Volume 73, No.1: 2021 Siriraj Medical Journal
https://he02.tci-thaijo.org/index.php/sirirajmedj/index
32
Srisuwan et al.
Witsanee Srisuwan, M.D.*, Wison Laochareonsuk, M.D.*, Panu Wetwittayakhlang, M.D.**, Supika Kritsaneepaiboon,
M.D.***, Surasak Sangkhathat, M.D.*
*Department of Surgery, **Department of Internal Medicine and NKC Institute of Gastroenterology and Hepatology, ***Department of Radiology,
Faculty of Medicine, Prince of Songkla University, Hat Yai, Songkhla 90110, ailand.
Correlation of Transient Elastography and Biliary
Cirrhosis in Longterm Survivors of Biliary Atresia
ABSTRACT
Objective: is study aimed to use transient elastography (TE) to evaluate the correlation between liver stiness
measure (LSM) and functional status of native liver in longterm follow-up of pediatric patients with biliary atresia
(BA).
Methods: Twenty cases of BA who had undergone hepatic portoenterostomy and had good initial outcome (total
bilirubin < 2 mg/dL) were enrolled for a transient elastography. e LSMs derived from the study were analyzed
with clinical and radiological parameters and endoscopic ndings of esophageal varices.
Results: e median age at enrollment of the 20 cases was 8.4 years. Of the 20 cases, 15 were diagnosed as cirrhosis
by ultrasonography and 9 had esophageal varices detected by an endoscopy. Parameters that were signicantly
associated with LSM were history of cholangitis, splenomegaly, cirrhosis and esophageal varices. Signicantly higher
LSM was found to be correlated with hyperbilirubinemia, transaminitis, alkaline phosphatasemia, thrombocytopenia
and prolonged INR. On linear regression, LSM was signicantly correlated with pediatric end-stage liver disease
score at the r
2
of 0.32 and correlated with the aspartate transaminase to platelet ratio index at the r
2
of 0.70. e area
under the receiver operating characteristic curve that reected the performance of LSM in predicting esophageal
varices was 0.97. At the cut-o value of 10.2 kPa, the sensitivity and specicity of LSM in predicting esophageal
varices were 100% and 72.7%, respectively.
Conclusion: TE can be useful as a non-invasive, point-of-care evaluation of liver brosis in long term follow-up of
BA. A high LSM indicates surveillance for esophageal varices in these patients.
Keywords: Transient elastrography; biliary atresia; liver stiness measurement (Siriraj Med J 2021; 73: 32-37)
Corresponding author: Surasak Sangkhathat
E-mail: surasak.sa@psu.ac.th
Received 17 August 2020 Revised 8 October 2020 Accepted 9 October 2020
ORCID ID: http://orcid.org/0000-0003-3622-3233
http://dx.doi.org/10.33192/Smj.2021.05
INTRODUCTION
Biliary atresia (BA) is a progressive inammatory
cholangiopathy occurring during infancy period that
can lead to liver cirrhosis and death within 2 years of
life.
1
e incidence of BA is reported at approximately
1 in 10,000-20,000 live births with a higher incidence in
Asians.
2,3
e etiology of BA remains unclear although
some evidence suggests genetic involvement. Currently,
hepatic portoenterostomy (HPE, also known as Kasai’s
operation) is the initial surgical treatment of choice and
gives a good longterm outcome in 30%-50% of cases.
4
e procedure replaces a brotic extrahepatic bile duct
with an intestinal conduit interposed between the hepatic
portal plate and the jejunum.
Volume 73, No.1: 2021 Siriraj Medical Journal
https://he02.tci-thaijo.org/index.php/sirirajmedj/index
33
Original Article
SMJ
About 40% of BA cases who undergo HPE receive
no benet from the surgery. However, among the cases
who achieve biliary drainage aer the surgery, progression
of cirrhosis continues at varying rates. About half of the
patients who have long term survival aer HPE later
develop biliary cirrhosis, portal hypertension, esophageal
varices and/or liver failure.
4
Long term follow-up of the
liver functions and surveillance for esophageal varices and
cirrhosis are mandatory in BA cases. Various laboratory
tests are used as indices of liver function reserve in
pediatric chronic liver diseases. e aspartate transaminase
to platelet ratio index (APRI) has been proposed as an
indicator for liver brosis in pediatric patients with
chronic hepatitis C and also BA.
5,6
e Pediatric End-
stage Liver Disease (PELD) score has been widely used
to predict life-year benet from a liver transplantation.
7
Imaging evaluation of cirrhosis is usually performed
by an ultrasound study while evaluation of esophageal
varices needs an esophagoscopy. In our institute, an
esophagoscopy is usually considered when a child with
BA reaches 2-3 years of age or when there are clinical
signs of portal hypertension.
e recent development of transient elastometry
(TE) provides a non-invasive technique to evaluate the
extent of liver brosis.
8
TE emits a 50-MHz ultrasound
wave from its probe and measures the velocity of the
shear wave passing through the liver parenchyma. e
shear wave velocity is then converted into a liver stiness
measurement (LSM), which is expressed in kilopascals
(kPa).
