to as “enema ant. The medical records of all children who were
the presence of intussusception in children, reduction”) is performed in the radiology de- referred to the radiology department at our institu-
with high reported sensitivity (98–100%) and partment with the use of water-soluble or air tion for clinical suspicion of intussusception from
specificity (88–100%) and negative predictive contrast media under fluoroscopic or ultra- 2005 to 2013 were retrospectively reviewed. Re-
value of 100% [5, 8]. sound guidance, depending on the training trieval of the imaging records of the children was
In the absence of contraindications, in- and experience of the radiologist [9, 10]. Fail- performed using a search engine (Softek Illumi-
cluding either clinical or imaging findings ure rates for intussusception reduction after nate, Softek Solutions). The keywords that were
of colonic perforation, peritonitis, or hypo- contrast enema reduction range from 10% to used to identify eligible subjects included “intus-
volemic shock, nonoperative reduction of 60%, depending on the tertiary care center susception,” “ultrasound,” “radiography,” “fluo-
intussusception with the use of an enema facilities, and these rates are different in de- roscopy,” and “contrast enema.”

A B
Fig. 2—6-year-old girl with intussusception secondary to enteric duplication.
A, Gray-scale ultrasound image shows ileocolic intussusception (arrowheads) in right upper quadrant. Cystic structure involved in
intussusception appears to be lead point. Sonographic visualization of double wall sign (arrows) consisting of inner hyperechoic rim
and outer hypoechoic layer is highly suggestive of enteric duplication cyst. Two layers correspond to mucosa-submucosa interface
and muscularis propria. Hyperechoic debris (asterisk) is seen in cyst lumen. L = liver, RK = right kidney.
B, Gray-scale ultrasound image shows trapped fluid (arrow) in intussusception mass. Fluid insinuates between layers of
intussusceptum and intussuscipiens.

AJR:207, August 2016 425


Downloaded from www.ajronline.org by 180.251.162.88 on 02/13/18 from IP address 180.251.162.88. Copyright ARRS. For personal use only; all rights reserved
Fig. 3—3-month-old male infant with bloody stools.
Image obtained during administration of air contrast
enema shows intussusception (arrow) that was
irreducible at level of sigmoid colon.
During the 8 years reviewed, 4850 children who
were clinically suspected of having intussusception
were referred to the department of radiology to un-
dergo imaging evaluation, which included a vari-
able combination of abdominal radiography, ul-
trasound, and contrast enema for the diagnosis of
intussusception, intussusception reduction, or both.
Standard abdominal radiography was per-
formed with the patient in the supine and decubi-
tus positions, and ultrasound was performed using
HDI 5000 and iU22 systems (Philips Healthcare)
with C5-8–MHz curved array or L12-5–MHz lin-
ear transducers. Air contrast enema, water-soluble
contrast enema, or a combination of these modal-
ities was subsequently performed, depending on
the training and experience of the radiologist.
Pneumatic enema reduction was performed
using the Shiel technique. A catheter was placed
into the rectum and was secured in place with
tape. The air insufflator device, which included a
pressure monitor and a pop-off valve, was then at-
tached to the catheter. The valve was set at 120
mm Hg to ensure that the insufflated air pressure
never exceeded that limit. Air was slowly insuf-
flated into the colon under fluoroscopic visualiza-
tion, with the use of a handheld pressure gauge.
During the procedure, positioning of the patient
was alternated between the prone and supine posi-
tions to promote air passage.
Hydrostatic enema reduction was performed
using the isotonic water-soluble contrast medium
iothalamate meglumine (17.2%, Cysto-Conray II,
Mallinckrodt), which was instilled into the rectum
via gravity, with the bag suspended at a height ap-
proximately 60 cm above the level of the abdomen
in the patient.
A total of 543 children were identified as having
undergone contrast enema. For these children, in-
formation on demographic characteristics, clinical
426 AJR:207, August 2016
Ntoulia et al.
A B
Fig. 4—Two infants with colic and abdominal pain.
A, 3-year-old boy. Fluoroscopic image obtained during administration of water-soluble contrast enema shows
excessive length of bowel loops with twisted configuration (arrows) consistent with colonic redundancy.
B, 6-month-old male infant. Fluoroscopic image obtained during administration of air contrast enema shows
excessive length of bowel loops with twisted configuration (arrows) consistent with colonic redundancy.
presentation, the enema reduction technique used, findings for which histopathologic confirmation was
the contrast medium used for intussusception re- available, and the length of the hospital stay.
duction, and the outcome of contrast enema was The ultrasound findings for children for whom
recorded. Intussusception reduction failed in 74 intussusception reduction failed were compared
children. The available ultrasound images of these with those for a control group of 57 age-matched
children were retrospectively evaluated to char- children who underwent successful contrast enema
acterize the type of intussusception mass, identi- reduction. Contrast enema findings were also com-
fy any underlying pathologic findings in the bowel pared for children for whom intussusception reduc-
that could act as a pathologic lead point for intus- tion failed and a group of 74 age-matched children.
susception, and detect imaging features potentially Review of all available imaging examinations of el-
associated with intussusception irreducibility, in- igible study subjects was performed by two radiolo-
cluding the presence of either trapped fluid within gists, in accordance with consensus criteria. Figure
the intussusception layers or peritoneal free fluid 1 presents detailed information on our population
as well as the absence of blood flow on color Dop- study groups.
pler imaging, power Doppler imaging, or both.
Moreover, available images obtained during Results
administration of contrast enema were retrospec- Contrast enema was performed for 543
tively reviewed to identify the colon sites involved of the 4850 children (11.2%) identified, 365
in the intussusception at the onset of enema reduc- (67.2%) of whom were boys (mean age, 25.2
tion and the colon sites where the intussusception months; range, 2 months to 14.9 years) and
persisted after unsuccessful attempts at enema re- 178 (32.8%) of whom were girls (mean age,
duction, as well as to reveal any imaging signs 20 months; range, 1 month to 8.3 years). The
associated with potential irreducibility of intus- clinical symptoms at the time of presentation
susception, including colon redundancy and the were colicky abdominal pain with or with-
dissecting sign. Colon redundancy was considered out palpable abdominal mass, bloody stools,
present when rectosigmoid length was excessive and emesis. Symptoms of feeding intoler-
and had elongated and twisted loops, whereas the ance, irritability, or lethargy were less fre-
dissecting sign was considered present when con- quently reported. The mean duration of clin-
trast medium was trapped between the intussus- ical symptoms was 3.2 days (range, 4 hours
ceptum and intussuscipiens layers. to 5.0 days).
The medical records of children for whom intus- Table 1 presents the outcomes in terms of
susception reduction failed were also reviewed to the success or failure of the intussusception
determine the duration of symptoms before hospi- reduction after contrast enema was adminis-
tal admission and the subsequent management strat- tered with the use of different contrast media
egy, including the surgical approach used, surgical and imaging guidance techniques.
 TABLE 1: Outcome of Contrast Enema Performed for Intussusception ing the presence of colonic intussusception
Radiologic and Other Findings of Failed Intussusception Reduction in Children
­Reduction With the Use of Different Contrast Media Under the that could not be reduced.
Guidance of Different Imaging Modalities In 46 of the 57 children, ultrasound re-
Total No. of
vealed multiple enlarged lymph nodes asso-
Type of Contrast Enema, Outcome Children No. of Boys No. of Girls ciated with the intussusception mass, and in
two children, a duplication cyst was detected
Fluoroscopically guided air contrast enema 270 176 94
(Fig. 2). No associated bowel pathologic find-
No intussusception at the time of enemaa 43 23 20 ings were seen in the remaining nine chil-
Successful reduction 199 132 67 dren. Trapped fluid insinuated between the
Downloaded from www.ajronline.org by 180.251.162.88 on 02/13/18 from IP address 180.251.162.88. Copyright ARRS. For personal use only; all rights reserved

