A New Diagnostic Approach to Biliary Atresia With Emphasis

on the Ultrasonographic Triangular Cord Sign: Comparison

of Ultrasonography, Hepatobiliary Scintigraphy, and Liver Needle
Biopsy in the Evaluation of Infantile Cholestasis
By Woo-Hyun

Park,

Soon-Ok

Choi,

Hee-Jung
Taegu,

Lee,
Republic

&&ground/Purpose:
The authors
evaluated
prospectively
the utility
of ultrasonography,
Tc-99m-DISIDA
hepatobiliary
scintigraphy,
and liver needle
biopsy
in differentiating
biliary
atresia
(BA) from
intrahepatic
cholestasis
in 73 consecutive
infants
who had cholestasis.
Methods:
Sixty
three
ultrasonographic
examinations
of 61
infants
with 7.0-MHz
transducer
were carried
out, focusing
on
the fibrous
tissue
at the porta
hepatis.
The authors
defined
the triangular
cord (TC) as visualization
of a triangular
or
tubular
shaped
echogenic
density
just cranial
to the portal
vein bifurcation
on a transverse
or longitudinal
scan.
Results:
Although
17 of 20 ultrasonographic
examinations
from
infants
who had BA denoted
TC, 43 ultrasonographic
examinations
from infants
with either neonatal
hepatitis
(NH)
or other
causes
of cholestasis
denoted
no TC, showing
a
diagnostic
accuracy
of 95% with 85% sensitivity
and 100%
specificity.
Investigation
with Tc-99m-DISIDA
hepatobiliary
scintigraphy
showed
that 24 of 25 infants
who had BA had no
gut excretion,
and 16 of 46 infants
who had either NH or other
causes
of cholestasis
had gut excretion,
showing
a diagnostic
accuracy
of 56% with
96% sensitivity
and 35% specificity.
Therefore,
gut excretion
of tracer
excluded
BA, but no gut
excretion
of tracer
needed
further
investigations
as liver
needle
biopsy.
Forty-four
liver
needle
biopsies
were
carried
out in 19
infants
who had BA and 24 infants
who had either NH or other

NFANTILE CHOLESTATIC JAUNDICE, characterized by persistent conjugated hyperbilirubinemia, has

remained a major diagnostic challenge despite continual
improvement of diagnostic tests and increasing knowledge regarding its pathogenesis. It is important that the
infants who have biliary atresia (BA) should be identified
early because the success of the Kasai procedure is
inversely related to age and is greatest if performed
during the first two months of life.-3 No single test or
imaging modality can reliably define the causes of
infantile cholestasis. Ultrasonography has played a role
in screening infantile cholestasis, mainly focusing on
shape or contractility of the gallbladder.4-10 Lately, we
reported that ultrasonographic triangular code (TC),
which represents a cone-shaped fibrotic mass cranial to
the bifurcation of the portal vein in BA infants, was very
useful in the diagnosis of BA.1,12
Journal

of Pediatric

Surgery,

%I 32, No 11 (Novem

her),

1997: pp 1555-1559

Sang-Pyo

Kim,

Seok-Kil

Zeon,

and

Sang-Lak

Lee

of Korea

causes
of cholestasis.
Although
18 of 20 biopsy
findings
in
infants
who had BA were correctly
interpreted
as having
BA,
23 of 24 biopsy
results
in infants
who had either NH or other
causes
of cholestasis
were
correctly
diagnosed,
showing
a
diagnostic
accuracy
of 93% with 90% sensitivity
and 96%
specificity.
Conclusions:
Since the introduction
of ultrasonographic
TC
sign in the diagnosis
of BA by our institution,
we have found
that it seemed
to be a simple,
time-saving,
highly
reliable,
and non-invasive
tool in the diagnosis
of BA from
other
causes
of cholestasis.
The authors
propose
a new diagnostic
strategy
in the evaluation
of infantile
cholestasis
with emphasis on ultrasonographic
TC sign as first priority
of investigations.
When
the TC is visualized,
prompt
exploratory
laparotomy
is mandatory
without
further
investigations.
When
the TC is not visualized,
hepatobiliary
scintigraphy
is the next
step. Excretion
of tracer
into the small bowel
actually
rules
out BA. Liver needle
biopsy
is reserved
only for the infants
with no excretion
of tracer. The authors
believe
that a correct
decision
regarding
the need for surgery
can be made
in
almost
all cases with infantile
cholestasis
by this multidisciplinary
approach.
J Pediatr Surg 32:1555-7559.
Copyright
o 1997 by W.B. Saunders
Company.
INDEX WORDS:
Biliary
nography,
hepatobiliary