9
TE has been widely used in various chronic liver
diseases in adults including chronic hepatitis and fatty
liver.
10,11
e scanning procedure can be performed
on an out-patient basis and usually takes less than 10
minutes. Studies involving BA patients have shown a
correlation between liver stiness and degree of liver
brosis in both before and aer HPE.
12-15
In 2011, a
study of liver stiness in post-HPE patients suggested
that the stiffness value was correlated well with the
presence of esophageal varices. A more recent study
of liver stiness measurements performed at 3 months
post-HPE signicantly predicted the risk of liver cirrhosis
and the need for a transplantation.
15
To our knowledge,
to date there have been only a few studies of TE in
longterm BA survivors and these have focused on the
correlation between the liver stiness and the risk of
developing portal hypertension or esophageal varices.
is study aimed to evaluate the correlation between
liver stiness measurement by Fibroscan and clinical/
laboratory proles of liver functions in cases of BA who
had good outcome aer HPE with more than 2 years of
follow-up period.
MATERIALS AND METHODS
Patients
is cross-sectional study included 20 cases of BA
who underwent HPE in our institute between 2001-
2015 and had good initial outcome dened as having
post-operative bilirubin level of less than 2 mg/dL. On
enrolment under informed consent of the parents or
other guardians, the patients were clinically examined for
growth anthropometry and signs of portal hypertension
by the attending surgeon. Additional evaluation consisted
of liver function study (total bilirubin, direct bilirubin,
total protein, albumin, alkaline phosphatase, aspartate
aminotransferase (AST) and alanine aminotransferase),
platelet count, hepatobiliary ultrasonography and upper
gastrointestinal endoscopy. Clinically signicant esophageal
varix was dened as the varix grade 2 or more according
to the Japanese Research Society for Portal Hypertension
denition (1991).
16
e aspartate transaminase to platelet
ratio index (APRI) was calculated as AST (×40) × 100/
platelet count (109/L). A Fibroscan was then performed
within 1 month of the laboratory and endoscopic tests.
e study was approved by the Human Research Ethics
Committee of the Faculty of Medicine, Prince of Songkla
University.
Transient elastography
e TE procedures were performed by trained internists
using a Fibroscan 502 Touch system (Echosens, Paris,
France). Two types of probe were used in this study: an S
probe for thoracic diameters < 45 cm and an M probe for
thoracic diameters > 45 cm. Scanning was performed in a
supine position with maximal abduction of the right arm.
For quality control, the median values of liver stiness
measurements from 10 validated measurements with
interquartile range (IQR) < 25% and reliability measured
as IQR/M < 25% were used.
17
All studies were performed
without sedation and under non-fasting conditions.
Statistical analysis
Data are presented as arithmetic means with ranges
when they have normal distribution and medians with
interquartile ranges otherwise. LSMs from the Fibroscan
studies were cross tabulated with various clinical and
investigative factors that are indicators of cirrhosis and
portal hypertension including the PELD scores. Association
analysis used Student t-test or Mann Whitney U test as
appropriate. Correlations between the LSMs and liver
cirrhosis as diagnosed by an ultrasonography and also
the presence of esophageal varices were analysed by
using a receiver operating characteristic (ROC) curve.
Correlations between LSMs and PELD scores were analysed
Volume 73, No.1: 2021 Siriraj Medical Journal
https://he02.tci-thaijo.org/index.php/sirirajmedj/index
34
Srisuwan et al.
by linear regression analysis. A P-value of less than 0.05
was considered statistically signicant.
RESULTS
e twenty BA patients enrolled in this study consisted
of 10 male and 10 female patients in an age range from
2.3 years to 21.0 years. e median age at enrolment
was 8.4 years (IQR 5.6 - 21.0 years). e mean age at
HPE in these patients was 73 days (range 30-119 days)
with 2 cases having their operation later than 90 days
of age. Twelve cases had ever had at least 1 episode of
cholangitis. ree patients had growth parameters of less
than the tenth percentile of the normal growth curve of
ai children. Ultrasound diagnosed liver cirrhosis in 15
cases, splenomegaly in 11 cases and presence of ascites
in 2 cases. Esophagogastroscopy detected esophageal/
gastric varices in 9 cases.
e median LSM was 12.4 kPa (IQR 6.0-32.4 kPa).
Clinical and radiologic parameters that were signicantly
associated with LSM were history of cholangitis,
splenomegaly, cirrhosis and presence of esophageal
varices (Table 1). When LSM was analysed with the
liver function test, signicant associations were found
between the stiness value and total bilirubin, direct
bilirubin, serum albumin, aspartate aminotransferase,
alanine aminotransferase and alkaline phosphatase
(Table 2). Higher LSM values were also signicantly
associated with thrombocytopenia and high prothrombin
time INR. On linear regression, LSM was signicantly
correlated with the PELD score at the p-value of 0.02 and
r
2
at 0.32. In addition, there was signicant correlation
between LSM and APRI at the p-value of < 0.01 and r
2
at 0.70.