Failed reduction 28 21 7 intussusception layers was detected in only

nine of the 57 children, and peritoneal free
Fluoroscopically guided water soluble contrast enema 200 129 71
fluid was detected in five children. Blood
No intussusception at the time of enemaa 72 46 26 flow was present in the intussusception mass
Successful reduction 106 67 39 in all 53 children who underwent color Dop-
Failed reduction 22 16 6 pler examination, power Doppler examina-
tion, or both.
Fluoroscopically guided enema with air and 49 37 12
water-soluble contrast media used in the same
Air was the contrast medium used for 28
session during successive attempts children, water-soluble contrast medium was
used for 26 children, and 20 children received
No intussusception at the time of enemaa 1 1 0
a combination of air and water-soluble con-
Successful reduction 28 20 8 trast media during successive enema attempts.
Failed reduction 20 16 4 Of the 20 children who received both contrast
Ultrasound-guided water-soluble contrast enema 24 19 1 media, air was used first and was followed by
water-soluble contrast medium for 16 chil-
Successful reduction 20 19 1
dren, whereas water-soluble contrast medium
Failed reduction 4 4 0 was used first and was followed by air for four
aEnema was performed on the basis of high clinical suspicion and abdominal radiography findings. No
children. The mean number of failed enema
ultrasound examination was performed before contrast enema. In these cases, there was no evidence of attempts was 2.9 (range, 1–6 failed attempts).
intussusception at the time of enema.
In most children in whom enema reduction
failed, the intussusception mass was encoun-
Air contrast enema successfully reduced in- and 17 girls [mean age, 14.1 months; range, tered at the level of the hepatic flexure (n =
tussusception in 87.6% of children (199/227), 1.5 months to 6.9 years]). 27) or the transverse colon (n = 22). Other less
whereas water-soluble contrast enema was Of the 74 children for whom intussuscep- common locations included the sigmoid colon
successful in 82.8% of children (106/128). No tion reduction failed, 57 underwent high-res- (n  = 10), the splenic flexure (n = 8), the ce-
statistically significant difference in the suc- olution gray-scale ultrasound before contrast cum (n = 5), and the ascending colon (n = 2).
cess rate associated with intussusception re- enema was attempted, and 53 of these 57 In most cases, intussusception reduction failed
duction was observed with the use of air con- children also underwent color Doppler ex- at the level of the ileocecal valve (n = 64). In a
trast medium (success rate, 89.6%) versus amination, power Doppler examination, or few cases, the intussusception could not move
water-soluble contrast medium (p > 0.05). both. Screening ultrasound confirmed the di- beyond the hepatic flexure (n = 6), the trans-
agnosis of colonic intussusception in 56 of verse colon (n  = 2), the splenic flexure (n =
Failure of Intussusception Reduction After the 57 children. For one child, ultrasound 1), or the sigmoid colon (n = 1) (Fig. 3). Co-
Contrast Enema findings favored the diagnosis of ileoileal in- lon redundancy was observed in 10 children
Failure of intussusception reduction oc- tussusception. However, because of the per- (13.5%) and the dissecting sign was observed
curred in 74 patients (57 boys [mean age, sistence of clinical symptoms, air contrast in 11 children (14.9%) for whom enema re-
25.0 months; range, 3.8 months to 10.9 years] enema was subsequently performed, reveal- duction failed (Figs. 4 and 5).