atresia,
infantile
scintigraphy,

cholestasis,
liver
needle

ultrasobiopsy.

The aim of this study is to reassess the relative accuracy and the role of ultrasonography (US), hepatobiliary
scintigraphy (HS), and liver needle biopsy (NBx) with
particular emphasis on ultrasonographic TC in differentiating BA from other causes of intrahepatic cholestasis

From the Departments

qf Pediatric
Surgery, Radiology,
Pathology,
N&ear
Medicine,
and Pediatrics,
Keimyung
Universiry
Dongsan
Medical Center, Taegu, Korea.
Presented at the 30th Annual Meeting of the Pacific Association
of
Pediatric Surgeons, Phoenix, Arizona, May 9-13, 1997.
Supported
by a grant (1996) of the Institute for Medical
Science,
Keimyung
University.
Address
reprint requests to Woo-Hyun
Park, MD, Department
of
Pediatric
Surgery, Keimyung
UniversiQ
Dorzgsan Medical Centel; I94
Dongsan Dong, Taegu 700-712, Rrepublic
of Korea.
Copyright
o 1997 by WB. Saunders Cowtpany
0022.3468/97/3211-0008$03.00/O
1555

1556

PARK ET AL

and to propose a new diagnostic approach in the evaluation of infantile cholestasis.

MATERIALS

AND

METHODS

Seventy-three
consecutive
infants who were suspected of having BA,
neonatal hepatitis (NH), or other causes of cholestatic
jaundice
were
evaluated prospectively,
using US, HS, or NBx during the period from
March 1992 to October 1996 at the Departments
of Pediatric Surgery,
Radiology,
Pathology,
Nuclear
Medicine,
and Pediatrics,
Keimyung
University
Dongsan Medical Center. Choledochal
cyst was excluded in
this study. The patients undergoing
study ranged in age from 12 to 120
days. The infants presented with conjugated
hyperbilirubinemia
(direct
bilirubin, 2 mg% or more) or clay-colored
stool. The averages of total
and direct serum bilirubin
level were 10.9 mg/dL and 6.3 mg/dL,
respectively.
US examinations
were carried out by one of the authors (HJL),
focusing on the fibrous tissue at the porta hepatis. We defined the TC
sign as visualization
of a triangular or tubular shaped echogenic density
just cranial to the portal vein bifurcation
on a transverse or longitudinal
scan (Fig 1). ACUSON
XP-10 ultrasound
equipment
(Acuson, Mountain View, CA) with a 7.0-MHz
transducer
was used in the majority
of
the infants. As described in our previous reports,11.12 the presence of TC
was interpreted
as having BA, and absence of TC was interpreted
as
having NH or other causes of cholestasis.
During the US studies, we
also measured the size and shape of the gallbladder,
its contractility
after feeding, and the dilatation
of the intra- and extrahepatic
biliary
duct. To evaluate the presence of the gallbladder. the infant fasted for at
least 4 hours.
Tc-99m-DISIDA
hepatobiliary
scintigraphy
(HS) was performed
after a 3- to 5-day course of phenobarbital
pretreatment
(5 mg/kg/d).

