This is a 5-month old female who presented to a 
rural emergency department because of lethargy, 
fussiness, vomiting, and blood in the stool.  An 
abdominal mass was noted on exam and 
intussusception was suspected.  Abdominal films and a 
barium enema were done.  The barium enema 
demonstrated intussusception which was successfully 
reduced with the BE.  This child was then transported to 
a children's hospital for hospitalization under the care of 
a pediatric surgeon.  Overnight she was kept NPO with 
IV fluids infusing.  After improving overnight, she was 
started on oral fluids in the morning.  However, she 
vomited twice that afternoon.  Her abdominal exam was 
benign but because of her previous intussusception, an 
abdominal series was ordered.

View Abdominal Series:  Upright view.


View Abdominal Series:  Supine view.

     These radiographs showed some dilated loops and 
a paucity of bowel gas.  There is a small amount of 
barium remaining (following the barium enema done at 
the rural hospital).  The liver edge is not well defined.  
This is sometimes seen in intussusception but it is not 
specific.  The more specific plain film signs of 
intussusception (the crescent sign and the target sign) 
are not present (see Case 2 in Volume 1).  Her 
abdominal radiographs were felt to be non-specific at 
this point.    She continued to feed and her symptoms 
resolved by morning at which time she was discharged.
     Her abdominal series discovered a congenital 
dislocated left hip (identified by the radiologist) which 
was previously undiagnosed on routine newborn 
screening exams.  She was referred to an orthopedic 
surgeon for this condition.
     Congenital hip dislocation is part of the spectrum of 
congenital dysplasia of the hip.  Hip dysplasia can 
range from a barely detectable instability of the hip to 
a non-reducible dislocation of the hip.  Very mild cases 
may resolve on their own without any long term 
morbidity.  More severe cases result in the progression 
of symptoms and if not diagnosed at an early stage can 
lead to an irreversibly damaged hip.
     Early diagnosis and treatment of congenital hip 
dysplasia is required to prevent permanent injury to the 
hip.  Physicians providing primary care for infants 
routinely screen for hip dysplasia through routine 
screening examination methods.  However, there is no 
clinical finding or screening method that can reliably 
detect all cases.  For this reason, patients with an 
undiagnosed occult hip dysplasia, may seek treatment 
in an E.D. for an abdominal condition.  If an abdominal 
radiograph is ordered on such a patient, the diagnosis 
may be apparent radiographically if one is astute 
enough to survey the bony structures.
     Failing to identify a congenital hip dislocation on a 
radiograph may result in liability if the patient's condition 
is ultimately diagnosed after irreversible hip damage is 
sustained.  Protection from liability is not afforded even 
if the radiograph was ordered for a different reason.  
Radiographs are usually kept indefinitely.  Thus, at any 
time in the future, our interpretation of a radiograph may 
be subjected to a retrospective review.
     In older children with a well ossified hip, it is usually 
obvious if the hip is dislocated since the well developed 
femur head is easily seen to fit symmetrically within the 
acetabulum bilaterally.  In younger children where the 
acetabulum is not fully ossified and the femur head is 
non-ossified or is just beginning to ossify, the 
relationship between the femoral head and the 
acetabulum is not easy to assess radiographically.  In
infants, ultrasound of the hip may be the diagnostic
method of choice in defining the anatomy of a 
suspected dysplastic hip.

View Hip Diagram.

     A horizontal line is drawn as shown (through the 
triradiate cartilages).  Vertical lines are drawn as shown 
(through the lateral edge of the acetabulum).  The black 
arc drawn over the patient's left hip (right side of the 
image) outlines the upper border of the acetabulum.  
The vertical lines are drawn through the lateral edge of 
this border.  These perpendicular lines form four 
quadrants about each hip.  The ossified femoral head 
should be normally located in the inferior medial 
quadrant.  In this case, the femoral heads are not 
ossified so it is not visible on the radiograph.  It is not 
hard to imagine the position of the femoral head 
(epiphysis) located proximal to the femur along its long 
axis.  The patient's left femoral head (right side of the 
image) would be in the superior lateral quadrant 
(indicating dislocation), while the patient's right femoral 
head (left side of the image) would be in the inferior 
medial quadrant (normal position).  If the femoral head 
is not ossified, the medial "beak" of the proximal 
femoral metaphysis should be in the medial inferior 
quadrant.  In this case the white arrows point to the 
"beaks" on both sides.  The patient's left femoral beak 
is in the inferior lateral quadrant (indicating dislocation), 
while the patient's right femoral beak is located in the 
medial inferior quadrant (normal position).
     Shenton's line is a smooth arc drawn from the 
medial femoral neck through the superior margin of the 
obturator foramen.  If this arc cannot be drawn 
smoothly, dislocation is suspected.  Note the smooth 
arc of Shenton's line drawn on the patient's normal right 
hip (white arc) while this arc cannot be drawn over the 
patient's dislocated left hip.  In the abnormal left hip, the 
arc is discontinuous.
     Children may occasionally present to the E.D. for 
evaluation of a limp if they have no primary care 
provider or if the patient's primary care provider fails to 
properly evaluate the child to the parent's satisfaction.  
When the child begins walking, the most common 
diagnostic sign is an abnormal gait.  This can be so 
subtle that it may be difficult to appreciate.  It is often 
best to rely on the parents' assessment of the child's 
gait since they are usually more sensitive at observing 
such subtle abnormalities.  Other signs may be subtle 
and difficult to appreciate.  There may be asymmetry in 
lower extremity length measurements.  Because 
asymmetry may not be present in bilateral hip 
dysplasia, this may be more difficult to detect than 
unilateral disease.  Radiographs should be more 
reliable in identifying abnormalities in the well ossified 
hip.  Ultrasound is more accurate in defining the hip 
anatomy in infants.
     Most infants with congenital hip dysplasia can be 
detected by screening exam measures.  This 
responsibility rests largely with primary care physicians.  
However, not all children have access to a primary care 
physician.  Infant growth and development changes the 
musculoskeletal structure of the hip such that screening 
exam measures become less reliable in detecting hip 
dysplasia as the infant grows.  The Ortolani and Barlow 
signs have usually disappeared by six weeks of age.  
Detection of congenital hip dysplasia outside of this age 
relies on the identification of subtle findings such as 
lower extremity length asymmetry, loss of the normal 
hip flexion contracture, buttock fullness, a hollow 
anterior groin, asymmetry of the perineal, gluteal or 
thigh folds/creases, or a palpably dislocated femoral 
head.
     The Ortolani sign (reduction of the dislocated hip by 
abduction) and the Barlow sign (dislocation of a 
reduced hip by adducting and applying longitudinal 
pressure on the femur) used in newborn screening refer 
to the sequential series of maneuvers testing the 
reducibility and dislocatability, respectively, of the hips.  
If the nomenclature is confusing, it may be more useful 
to describe a hip as being dislocatable/reducible and/or 
as stable/unstable.

References
     The Hip.  In:  Renshaw TS.  Pediatric Orthopedics.  
Volume 28 in the Series, Major Problems in Clinical 
Pediatrics.  Philadelphia, W.B. Saunders, 1986, pp. 
63-89.
     Herring JA.  Congenital Dislocation of the Hip.  In:  
Morrissy RT (ed).  Lovell and Winter's Pediatric 
Orthopedics, 3rd edition.  Philadelphia, J.B. Lippincott 
Co., 1990, pp. 815-851.