Fig. 5—Two infants with irreducible

intussusception.
A, 3-month-old male infant. Image
obtained during administration
of water-soluble contrast enema
shows dissecting sign (arrows)
where contrast material tracks
between intussusceptum and
intussuscipiens layers, resulting in
loss of pressure at dissection.
B, 19-month-old female
infant. Image obtained during
administration of air contrast enema
shows dissecting sign (arrows)
where contrast material tracks
between intussusceptum and
intussuscipiens layers, resulting in
loss of pressure at dissection.
A B

AJR:207, August 2016 427

 Ntoulia et al.

reduction of intussusception. Of these chil-

72 Children underwent dren, 26 (36.1%) underwent laparoscopy and
surgical management
of intussusception
38 (52.8%) underwent laparotomy, whereas
eight children (11.1%) required conversion
from laparoscopy to laparotomy.
8 Children underwent
26 Children underwent 38 Children underwent
Downloaded from www.ajronline.org by 180.251.162.88 on 02/13/18 from IP address 180.251.162.88. Copyright ARRS. For personal use only; all rights reserved

conversion from Surgical Outcomes

laparoscopy
n=3
Self-reduced
intussusception
n = 21
Surgical reduction
(laparoscopic
instrumentation)
n=2
Small bowel
resection
Fig. 6—Flowchart shows surgical approach used and procedures performed for children for whom
intussusception reduction failed.
Management of Children for Whom Contrast
Enema Reduction Failed
Of the 74 children for whom intussus-
ception reduction failed, 72 (56 boys [mean
age, 24.8 months; range, 3.8 months to 10.9
years] and 16 girls [mean age, 14.2 months;
range, 1.5 months to 6.9 years]) subsequent-
ly underwent surgical treatment of intussus-
ception. The remaining two children did not
laparotomy
laparoscopy to laparotomy
Five children (6.9%) who underwent sur-
gery had no intraoperative evidence of in-
n=2 tussusception and required no surgical ma-
Self-reduced neuver. However, the presence of indirect
intussusception
findings, including edematous distal termi-
nal ileum with areas of bowel wall indenta-
tion, was suggestive of spontaneous reduc-
n = 23 tion. Therefore, it appeared that either enema
Surgical reduction
(manual) reduction had been successful or the intus-
susception had spontaneously reduced sub-
sequent to the enema attempt. Of these five
n=6 n=6
children with self-reduction of the intussus-
Small bowel Small bowel ception, three had undergone laparoscopy
resection resection and two had undergone laparotomy.
In 44 children (61.1%), surgical reduction
of intussusception was performed either us-
ing laparoscopic instrumentation (n = 21) or
n=7 n=2
Ileocectomy Ileocectomy manually during laparotomy (n = 23). Ileo-
cecal-sparing small bowel resection was re-
quired for 14 of the 72 children who under-
went surgical reduction of intussusception
(19.4%), and ileocecectomy was performed
for nine children (12.5%).
have to undergo surgical treatment because Figure 6 presents the surgical procedures
clinical stability and repeated ultrasound ex- performed for treatment of intussusception.
amination confirmed that interval self-re- Ileocolic intussusception was the most
duction of the intussusception occurred after common type of intussusception, account-
contrast enema was performed. ing for 85.1% of cases of failed reduction,
followed by ileoileocolic and ileocoloco-
Surgical Approach lic intussusception, which accounted for
Seventy-two children were subsequently 11.9% and 3% of cases of failed reduction,
transferred to the operating room for surgical respectively.

A B C
Fig. 7—5-month-old male infant with abdominal distention, emesis, and lethargy.
A, Intraoperative photograph shows irreducible ileocolic intussusception with ischemic changes in bowel wall. Appendix (arrow) with vascular congestion is also shown.
B, Color photograph shows gross specimen obtained from same patient after resection. Forceps is grasping appendix (arrow).
C, Color Doppler sonogram shows vascular flow within intussusception mass. Small lymph nodes (arrows) are also noted.