The infants underwent imaging after intravenous

injection of 1 .O mCi of
Tc-99m-DISIDA.
Images were obtained at 3,5, 10, 15,30,45,60,
120,
and 240 minutes. If necessary, delayed images were obtained up to 24
hours after injection. The images were reviewed
by one of the authors
(SKZ).
Hepatocyte
clearance
and transit time were evaluated
but
interpretation
was mainly dependent on the transit time. No excretion of
tracer in 24 hours was indicative
of having BA, whereas excretion
of
tracer in 24 hours was indicative
of having either NH or other causes of
cholestasis.
Percutaneous
liver needle biopsy (NBx) under the ultrasound
guidance was performed
with local anesthesia using 5-automatic
cutting
needle (18 gauge, Manan Medical Products, Inc, Northbrook,
IL) The
histopathology
was reviewed by one of the authors (SPK). Interpretation of the biopsy specimens was carried out without knowledge
of the
clinical data. Bile duct proliferation
in all visible portal areas was used
as the most valuable criterion
of BA. Ductal proliferation
(more than
average 5 ducts per one portal area) with portal fibrosis was interpreted
to be consistent
with BA. Hepatocellular
degeneration
with no ductal
proliferation
(average 1 to 2 ducts per one portal area) was interpreted to
be consistent
with NH or other type of cholestasis.
Mild ductal
proliferation
(average 3 to 4 ducts per one portal area) with mild or no
portal fibrosis was interpreted
to be consistent
with probable BA. Less
than 0.5 ducts per one portal area was interpreted
to be consistent with
paucity of intrahepatic
bile duct.
The criteria used for the final diagnosis of BA include (1) fibrotic
obstruction
of extrahepatic
biliary ducts grossly at surgery or (2) the
presence of bile duct proliferation,
portal fibrosis, and canalicular
bile
stasis in the liver biopsy specimen.
The criteria used for the final
diagnosis of NH or other causes of cholestasis
were (1) gut excretion
of
tracer in 24 hours on hepatobiliary
scintigraphy
and evidence of clinical

Fig 1. (A) Transverse

and (B) longitudinal
ultrasonograms
disclose a
triangularor tubular-shaped
echogenic
density
(asterisks)
just cranial
to the portal vein (PV) in an infant who has biliary atresia.
(C) The
resected
specimen
of atretic
extrahepatic
biliary tree of the same
infant. FM, fibrotic mass at the porta hepatis; GE, gallbladder.

ULTRASONOGRAPHIC

TRIANGULAR

CORD

IN BA

Table 2. Diagnostic
Accuracy
of Tc-99m-DISIDA
Hepatobiliary
Scintigraphy

improvement,
(2) hepatocellular
degeneration
without portal change in
liver biopsy specimen or evidence of clinical improvement
of 3 months
or more follow-up,
and (3) patent extrahepatic
bile ducts noted via
cholangiography.
The 73 infants turned out to be composed
of 25 cases
with BA, 42 cases with NH, 5 cases with total parenteral
nutrition
(TPN)-associated
cholestasis,
and one case with Alagille syndrome.

Sixty-three US examinations of 61 infants were carried

out, focusing on the fibrous tissue at the porta hepatis.
Twenty US examinations were carried out in 1X infants
who had BA. Seventeen examinations denoted TC
whereas three examinations did not. In two infants who
had BA, the first US examination performed at 33 days
and at 60 days of age, respectively showed absence of TC
but a second US examination performed at 84 days and at
80 days of age respectively, showed positive TC. Of
particular interest is the remaining one BA infant who had
negative TC. It turned out that the fibrotic cord was
masked by an abnormally high and large right hepatic
artery running precisely anterior to the relatively small
fibrotic mass at surgery. Forty-three US examinations
from the 43 infants who had either NH or other causes of
cholestasis showed absence of TC. US in the diagnosis of
BA denoted a diagnostic accuracy of 95% with 85%
sensitivity and 100% specificity (Table 1). With respect to
the US examination of the gallbladder in BA cases, the
gallbladder was invisible in 9 patients, small in 13
patients, and normal (1.5 cm or greater in length) in 3 patients.
Contractility of the gallbladder was noted in 3 patients.
Hepatobiliary Scintigraphy

Seventy-one infants underwent Tc-99m-DISIDA hepatobiliary scintigraphy. Twenty-four of 25 infants who had
BA had no gut excretion, and 16 of 46 infants who had
either NH or other causes of cholestasis had gut excretion, denoting a diagnostic accuracy of 56% with 96%
sensitivity and 35% specificity (Table 2). The mean direct
bilirubin concentration of the infants with intrahepatic
cholestasis was 5.4 mg/dL for those without gut excretion
of tracer and 4.3 mg/dL for those with gut excretion,
respectively.
Pecutaneous Liver Needle Biopsy
Forty-three infants underwent 44 needle biopsies. The
majority of the infants (42 of 43 infants) showed no
excretion of tracer in 24 hours on Tc-99m-DISIDA

NOTE.

