A 2 year old boy is brought to the emergency 
department at 2:30 a.m. complaining of throat pain, not 
drinking, drooling, difficulty swallowing, coughing, nasal 
congestion, and fever.  The throat pain, difficulty 
drinking, and drooling started in the afternoon and have 
worsened over the night.  He has had the coughing and 
nasal congestion for 3 days now.  He was given some 
oral acetaminophen 45 minutes ago, but he could not 
swallow it.  His past history is significant for frequent 
ear infections.  His immunizations are "up to date".
     At triage, the nurse notes that he is tired in 
appearance, but alert.  He is given 14 mg/kg of 
acetaminophen rectally.
     Exam VS T38.6 (rectal), P143, R28, BP not taken, 
oxygen saturation 100% in room air.  He is active and 
alert.  Not toxic.  Not irritable.  Drooling is noted but he 
is not tripodding.  Eyes clear.  TM's normal.  Nose with 
thick mucus.  Oral pharynx slightly red.  The palate, 
gums, and oral mucosa were clear of any ulcers or 
petechiae.  No tonsillar enlargement.  No exudates.  
The epiglottis could not be seen.  Neck supple, small 
nodes.  Heart regular without murmurs.  Lungs clear.  
No stridor.  Good aeration.  No retractions.  He coughs 
occasionally and the sound of the cough is not croupy.  
Abdomen benign.  Color and perfusion are good.
     Although he does not look toxic and his airway 
function is assessed as being good, a lateral neck 
radiograph is ordered at 3:00 a.m. to look for a 
retropharyngeal abscess or epiglottitis.

View lateral neck radiograph.

     The original radiograph displayed here was 
excessively dark.  It was read with the assistance of a 
hot light.  This scanned image shown here shows the 
radiograph scanned using standard light intensities and 
another view scanned using a hot light.  Thus the image 
on the right shows the soft tissue structures better while 
overexposing the bony structures.  The pre-vertebral 
soft tissue space is not widened.  This essentially rules 
out a significant retropharyngeal abscess.  The black 
arrow points to the tip of the epiglottis and the white 
arrow points to the pre-epiglottic space (vallecula).  The 
shape of the epiglottis is better seen on the 
overexposed (whiter) image on the right.  The epiglottis 
should normally be thin or triangular in appearance.  In 
this view, it appears to be rounded and somewhat 
thumb-like, which is similar to the classic radiographic 
description of epiglottitis.  The pre-epiglottic space is 
fairly well preserved but it is slightly shallow.  In 
epiglottitis, the pre-epiglottic space is obliterated or very 
shallow due to the inflammation of the epiglottis.  The 
positioning of the patient does not show his neck well 
extended.
     It was felt that this patient probably did not have 
epiglottitis, but the radiograph was suspicious enough 
to worry the physician.  A tongue depressor was used 
to visualize the epiglottis, but it still could not be seen.  
He was sent back to the imaging department to repeat 
his lateral neck radiograph with better positioning.

View repeat lateral neck radiograph.

     This radiograph was again somewhat dark.  The 
epiglottis still appeared to be enlarged (thumb-like) and 
the pre-epiglottic space was still shallow.  At 3:30, it 
was decided to get a radiologist's consultation by 
teleradiology.  The patient was given IM ceftriaxone 
while waiting for the radiologist's call.  The radiologist 
reviewed both radiographs and agreed that they 
appeared to show epiglottitis.
     The child continued to look good.  No stridor was 
noted, but he continued to drool.  VS at 3:40 a.m. T37.8 
(rectally), P145, R40, BP not done, oxygen saturation 
100% in room air.
     The physician was still convinced that the child did 
not have epiglottitis.  After a discussion with his mother, 
the physician looked at the epiglottis using a 
laryngoscope.  The epiglottis was swabbed for a 
culture.  The epiglottis was not enlarged, but it was 
slightly erythematous.
     At 4:30 a.m., an ENT surgeon was called.  He felt 
that it was satisfactory to hospitalize the child in a floor 
bed for observation.  He did not feel that intubation was 
indicated, nor admission to a more advanced or 
intensive care unit for observation.  
     Although the epiglottis was somewhat erythematous, 
its anatomic shape and size were normal.  How could a 
structurally normal epiglottis, cast an abnormal 
radiographic shadow ?  One could only conclude that 
the examiner did not view the epiglottis correctly during 
laryngoscopy or that the radiographic appearance is 
due to some type of artifact that is appearing on two 
separate radiographs.
     Over the next few hours, the patient did not worsen.  
The ENT surgeon reviewed the radiographs and agreed 
that the epiglottis appeared to be enlarged.  Fiberoptic 
laryngoscopy carried out later that morning showed 3+ 
adenoid hypertrophy, mucopurulent discharge, a 
non-enlarged epiglottis, 3+ edema of the vocal cords, 
and a clear subglottic region.
     A second radiologist who looked at the radiographs 
felt that both radiographs showed a normal epiglottis.  
The epiglottis can appear to be rounded if the lateral 
neck is taken at an oblique angle.
     In this case, I thought that it would be very difficult to 
view the epiglottis with a laryngoscope in an awake 2 
year old without sedation.  However, it was surprisingly 
easy.  With the patient supine, it was only necessary to 
use the laryngoscope blade to push down the posterior 
aspect of his tongue.  It was not necessary to insert the 
blade all the way to the epiglottis or the vallecula.  He 
gagged a little, but it was so easy to see the epiglottis 
that it was not necessary to leave the laryngoscope 
blade in place as one would need to for intubation.  The 
normal size and shape of the epiglottis on direct 
visualization in this case should have effectively ruled 
out epiglottitis.  The slight erythema of the epiglottis 
may have been due to a viral infection such as 
laryngitis.