428 AJR:207, August 2016

 Radiologic and Other Findings of Failed Intussusception Reduction in Children

TABLE 2: Histopathologic Confirmation for Cases of Failed Intussusception Reduction

Surgical Approach
Therapeutic Conversion From Laparoscopy
Histopathologic Confirmation Laparoscopy Laparotomy to Laparotomy All
Surgical reduction of intussusception without specimen obtained 10 5 0 15
Downloaded from www.ajronline.org by 180.251.162.88 on 02/13/18 from IP address 180.251.162.88. Copyright ARRS. For personal use only; all rights reserved

for pathologic analysis
No pathologic findings (idiopathic type) 13 18 0 31
Extensive transmural necrosis of the bowel wall 0 6 2 8
Lymphoid hyperplasia 2 3 2 7
Burkitt lymphoma 0 2 1 3
Meckel diverticulum 1 2 0 3
Enteric duplication cyst 0 0 2 2
Juvenile polyps 0 1 1 2
Adenovirus infection of appendix 0 1 0 1
All 26 38 8 72

Thirty children underwent appendectomy in
addition to intussusception reduction. For the
nine children who underwent ileocecectomy,
the appendix was part of the resected cecum.
Histopathologic Analysis
For 57 of the 72 children (79.2%) who un-
derwent surgical reduction of intussuscep-
tion, bowel specimens were resected and were
available for histopathologic analysis. For the
remaining 15 children (20.8%), surgical re-
duction only was performed, and no bowel
specimen was obtained for further analysis.
For 31 of the 57 bowel specimens collected,
histopathologic analysis did not reveal an un-
derlying bowel abnormality, and the intus-
susception was considered to be idiopathic.
In eight bowel specimens (14.0%), extensive
mucosal and submucosal inflammation as-
sociated with focal hemorrhagic transmural
necrosis was noted, consistent with a clinical
history of prolonged intussusception (Figs. 7
and 8). For 18 specimens (31.6%), pathologic
findings revealed the presence of an underly-
ing lead point as the cause of the intussuscep-
tion. The most common pathologic lead points
were lymphoid hyperplasia (n = 7), Burkitt
lymphoma (n = 3), Meckel diverticulum (n =
3), enteric duplication cyst (n = 2), juvenile
polyp (n = 2), and adenovirus appendicitis
(n = 1) (Fig. 9). The mean age of the children
who had a pathologic lead point was 2.6 years
(range, 1.5 months to 10.9 years).
Further histopathologic analysis of the 39
appendiceal specimens obtained from the 30
children who underwent appendectomy and
the nine children who underwent ileocecec-
tomy revealed no pathologic findings in 21
specimens (53.8%); acute mucosal inflam-
matory changes and vascular congestion in
17 specimens (43.6%), presumably because
of the prolonged involvement of the appendix
in the intussusception; and one case (2.6%)
for which adenovirus infection was recog-
nized as the pathologic abnormality underly-
ing intussusception (Fig. 5).
Table 2 presents data on histopathologic
confirmation for cases of failed intussuscep-
tion reduction.
Comparison of Ultrasound Findings With
Surgical Findings, Pathologic Findings, or Both
High-resolution screening ultrasound re-
vealed multiple enlarged lymph nodes in the
intussusception mass in 46 children with irre-
ducible intussusception. For 45 of these chil-
dren, surgical findings, pathologic findings,
or both were considered idiopathic on the ba-
sis of intraoperative findings, pathologic find-
ings, or both; however, in the remaining child,
lymphadenopathy was caused by Burkitt lym-
phoma. For this child with Burkitt lymphoma,
ultrasound suggested the correct diagnosis on
the basis of bulky and confluent lymphade-
nopathy and the age of the child at the time
of clinical presentation. Ultrasound accurate-
ly identified the lead points in each of the two
children who had duplication cysts, and these
cysts were subsequently confirmed and re-
sected during surgery.
For three of 57 children who underwent
surgical reduction of intussusception, ultra-
sound failed to diagnose the causative bow-
el abnormality (i.e., Meckel diverticulum,
Burkitt lymphoma, or appendicitis). Ultra-
sound was not performed before contrast en-
ema for five children who were later found
to have an underlying pathologic lead point
at the time of surgery. Of these patients, two
had Meckel diverticulum, two had juvenile
polyps, and one had Burkitt lymphoma.
Finally, color Doppler sonographic find-
ings complemented gray-scale ultrasound
findings for 53 of 57 children, revealing the

TABLE 3: Comparison of Several Imaging Features Evaluated With Ultrasound in 57 Children With Successful
­Intussusception Reduction Versus 57 Children With Failed Intussusception Reduction
Children With Successful Children With Failed
Ultrasound Feature ­Intussusception Reduction, No. (%) Intussusception Reduction, No. (%)
Trapped fluid between intussusception layers 9 (15.8) 5 (8.8)
Peritoneal free fluid 5 (8.8) 7 (12.3)
Presence of lymph nodes 29 (50.9) 46 (80.7)
Blood flow on color Doppler ultrasound, power Doppler ultrasound, or both 57 (100) 57 (100)