Sensitivity,

1. Accuracy

of Diagnosis

of Biliary

Atresia

Infantile cholestasis occurs in a number of disorders

such as infections, metabolic or genetic disorders, and
anatomic abnormalities. Cholestasis is diagnosed when
the conjugated bilirubin fraction comprises more than
20% of total bilirubin or more than 2 mg%.13 Among the
many causes of infantile cholestasis, investigation leads
to a diagnosis of BA or idiopathic NH in 70% to 80% of
infants.13 It is well established that the success of the
Kasai procedure is time dependent.-3 The success of the
Kasai procedrue is greatest if performed during the first 2
months of life.1-3 Clinical features and standard laboratory tests of liver function frequently cannot distinguish
BA from NH or other causes of cholestasis. A large
number of diagnostic tests have been attempted to
distinguish BA from NH or other causes of cholestasis.
Despite continual improvements and changes in the

Biliary

atresia

Neonatal
NOTE.

(n = 20)

hepatitis

Sensitivity,

and other

(n = 43)

85%; specificity,

Accuracy

of Percutaneous

TC
Negative

17
0

3
43

100%; accuracy,

95%.

Liver

Interpretation

of US

TC
Positive

56%

DISCUSSION

by Ultrasonography
lnterm3tation

16

35%; accuracy,

hepatobiliary scintigraphy. In 20 biopsies from 19 infants

with BA, the histopathologic findings were consistent
with BA in 18, showing 90% sensitivity (Table 3). In a
54-day-old BA infant, the histopathologic findings were
those of neonatal hepatitis but exploratory laparotomy
findings showed biliary atresia. In another BA infant, the
histopathologic findings of a first biopsy performed at 43
days of age were those of neonatal hepatitis but follow-up
with a second biopsy obtained at 83 days of age disclosed
the typical findings of BA.
In 24 infants who had either NH or other causes of
cholestasis, the histopathologic findings of 23 infants
were consistent with NH or other causes of cholestasis,
showing 96% specificity. One false-positive case was
observed in an infant who had TPN-associated cholestasis who showed mild ductal proliferation (3 to 4 ducts per
portal area). This infant improved clinically after discontinuation of TPN.

Table 3. Diagnostic
Table

30
96%; specificity,

Scan

Neonatal
Hepatitis/Other

24

(n = 46)

Ultrasonography

by Hepatobiliary

Biiiary
Atresia

Biliary atresia (n = 25)

Neonatal hepatitis and other

RESULTS

Tc-99mDISIDA

Interpretation

Biliary
Atresia

Biliary atresia (n = 20)

Neonatal hepatitis and other
(n = 24)
NOTE.

Sensitivity,

Needle

of Needle Biopsy
Neonatal
Hepatitis/Other

18

1
90%; specificity,

96%; accuracy,

Biopsy

23
93%.

