The Stomach Flu? - The Target, Crescent, and Absent Liver Edge Signs
Radiology Cases in Pediatric Emergency Medicine
Volume 1, Case 2
Lynette L. Young, MD
Loren G. Yamamoto, MD, MPH
Kapiolani Medical Center For Women And Children
University of Hawaii John A. Burns School of Medicine
     An 11 month old male with history of "stomach flu" 
symptoms two weeks ago that had resolved, now 
presents to the ED with emesis five times the night 
prior, without blood or bilious material.  In the morning 
he had three loose stools with blood but no mucous.  
There are no URI symptoms, and no history of fever.  
He cries intermittently in cycles of 10 to 20 minutes.  
His past medical history is unremarkable.
     Exam:  Vital signs T36.5Ax, P118, RR40, Wt 
50%ile.  He is alert, smiling, and not toxic appearing.  
Skin exam shows good perfusion (capillary refill time 2 
seconds).  Pupils reactive.  Tympanic membranes no 
erythema.  Oral mucosa moist.  Heart regular, no 
murmur.  Lungs clear breath sounds, good aeration.  
Abdomen soft, flat, active bowel sounds, no mass 
palpated.  Testes descended bilaterally, nontender.  No 
anal fissure, stool heme-positive.  Pulses were good.
     A stool culture was sent and an abdominal series 
was obtained.

View abdominal series:  Supine (Flat) view


View abdominal series:  Upright view


     There is a suspicion of a soft-tissue mass in the right 
upper quadrant.  There is some distention of a single 
loop of small bowel in the mid-abdomen and gaseous 
distention of the transverse colon and proximal left 
colon.  No peritoneal free air.  The liver edge is not 
easily identified in these views (the absent liver edge 
sign).  There is a paucity of bowel gas.
     A barium enema demonstrated an intussusception 
at the hepatic flexure which was successfully reduced.

Teaching points and Discussion:
     1.  Intussusception is a common abdominal 
emergency in young children.  A delay in establishing 
the diagnosis leads to a delay in treatment, bowel 
ischemia, and bowel infarction.  An early diagnosis is 
essential.
     2.  The most common is ileocolic, with the lead point 
proximal to the ileocecal valve.  Bloody mucousy stool 
(currant-jelly stool) is a late sign, resulting from 
engorgement of the intestine, edema, and then bleeding 
from the mucosa.  Although this finding is known as 
currant jelly stools, it can resemble blood mixed with 
stool as in dysentery.  This can easily be dismissed as 
being caused by gastroenteritis due to shigella or 
salmonella.  This pitfall can be avoided by considering 
the diagnosis of intussusception in all cases of bloody 
diarrhea and bloody stools.
     3.  Males outnumber females 2:1.  The 3 - 12 month 
old age group is the most common.
     4.  The triad of symptoms:  a) intermittent crampy 
abdominal pain (episodic pain, child may appear 
comfortable in between episodes); b) emesis, and c)
passage of bloody, mucousy stools.  Most patients with 
intussuception do not present with this triad, therefore it 
is not useful to use this set of findings to rule out 
intussusception.
     5.  An abdominal mass is not part of the triad, but 
this finding, that represents the leading head of the 
intussusception, may be helpful in establishing the 
diagnosis.  The mass may be present in any part of the 
abdomen depending on where the intussuception 
originates and where it ends.  This mass is usually 
palpated in the right abdomen, but in severe cases, it 
may be present in the left abdomen if the 
intussusception has passed the splenic flexure and has 
entered the descending colon.
    6.  Plain abdominal films may be normal.  There may 
be evidence of bowel obstruction after 6-12 hours of 
symptoms.  Thus, plain abdominal films cannot be used 
to rule out intussusception.  However, plain films may 
be used to add to the body of clinical evidence 
prompting one to do a barium enema.
     7.  Radiographic signs on plain abdominal films 
include the target sign, the crescent sign, the absent 
liver edge sign, and other signs that are less specific for 
intussusception.

     Target sign:  Two approximately concentric circles of 
fat density to the right of the spine, due to layers of 
peritoneal fat surrounding and within the 
intussusceptum alternating with layers of mucosa and 
muscle.  This sign resembles a very faint target, or 
bull's eye, or doughnut appearance.

View a target sign example. 


     This radiograph shows a classic target sign in the 
right upper quadrant just below the liver.  It resembles a 
chubby doughnut with a puffy center.  It is very subtle.  
You may need to adjust the contrast control on your 
monitor to appreciate it.   This radiograph also shows 
the absent liver edge sign and the crescent sign.  A 
paucity of bowel gas is also noted.

     Crescent sign:  Soft-tissue density mass of the 
intussusceptum projecting into the colon  (leading 
edge).  If the head of the intussusceptum is projecting 
into a gas filled pocket, it will show itself.  It often takes 
on a crescent shape; however, it may also merely 
resemble a protruding head into a gas filled pocket.