AJR:207, August 2016 429


presence of blood flow within the intussus-
ception mass in all 53 children. However, in
seven cases, although color Doppler exami-
nation demonstrated blood flow within the
involved bowel segments, extensive ischemic
changes with transmural necrosis were found
Downloaded from www.ajronline.org by 180.251.162.88 on 02/13/18 from IP address 180.251.162.88. Copyright ARRS. For personal use only; all rights reserved
during subsequent surgery.
Comparisons Between Children With Failed
Intussusception and Control Subjects
Tables 3 and 4 present comparisons of sev-
eral imaging features evaluated using ultra-
sound and contrast enema in children with
successful intussusception reduction versus
children with failed intussusception reduc-
tion. The frequency of trapped fluid insinu-
ating between the intussusception layers and
the presence of peritoneal free fluid was sim-
ilar between the two groups. The presence
of enlarged lymph nodes within or adjacent
to the intussusception mass was statistically
significantly more common in children with
failed intussusception reduction (p < 0.05).
Of interest, blood flow was observed in all
children in both comparison groups, includ-
ing children with intraoperative findings of
extensive ischemic bowel necrosis.
In terms of contrast enema findings, colon
redundancy rates were similar between the
two comparison groups, whereas the dissect-
ing sign occurred in only 14.9% of the study
population and was significantly more com-
mon in the group for whom intussusception
reduction failed. Finally, the results of the
present study show that the more distally the
mass was initially located, the higher the risk
that intussusception reduction would fail.
Complications
Two children had bowel perforation occur
during enema reduction. One perforation oc-
curred when water-soluble contrast enema was
performed, and the other occurred when air
contrast enema was performed. Both children
were immediately and safely transferred to the
operating room, where they underwent laparot-
omy. In the first case, the intussusception could
not be completely reduced, and an ileocecec-
tomy was performed. The site of perforation
could not be identified macroscopically, and
only on histopathologic analysis was the site of
mucosal ulceration found to be proximal to the
terminal ileum. In the second case, there was
free air on entering the abdomen, and there was
no fecal contamination. Extensive bowel ex-
ploration failed to reveal the site of perforation.
The intussusception was manually reduced,
and an appendectomy was also performed.
430 AJR:207, August 2016
Ntoulia et al.
TABLE 4: Comparison of Several Imaging Features Evaluated With Contrast
Enema in 74 Children With Successful Intussusception Reduction
Versus Children With Failed Intussusception Reduction
Children With Successful
Intussusception Reduction,
Contrast Enema Feature
Colon redundancy
Dissecting sign
Initial location of intussusception mass
Ileocecal valve
Hepatic flexure
Transverse colon
Splenic flexure
Sigmoid colon
Length of Stay in the Hospital
The mean (± SD) length of stay in the hos-
pital for children who underwent laparosco-
py after failed enema reduction was 2.9 ± 2.2
days and was statistically significantly short-
er than the length of stay for children who
underwent laparotomy, including those who
required conversion of the surgical proce-
dure to laparotomy (p = 0.019). In this group
of children, the mean length of hospital stay
was 5.9 ± 5.6 days.
Recurrence
No case of intussusception recurrence af-
ter surgical reduction was noted.
Discussion
The present study is a retrospective re-
view of all cases of intussusception recorded
from 2005 to 2013 for which reduction was
performed using a contrast enema technique
failed. During this prolonged period, signifi-
cant advances in imaging modalities signifi-
cantly improved diagnostic imaging and in-
fluenced medical practices. High-resolution
ultrasound has been established as the im-
aging modality of choice for the diagnosis
of intussusception in children. Although ul-
trasound is well known to be largely operator
dependent, several studies have shown that
ultrasound has a high accuracy for the reli-
able diagnosis of intussusception, with sen-
sitivity of up to 98–100%, specificity of 88–
100%, and negative predictive value of 100%
[5, 8]. False-negative results may occur in as-
sociation with transient small bowel intussus-
ception and the presence of thickened bowel
wall in some children with acute gastrointes-
tinal conditions that may produce ultrasound
features that mimic intussusception. Liouba-
shevsky et al. [16] previously showed that a
Children With Failed
­Intussusception Reduction,
No. (%) No. (%)
9 (12.2) 10 (13.5)
3 (4.1) 11 (14.9)
15 (20.3) 7 (9.5)
42 (56.8) 27 (36.5)
10 (13.5) 22 (29.7)
4 (5.4) 8 (10.8)
1 (1.4) 10 (13.5)
suspected intussusception mass with a mean
diameter of 1.42 cm (range, 0.8–3.0 cm) cor-
related with ileoileal intussusception. In our
study population of children with failed in-
tussusception reduction, screening ultra-
sound correctly identified the presence of il-
eocolic intussusception in 98.2% of cases. In
one case only, the smaller diameter of the de-
picted mass erroneously indicated a diagnosis
of small bowel intussusception, although the
mass was subsequently proven to be ileocolic
intussusception on contrast enema.
Moreover, in 116 of 543 children (21.4%),
contrast enema was performed without a pre-
vious ultrasound examination having been
performed, on the basis of high clinical sus-
picion and abdominal radiography findings.
These children had no evidence of intussus-
ception at the time that they underwent enema.
These cases were encountered before ultra-
sound was established as a first-line imaging
tool for intussusception. Later, as screening ul-
trasound began to be used more frequently, a
significant decrease in the number of unnec-
essarily performed contrast enemas was not-
ed, and limiting enemas to only therapeutic
use for intussusception reduction resulted in a