PARK ET AL

diagnostic accuracy of these tests, none has been constantly reliable. Therefore, a multidisciplinary approach
to the evaluation of infantile cholestasis is required.
Ultrasonography is a simple and noninvasive procedure that has been used in all cases with infantile
cholestasis as a first step in the work-up of these patients.
An abdominal US is useful in the detection of a choledochal cyst, but it is not so helpful in differentiating BA
from NH or other causes of intrahepatic cholestasis. Most
investigators have focused on the shape or change in size
of the gallbladder.4-*0 Nonvisualization of the normal
gallbladder was thought to suggest BA,4-10 but this may
occur in association with both BA and severe intrahepatic
cholestasis.4,xFurthermore, several investigators have
documented BA cases with a normal gallbladder as seen
in three of 25 infants who had BA in our study. Although
the gallbladder is usually small or nonvisible in BA, a
normal gallbladder does not exclude the diagnosis of BA.
Conversely, nonvisualization of the gallbladder does not
exclude the diagnosis of NH or other intrahepatic cholestatic disorders.
Green and Carroll6 and Ikeda et al have suggested that
BA can be excluded when the change in the gallbladder
size is observed after feeding on serial US examinations.
Although in infants who have BA, the gallbladder is
usually small and difficult to identify, it is not surprising
to demonstrate the contractility of the gallbladder after
feeding, because a patent bile duct from the gallbladder to
the duodenum can be seen in 19% to 22% of cases.2,3,5
Three of 25 BA infants demonstrated contractility of the
gallbladder after feeding in this study. We believe that
ultrasonographic evaluation of the gallbladder shape,
size, and contractility is not such a reliable test for the
differentiation of BA from intrahepatic cholestasis.
With the development of high-resolution gray-scale
recording and real-time imaging technique with 7.0 MHz
or 10 MHz transducer, we recently reported the sonographic feature of a fibrous cone at the porta hepatis as a
triangular-shaped or tubular-shaped echogenic density
just cranial to the portal vein bifurcation on transverse or
longitudinal scan. 11-12The echogenic density was confirmed to represent the fibrous remnant in the porta
hepatis of BA cases at surgery. We named this echogenic
density the triangular cord (TC). Although we are still in
a learning period, the TC appears to be a very specific
ultrasonographic finding representing the fibrous cone in
BA cases, showing overall diagnostic accuracy of 95%
with 85% sensitivity and 100% specificity on the basis of
our study. A very experienced and enthusiastic radiologist
is indispensible for precise interpretation of the TC.
In light of our experience, there can be pitfalls with this
proposal. False-negative TC may occur in some BA
infants because of the patterns of unusual hepatic radicles
such as hypoplastic, aplastic, or fibrous hepatic ducts,3
and in the early stage of some BA cases. The fibrotic

remnant at the porta hepatis may be inconspicuous at the

early stage of the disease but it can be recognizable with
time as seen in our two cases. A false-negative TC caused
by a large right hepatic artery running over the relatively
small fibrotic mass at the porta hepatis as seen in our one
BA case, may also occur.
Tc-99m-IDA hepatobiliary scintigraphy can be used to
evaluate the degree of hepatocyte dysfunction using
hepatocyte clearance and bile duct patency using excretion of tracer into the small bowel. The majority of the
infants who have BA are expected to have no excretion of
tracer into the small bowel. The diagnosis of infantile
cholestasis is mostly dependent on the presence or
absence of tracer excretion rather than degree of hepatocyte clearance.14-lg The majority of the literature report
that the sensitivity of the Tc-99m-IDA hepatobiliary
scintigraphy for the diagnosis of BA is as high as 97% to
lOO%, whereas the specificity varied from the reports,
ranging from 33% to 91%.14-lg In our study, 16 of 46
infants who had either NH or other causes of cholestasis
showed excretion of tracer, denoting 35% specificity.
This discrepancy in specificity is thought to be attributable to the different age, diagnosis, and the severity of
diseases in the study population. Thus, excretion of tracer
can exclude BA but no excretion of tracer requires further
investigation such as percutaneous needle biopsy because
of its low specificity. Infants with low birth weight less
than 2,200 g, premature infants, infants with TPN, and
infants who have severe NH are prone to have no gut
excretion of tracer.1s*1g
In our institution, percutaneous liver needle biopsy is
usually performed to confirm BA before exploratory
laparotomy when there is no excretion of tracer on
hepatobiliary scintigraphy. Several portal areas should be
included in the specimen for proper diagnosis.*JO Although there are some histological similarities between
BA and NH, most investigators agree that the most
reliable criteria for the diagnosis of extrahepatic biliary
obstruction is ductal proliferation, portal fibrosis, and
canalicular bile stasis.21-25Recognition of the degree of
ductal proliferation requires quantification of the bile
duct to portal space ratio in the liver biopsy specimen. We
interpreted the results using Watkins definition that bile
duct to portal space ratio in normal infants is between 0.9
and 1.8.26 Bile duct proliferation of more than average
five per one portal area with portal fibrosis was interpreted as having BA, and average three to four ducts per
one portal area as having probable BA. No ductal
proliferation (average 1 to 2 ducts per one portal area)
with hepatocellular degeneration was interpreted as having NH or other causes of cholestasis, and less than 0.5
ducts per one portal area was interpreted as having
paucity of intraheaptic bile duct.
In this study, NBx was carried out in the majority of the
infants with no gut excretion of tracer in 24 hours on