View a crescent sign example.


     This radiograph shows a classic crescent sign in the 
left upper quadrant.  This radiograph indicates that the 
head of the intussuception is in the distal transverse 
colon.  Also note that this radiograph demonstrates the 
target sign and the absent liver edge sign.  The target 
sign is again subtle and less noticeable in this film.  You 
may need to adjust the contrast control on your monitor 
to appreciate it.
     Go back to the target sign radiograph to see
if you can appreciate the crescent sign on this
radiograph (left upper quadrant).  In this case, the
crescent is not as crescent shaped, because the 
colonic air pocket surrounding the intussusceptum is 
larger than in the crescent sign film.

View an atypical crescent sign.


     This radiograph shows an atypical crescent sign in 
the right upper quadrant just below the liver.  The head 
of the intussusception is coming up the ascending 
colon.  It can be seen protruding upward into the gas 
filled transverse colon at the hepatic flexure.

     Other non-specific radiographic signs that may 
suggest an intussusception are as follows:
     a.  Abdominal mass:  An absence of bowel gas in 
the area suggesting indirectly that something is pushing 
normal bowel out of the way.
     b.  Small bowel obstruction:  Dilated bowel loops 
and air-fluid levels.  Examine the flat plate (supine view
of this case  

Notice the dilated bowel segments.  They exhibit
smooth bowel walls lacking normal haustrations.  
This is suggestive of a small bowel obstruction.
     c.  Paucity of gas:  Distal to obstruction.
     d.  Loss of subhepatic angle:  The absent liver edge 
sign.
 
     The target and crescent signs are the most accurate 
in making the diagnosis of intussusception on a plain 
film.  The target sign is seen twice as often as the 
crescent sign.  An abdominal mass is most commonly 
seen on radiographs, but it is non-specific.
     8.  Barium enema is the gold standard of diagnosis.  
It often results in a successful reduction of the 
intussusception as well.  Ultrasound and air contrast 
enemas have also been used to diagnose 
intussusception.  The two contraindications to 
performing a barium enema include shock and/or 
radiographic or clinical evidence of bowel perforation.  
Patients with hypovolemic shock should first have their 
intravascular volume restored before undergoing a 
barium enema.  Any patient with evidence of bowel 
perforation should be taken immediately to surgery.
     9.  Vomiting is a common reason to seek emergency 
or acute care.  It is usually the result of a benign cause.  
However, it may be difficult to distinguish serious 
causes from benign causes if the evaluation is 
superficial.  Whenever the chief complaint is vomiting, 
the diagnosis of intussusception should be considered.  
The history and examination should be directed at 
determining whether intussusception is possible based 
on clinical grounds.  The chart should include 
comments in the history regarding the frequency of 
vomiting, the color of the emesis, the presence or 
absence of abdominal pain, the frequency of abdominal 
pain, and the activity level of the child.  Intussusception 
is more likely if the emesis is bilious and/or frequent.  
Intussusception is more likely if the pattern of the pain 
is colicky in nature (intermittent and severe in regular 
cycles 5-20 minutes apart).  Intussusception is more 
likely if the child exhibits lethargy.  The absence of 
these symptoms does not rule out intussusception.  
Patients with intussusception may have all, some, or 
none of these symptoms.  The physical exam portion of 
the chart should document the presence or absence of 
lethargy and an abdominal mass.  The exam should 
include the testes (in males) and the inguinal region 
looking for incarcerated hernias.  The rectal exam and 
stool guaiac results should also be recorded.  Ideally, 
the chart should comment on whether the examiner has 
noted a colicky abdominal pain pattern observed during 
the evaluation period.
     10.  Infants presenting purely with lethargy (no 
vomiting) have often been evaluated for possible 
sepsis.  However, lethargy is a common presentation 
for intussusception despite the absence of all the other 
signs of intussusception. 

References
     1.  Waisman Y.  Intussusception.  In:  Barkin RM 
(ed).  Pediatric Emergency Medicine Concepts and 
Clinical Practice. Chicago, Mosby Year Book, 1992, pp. 
784-786.
     2.  Schnaufer L, Mahboubi S. Intussusception.  In:  
Fleisher GR, Ludwig S (eds).  Textbook of Pediatric 
Emergency Medicine, third edition.  Baltimore, 
Williams and Wilkins, 1993, pp. 1314-1316.
     3.  Ratcliffe JF, Fong S, Cheong I. O'Connell P.  The 
Plain Abdominal Film in Intussusception:  The Accuracy 
and Incidence of Radiographic Signs.  Pediatric 
Radiology 1992; 22:110-111.

Return to Radiology Cases In Ped Emerg Med Case Selection Page

Return to Univ. Hawaii Dept. Pediatrics Home Page

Web Page Author:
Loren Yamamoto, MD, MPH
Associate Professor of Pediatrics
University of Hawaii John A. Burns School of Medicine
loreny@hawaii.edu