significant reduction in the overall exposure of
children to radiation for diagnostic purposes.
In addition to accurately establishing the
diagnosis of intussusception, ultrasound can
also be very useful in correctly identifying the
underlying lead point, which may significant-
ly influence the treatment options for these pa-
tients. Ultrasound correctly identified lymph
nodes in all cases of idiopathic intussuscep-
tion and in cases in which enteric duplica-
tion cysts were the lead points. However, ul-
trasound failed to detect Meckel diverticulum
or appendicitis as a lead point in any case, and
it identified only one case of Burkitt lesions.
 Radiologic and Other Findings of Failed Intussusception Reduction in Children
Of interest, and contrary to the findings of
earlier reports, color Doppler imaging was
not helpful in identifying the nonvascular-
ized tissue and the potential for failed reduc-
tion in seven patients with extensive isch-
emic changes and transmural necrosis [17].
Downloaded from www.ajronline.org by 180.251.162.88 on 02/13/18 from IP address 180.251.162.88. Copyright ARRS. For personal use only; all rights reserved
It is possible, however, that these ischemic
changes developed in the period between the
ultrasound examination and subsequent sur-
gical management.
Previous studies have described specific
imaging signs noted during enema reduction
that are considered to be highly associated
with intussusception irreducibility, includ-
ing the dissecting sign and colon redundan-
cy. Fishman et al. [18] showed that irrevers-
ible entrapment of contrast medium between
the intussusceptum and the intussuscipiens is
an indication to discontinue reduction and in-
stitute surgical treatment, especially for pa-
tients younger than 2 years. Similarly, colon
redundancy was reported to prevent adequate
pressures being achieved during an attempt at
enema reduction [19]. These signs were rare
A
Fig. 8—3-month-old male infant with bloody stools and abdominal pain.
A, Image obtained during administration of air contrast enema shows intussusception mass (arrows) in right upper quadrant.
B, Photograph of gross specimen of resected ileocecal region shows multifocal areas of transmural hemorrhage and necrosis involving mucosa and submucosa of bowel wall.
C, Photomicrograph (H and E, ×50) of histopathologic specimen of same area shows hemorrhagic necrosis (arrows) with absence of mucosal glands.
AJR:207, August 2016 431
in the present study, with the dissecting sign
present and sigmoid colon redundancy noted
in 14.9% and 13.5% of failed cases of reduc-
tion, respectively. In the current study, screen-
ing ultrasound was able to detect trapped fluid
between the intussusceptum and the intussus-
cipiens layers in 15.8% of the children with
failed intussusception reduction who under-
went ultrasound before undergoing contrast
enema. Contrast enema, on the other hand,
similarly revealed the presence of the dis-
secting sign and colon redundancy in some
children for whom intussusception reduction
failed, accounting for 14.9% and 13.5% of
failed cases of reduction, respectively.
In the present study, successful reduction
of intussusception occurred in 82.7% of chil-
dren who underwent reduction with the use
of an enema technique [6, 20, 21]. This is
consistent with success rates of 70–92% re-
ported in the published literature.
Contrast enema failed to reduce intussus-
ception in 13.6% of 74 children, and these
patients therefore underwent surgical man-
agement; 36.1% of these surgically treated pa-
tients were subsequently treated laparoscopi-
cally. Laparoscopic reduction has been shown
to be a safe and effective approach in chil-
dren [22–24]. Recent advances in minimally
invasive surgery have resulted in high reduc-
tion rates, low complication rates, low overall
postoperative morbidity, and a shorter length
of stay in the hospital compared with those
noted with use of the open surgical approach.
As the practice of laparoscopy has increased,
this approach has become a reasonable alter-
native—or even the preferred approach—for
some surgeons. Laparotomy was performed
for 52.8% of the children in our study. In two
of these children, the intussusception self-re-
duced by the time of surgery, and primary lap-
B C
aroscopy may have spared the patients from
undergoing an open procedure.
Conversion to open surgery may be neces-
sary in some cases. In the present study, the
surgeon opted to convert to an open reduction
in 11.1% of cases because of the presence of
pathologic lead points that did not allow lapa-
roscopic reduction. The overall mean length
of stay in the hospital was significantly shorter
in the laparoscopically treated group than in
the group who underwent laparotomy (2.9 ±
2.2 vs 5.9 ± 5.6 days), with no complications
and no recurrence rates reported. These find-
ings further support an initial attempt at lapa-
roscopic management.
Surgical findings, histologic findings, or
both showed that most of the intussuscep-
tions (75.7%) in the present study were idio-
pathic. This finding is somewhat lower than
the higher frequency (90–95%) of idiopathic
intussusception reported in previous studies
[7–9], but it may be a reflection of the wider
age range of the patients in the present study.
Idiopathic intussusception is likely related to
hypertrophied lymphoid tissue in the terminal
ileum. The higher incidence noted in spring
and fall suggests that there may be a predis-
posing viral cause. Several viral agents have
been considered in the pathogenesis of intus-
susception, including adenovirus, rotavirus,
enterovirus, human herpesvirus 6 and hu-
man herpesvirus 7, cytomegalovirus, and Ep-
stein-Barr virus. It has recently been reported
that the risk of intussusception increases af-
ter administration of rotavirus vaccines [25].
An association between viral replication and
intussusception has been suggested, but the
mechanism is unknown. Postlicensing stud-
ies have shown that this risk is higher during
the first week after administration of the first
or second dose [26, 27].
 Ntoulia et al.