ULTRASONOGRAPHIC

TRIANGULAR

CORD

1559

IN BA

Tc-99m-DISIDA hepatobiliary scintigraphy. Our diagnostic accuracy rate of 93% is similar to those of previous
reports.17J1.Z3Although NBx is considered to be a very
valuable test with a diagnostic accuracy of over 93%,17.1.23
NBx biopsy is invasive and can be misleading in certain
cases.17,21,22.24,25
If it is performed before 4 to 8 weeks of
age in infants who have BA, it may not show the
characteristic ductal proliferation or portal fibrosis24.25
necessitating a second biopsy as demonstrated in our two
false-negative cases. False-positive results can occur in
some cholestatic liver diseases such as bile plug syndrome, alpha-1-antitrypsin deficiency, and some neonatal
hepatitis.i7-*I We also experienced one false-positive case
of TPN-associated cholestasis that showed mild ductal
proliferation (3 to 4 ducts per one portal area). In these
cases, we have to evaluate infants who have cholestatic
jaundice in the context of clinical and laboratory data.
Since the introduction of the TC sign on US in the
diagnosis of BA by our institution, we found that it
seemed to be a simple, time saving, highly reliable, and
noninvasive tool in the diagnosis of BA from other causes
of cholestasis. The authors would like to propose a new
diagnostic strategy in the evaluation of infantile cholestasis. When the TC is visualized on US, prompt exploratory
laparotomy is mandatory without further investigations.

BILIARY ATRESIA

VS INTRAHEPATIC

CHOLESTASIS

1
Ultrasonography
Triangular A cord (+)
I
surgery

-k&i,,

cord (-)
I

Tc-99m-DISIDA

Excretion

into intestine

1
Conservative
\

Scan (after Phenobarbital

\
No excretion

into intestine

1
Liver Needle Biopsy

Rx
Neonatal

Rx)

H-----l\
hepatitis

Inconclusive
1
Optional

Biliary atresia
I
surgery

Fig 2. Algorithm
for workup
of infantile
cholestasis.
*Optional:
Minilaparotomy
or laparoscopy
is advised in older infants (more than
8 weeks)
but repeat ultrasonography,
liver biopsy, or hepatobiliary
scan are mandatory
in 2 weeks in younger
infants (less than 8 weeks).

When the TC is not visualized, hepatobiliary scintigraphy

is the next step. Excretion of tracer into the small bowel
actually rules out BA. Liver needle biopsy can be
reserved only for the infants without gut excretion of
tracer. We believe that a correct decision regarding the
need for surgery can be made in almost all cases of
infants who have cholestatic jaundice by this multidisciplinary approach (Fig 2).

REFERENCES
1, Kasai M, Kimura S. Asakura Y, et al: Surgical treatment of biliary
atresia. J Pediatr Surg 3:665-675,
1968
2. Karrer FM: Lilly JR. Stewart BA; et al: Biliary atresia registry,
1976 to 1989. J Pediatr Surg 25:1076-1081,
1990
3. Ohi R, Ibrahim M: Biliary atresia. Sem Pediatr Surg 1:115-124,
1995
4. Abramson
SJ, Treves S, Teele RL: The infant with possible biliary
atresia: Evaluation
by ultrasound
and nuclear medicine. Pediatr Radio1
12:1-5, 1982
5. Brun F, Gauthier
F, Doucher
D, et al: Ultrasound
findings in
biliary atresia in children. Ann Radio1 28:259-263,
198.5
6. Green D, Carroll BA: Ultrasonography
in the jaundiced infant: A
new approach. J Ultrasound
Med 5:323-329,
1986
7. Ikeda S, Sera Y, Akagi M: Serial ultrasonic
examination
to
differentiate
biliary atresia from neonatal hepatitis-special
reference to
change in size of the gallbladder.
Eur J Pediatr 148:396-400,
1989
8. Altman RP, Abramson
S: Potential errors in the diagnosis
and
surgical management
of neonatal jaundice. J Pediatr Surg 20:529-534,
1985
9. Kirks DR. Coleman RE, Filston HC, et al: An imaging approach to
persistent neonatal jaundice. Am J Roentgen01 142:461-465,
1984
10. Weinberger
E, Blumhagen
JD, Ode11 JM: Gall bladder contraction in biliary atresia. Am J Roentgen01 149:401-402.
1987
11. Choi SO, Park WH; Lee HJ. et al: Triangular
cord:
A
sonographic
finding applicable
in the diagnosis
of biliary atresia. J
Pediatr Surg 3 1:363-366,
1996
12. Choi SO, Park WH, Lee HJ: Ultrasonographic
Triangular
cord: The most definite finding for noninvasive
diagnosis of extrahepatic biliary atresia. Eur J Pediatr Surg [in press)
13. Balistreri WF: Neonatal cholestasis.
J Pediatr 106: 17 1-l 84, 1985
14. Majd M, Reba RC, Altman RP: Diagnostic
accuracy of hepatobiliary scintigraphy
with Tc-99m-PIPIDA
questioned.
Pediatrics
68:606607, 1981

15. Gerhold JP, Klingensmith

WC. Kuni CC: et al: Diagnosis
of
biliary
atresia with radionuclide
hepatobiliary
imaging.
Radiology
146:499-504,
1983
16. Manolaki
AG, Larcher VF. Mowat AP, et al: The prelaparotomy
diagnosis
of extrahepatic
biliary atresia. Arch Dis Child 58:591-594,
1983
17. Tolia V, Dubois
RS, Kagalwalla
A, et al: Comparison
of
radionuclear
scintigraphy
and liver biopsy in the evaluation
of neonatal
cholestasis. J Pediatr Gastroenterol
Nutr 5:30-34. 1986
18. Spivak WS, Sarkar S, Winter D, et al: Diagnostic
utility of
hepatobiliary
scintigraphy
with 99m-Tc-DISIDA
in neonatal cholestasis. J Pediatr 110:855-861,
1987
19. Park WH, Kim WS, Choi SO, et al: Comparison
of Tc-99mDISIDA hepatobiliary
scintigraphy
and percutaneous
needle biopsy in
the diagnosis of biliary atresia from intrahepatic
cholestasis.
Journal of
Korean Association
of Pediatric Surgeons (in press)
20. Desmet VJ: Congenital diseases of intrahepatic
ducts: Variations
on the theme ductal plate malformation.
Hepatology 16:1069-1083,
1992
21. Brough AJ, Bernstein J: Conjugated
hyperbilirubinemia
in early
infancy. Hum Path01 5~507-516, 1974
22. Altman RP, Stolar CJH: Pediatric hepatobiliary
disease. Surg
Clin North Am 65:1245-1267,
1985
23. Lai MW, Chang MH, Hsu SC, et al: Differential
diagnosis of
extrahepatic
biliary atresia from neonatal hepatitis: A prospective
study.
J Pediatr Gastroenterol
Nutr 18:121-127,
1994
24. Shah HA. Spivak W: Neonatal cholestasis.
New approaches
to
diagnostic
evaluation
and therapy. Pediatr Clin North Am 41:943-966.
1994
25. Park WH. Kim SP, Park KK, et al: Electron microscopic
study of
the liver with biliary atresia and neonatal hepatitis.
J Pediatr Surg
3 1:367-374,
1996
26. Watkins JB: Neonatal cholestasis:
Developmental
aspects and
current concepts. Sem Liver Dis 13:276-288,
1993