Pathologic lead points were identified in
25% of cases in the present study and includ-
ed lymphoid hyperplasia, Burkitt lympho-
ma, Meckel diverticulum, enteric duplication
cyst, juvenile polyps, and adenovirus infec-
tion of the appendix. Although previous stud-
Downloaded from www.ajronline.org by 180.251.162.88 on 02/13/18 from IP address 180.251.162.88. Copyright ARRS. For personal use only; all rights reserved
ated. Incidental appendectomy was also per-
formed for 41.6% of cases, whereas the ap-
pendix was resected during ileocecectomy in
12.5% of cases. It should be noted that true
adenovirus infection of the appendix was de-
tected in one case only.
bowel, or the uncovering of an occult perfora-
tion during reduction. The section of the colon
surrounding the intussusceptum has the high-
est risk for perforation [29].
The results of the present study indicate
that perforation presumably was contained

ies have reported the presence of lead points
in an older group of patients, with more than
20% of lead points noted in patients older
than 2 years, the mean age of the patients with
lead points in the present study was 2.6 years
(range, 1.5 months to 10.9 years) [28].
It is worth mentioning that although most
children for whom enema reduction failed
were found to have idiopathic intussuscep-
tion at surgery, bowel biopsy and subsequent
histopathologic analysis of the specimens
were performed for 80% of children evalu-
In the present study, we encountered two
cases of bowel perforation during contrast en-
ema reduction (one performed with air con-
trast medium and one performed with wa-
ter-soluble contrast medium), accounting for
0.4% of all enema reduction attempts. Howev-
er, the perforation site could not be identified
during laparotomy, and no fecal contamina-
tion was seen in the abdominal cavity. Pre-
vious studies have shown that colonic perfo-
ration might be caused by a sudden increase
in intraluminal pressure, a rupture of necrotic
within the intussusception in the first case
and that microperforation might have oc-
curred in the second case. At our institution,
the surgical team is always notified before an
attempt at intussusception reduction begins,
and a surgeon is always present during ra-
diologic reduction of intussusception in case
procedural complications occur that require
immediate intervention, such as bowel per-
foration and tension pneumoperitoneum. Re-
cently, there has been increased discussion
in the literature about the practical benefit of

A B

C D
Fig. 9—8-month-old male infant with crampy abdominal pain and diarrhea.
A, Image obtained during administration of water-soluble enema for intussusception reduction shows persistent filling defect (arrows) at area of cecum and terminal ileum.
B, Gray-scale ultrasound image shows hyperechoic area (arrows) in center of intussusception mass with associated small volume of trapped fluid, which was
retrospectively correlated with Meckel diverticulum at time of surgery.
C, Photograph of postsurgical specimen of cecum reveals Meckel diverticulum (arrow).
D, Photomicrograph (H and E, ×100) of histopathologic specimen of same area shown in C confirms presence of Meckel diverticulum with gastric mucosa (arrows).

432 AJR:207, August 2016

 Radiologic and Other Findings of Failed Intussusception Reduction in Children
this approach [30]. In reality, even if a per-
foration occurs, the patient will not undergo
surgery in the radiology suite. Proper resus-
citation is immediately initiated, the operat-
ing room is prepared, and the patient eventu-
ally is safely transferred.
Downloaded from www.ajronline.org by 180.251.162.88 on 02/13/18 from IP address 180.251.162.88. Copyright ARRS. For personal use only; all rights reserved
The main limitation of the present study is
related to its retrospective design. By evalu-
ating previously performed imaging examina-
tions that were available on our PACS, we did
not have the advantage of real-time evalua-
tion. In addition, these studies were performed
over a relatively extended period during which
significant advances in ultrasound equipment
and imaging guidelines for intussusception
occurred that might have further improved the
diagnostic capabilities. Moreover, although
the same departmental standards were con-
sistently applied, it was not possible to expect
uniformity in the performance of ultrasound
and contrast enemas, given the period when
the study was performed and the number of
radiologists performing these studies. In the
same way, we retrospectively reviewed the
available surgical notes. The experience and
level of training of the surgeon were not taken
into consideration and were not analyzed in
terms of decision making.
Conclusions
Imaging signs that were previously con-
sidered to be potentially predictive of failed
reduction of intussusception are actually rare
in children with failed intussusception re-
duction. Among those signs, only trapped
fluid, as noted on ultrasound, and the pres-
ence of the dissecting sign on imaging when
enema is used appear to have a significantly
higher frequency in the failed cases. In ad-
dition, screening ultrasound can significant-
ly decrease the number of unnecessary con-
trast enemas performed, but it has limitations
in the detection of certain lead points, such
as Burkitt lymphoma and Meckel diverticu-
lum, as well as in the identification of evolv-
ing ischemia. Intussusception can be safely
and effectively reduced during laparoscopy
or laparotomy; however, intussusception re-
ductions performed using laparoscopy are
associated with a significantly shorter length
of stay in the hospital. In most cases, failure
of enema reduction is related to idiopath-
ic intussusception. Underlying pathologic
lead points appear in 25% of cases, the most
common of which are lymphoid hyperplasia,
Burkitt lymphoma, Meckel diverticulum, en-
AJR:207, August 2016 433
teric duplication cyst, juvenile polyp, and ad-
enovirus appendicitis.
References
1. Lochhead A, Jamjoom R, Ratnapalan S. Intussus-
ception in children presenting to the emergency de-
partment. Clin Pediatr (Phila) 2013; 52:1029–1033
2. Waseem M, Rosenberg HK. Intussusception.
­Pediatr Emerg Care 2008; 24:793–800
3. Parashar UD, Holman RC, Cummings KC, et al.
Trends in intussusception-associated hospitaliza-
tions and deaths among US infants. Pediatrics
2000; 106:1413–1421
4. Ekenze SO, Mgbor SO. Childhood intussuscep-
tion: the implications of delayed presentation.
Afr J Paediatr Surg 2011; 8:15–18
5. Daneman A, Navarro O. Intussusception. Part 1.
A review of diagnostic approaches. Pediatr
­Radiol 2003; 33:79–85
6. Daneman A, Navarro O. Intussusception. Part 2.
An update on the evolution of management.
­Pediatr Radiol 2004; 34:97–108; quiz, 187
7. Navarro O, Daneman A. Intussusception. Part 3.
Diagnosis and management of those with an iden-
tifiable or predisposing cause and those that reduce
spontaneously. Pediatr Radiol 2004; 34:305–312;
quiz, 369
8. Navarro O, Dugougeat F, Kornecki A, Shuckett B,
Alton DJ, Daneman A. The impact of imaging in
the management of intussusception owing to
pathologic lead points in children: a review of 43
cases. Pediatr Radiol 2000; 30:594–603
9. Williams H. Imaging and intussusception. Arch
Dis Child Educ Pract Ed 2008; 93:30–36
10. Applegate KE. Intussusception in children: imag-
ing choices. Semin Roentgenol 2008; 43:15–21
11. Chua JH, Chui CH, Jacobsen AS. Role of surgery
in the era of highly successful air enema reduc-
tion of intussusception. Asian J Surg 2006;
29:267–273
12. Naiditch JA, Rigsby C, Chin A. Delayed repeated
enema and operative findings after unsuccessful
primary enema for intussusception. Eur J Pediatr
Surg 2012; 22:404–408
13. Talabi AO, Sowande OA, Etonyeaku CA, A ­ dejuyigbe
O. Childhood intussusception in Ile-ife: what has
changed? Afr J Paediatr Surg 2013; 10:239–242
14. Samad L, Marven S, El Bashir H, Cameron JC,
Lynn R, Taylor B. Intussusception in children less
than 12 months of age: a UK national surveillance
study. J Pediatr Surg 2008; 43:2136
15. Ocal S, Cevik M, Boleken ME, Karakas E. A com-
parison of manual versus hydrostatic reduction in
children with intussusception: single-center expe-
rience. Afr J Paediatr Surg 2014; 11:184–188
16. Lioubashevsky N, Hiller N, Rozovsky K, Segev L,
Simanovsky N. Ileocolic versus small-bowel in-
tussusception in children: can US enable reliable
differentiation? Radiology 2013; 269:266–271
17. Lim HK, Bae SH, Lee KH, Seo GS, Yoon GS. As-
sessment of reducibility of ileocolic intussuscep-
tion in children: usefulness of color Doppler so-
nography. Radiology 1994; 191:781–785
18. Fishman MC, Borden S, Cooper A. The dissection
sign of nonreducible ileocolic intussusception.
AJR 1984; 143:5–8
19. Zanca P. Redundancy of the sigmoid colon. U S
Armed Forces Med J 1957; 8:1303–1310
20. Navarro OM, Daneman A, Chae A. Intussuscep-
tion: the use of delayed, repeated reduction at-
tempts and the management of intussusceptions
due to pathologic lead points in pediatric patients.
AJR 2004; 182:1169–1176
21. Beres AL, Baird R, Fung E, Hsieh H, Abou-Khalil
M, Gerstle JT. Comparative outcome analysis of
the management of pediatric intussusception with
or without surgical admission. J Pediatr Surg
2014; 49:750–752
22. Wei CH, Fu YW, Wang NL, Du YC, Sheu JC.
Laparoscopy versus open surgery for idiopathic
intussusception in children. Surg Endosc 2015;
29:668–672
23. Hill SJ, Koontz CS, Langness SM, Wulkan ML.
Laparoscopic versus open reduction of intussuscep-
tion in children: experience over a decade. J Lapa-
roendosc Adv Surg Tech A 2013; 23:166–169
24. Apelt N, Featherstone N, Giuliani S. Laparoscopic
treatment of intussusception in children: a system-
atic review. J Pediatr Surg 2013; 48:1789–1793
25. Haber P, Parashar UD, Haber M, DeStafano F. In-
tussuseption after monovalent rotavirus vaccine:
United States Vaccine Reporting System
(VAERS), 2008–2014. Vaccine 2015; 33:4873–
4877
26. Haber P, Patel M, Izurieta HS, et al. Postlicensure
monitoring of intussusception after RotaTeq vaccina-
tion in the United States, February 1, 2006, to Sep-
tember 25, 2007. Pediatrics 2008; 121:1206–1212
27. Haber P, Patel M, Pan Y, et al. Intussusception af-
ter rotavirus vaccines reported to US VAERS,
2006-2012. Pediatrics 2013; 131:1042–1049
28. Rubinstein JC, Liu L, Caty MG, Christison-Lagay
ER. Pathologic leadpoint is uncommon in ileo-
colic intussusception regardless of age. J Pediatr
Surg 2015; 50:1665–1667
29. Mahboubi S, Sherman NH, Ziegler MM. Barium
peritonitis following attempted reduction of intus-
susception. Clin Pediatr (Phila) 1984; 23:36–38
30. Nguyen HN, Kan JH, Guillerman RP, Cassady CI.
Intussusception revisited: is immediate on-site
surgeon availability at the time of reduction nec-
essary? AJR 2014; 202:432–436