To determine the diagnostic performance of LSM
and cirrhosis diagnosed by ultrasonography, an ROC
curve was plotted between sensitivity and 1-specicity to
predict esophagogastric varices by each LSM value. e
area under the ROC curve was 0.97 and the cut-o value
that best predicted the presence of esophageal varices was
at 10.2 kPa (sensitivity 100.0 % and specicity 72.7%)
(Fig 1). e area under the curve when analysed against
APRI in predicting esophageal varices was 0.99.
DISCUSSION
Gradual deterioration of liver function is a natural
history aer surgery in BA patients. Even in cases that
had achieved good bile flow after surgery, cirrhosis
and portal hypertension were eventually occurred.
Ascending cholangitis may accelerate the cirrhotic changes.
Evaluation of the severity of cirrhosis usually relies on
clinical evaluation and ultrasonography. In general, an
esophagoscopy is indicated when there are clinical or
radiological signs of portal hypertension. In this study,
we evaluated the values of transient elastography in BA
related cirrhosis by studying the association between LSM
and liver proles in good outcome BA cases who had
been followed more than 2 years. e study demonstrated
signicant associations between LSM and both APRI
and radiologic diagnosis of liver cirrhosis. In addition,
LSM was signicantly associated with other signs of
liver decompensation and portal hypertension including
hyperbilirubinemia and presence of esophageal varices.
In 2011, a study by Chongsrisawat and colleagues has
reported that both LSM and APRI could equally predict
the esophageal varices in BA cases with an acceptable
accuracy.
12
Consistent with that report, our data showed
good correlation between LSM and APRI and near
perfect performance of both parameters in predicting
signicant esophageal varices. e data suggested that
screening for esophageal varices during BA follow-up
can be performed in particular cases and Fibroscan
is the eective tool for case selection. Our study also
suggests that an LSM of 10.2 kPa is a practical cut-o
value for variceal prediction, a value which is similar
to the value range of 9.7-12.5 kPa as suggested by the
previous studies.
12,18,19
One of limitations in this study was its cross-sectional
design which meant that all variables analysed were
measured at a single time-point during the wide range
follow-up period. Chronological changes over time of
the LSM might be more useful in following the rate of
deterioration of the native liver function. e timing of
scanning has varied among earlier reports. One study in
pre-operative BA cases suggested that LSM was superior
to APRI in predicting histologically conrmed high-grade
brosis (F4 by the Metavir scoring system).
14
Another
study performed at 3 months post HPE demonstrated
good correlations between high LSMs and liver-related
events such as ascites, variceal bleeding or death during
the follow-up period which suggested the earlier need
of transplantation.
13
CONCLUSION
In conclusion, our study performed transient
elastography during long term follow-up in BA patients
who achieved good biliary drainage aer HPE. As in other
studies, we found correlations between LSMs and other
clinical/radiological evidence of portal hypertension.
Using a cut-o of 10.2 kPa, TE predicted esophageal
varices with high sensitivity.
Volume 73, No.1: 2021 Siriraj Medical Journal
https://he02.tci-thaijo.org/index.php/sirirajmedj/index
35
Original Article
SMJ
TABLE 1. Association between liver stiness measure (LSM) by a transient electrography and clinical signs,
esophagogastroscopic detection of varices and ultrasound ndings.
Cases (%) LSM (kPa) P-value*
History of cholangitis
No 8 (40%) 8.12 0.03
Yes 12 (60%) 30.33
Weight or height <10
th
percentile
No 17 (85%) 21.75 0.31
Yes 3 (15%) 19.77
Splenomegaly by examination
No 12 (60%) 8.22 < 0.01
Yes 8 (40%) 41.29
Esophageal varices
Absent 11 (55%) 7.66 < 0.01
Present 9 (45%) 38.30
Cirrhosis by ultrasound
Absent 5 (25%) 5.18 < 0.01
Present 15 (75%) 26.87
Splenomegaly by ultrasound
Absent 9 (45%) 6.61 < 0.01
Present 11 (55%) 33.59
Ascites by ultrasound
Absent 18 (90%) 21.35 0.61
Present 2 (10%) 22.35
Intrahepatic duct dilatation
Absent 16 (80%) 17.76 0.32
Present 4 (20%) 36.22
Fatty change
Absent 19 (95%) 22.31 -
Present 1 (5%) 5.10
*Mann Whitney U test
Volume 73, No.1: 2021 Siriraj Medical Journal
https://he02.tci-thaijo.org/index.php/sirirajmedj/index
36
Srisuwan et al.
TABLE 2. Association between liver stiness measure (LSM) by a transient electrography and liver function proles,
prothrombin time-INR and platelet count.
Cases (%) LSM (kPa) P-value*
Total bilirubin (mg/dL)
TB < 2.0 16 12.45 < 0.01
TB > 2.0 4 57.45
Direct bilirubin (mg/dL)
DB < 2.0 17 16.13 0.02
DB > 2.0 3 51.60
Serum albumin (g/dL)
Alb < 3.5 2 75.00 0.02
Alb > 3.5 18 15.50
Aspartate aminotransferase (U/L)
AST < 35 15 7.30 0.04
AST > 35 5 26.17
Alanine aminotransferase (U/L)
ALT < 35 8 7.15 < 0.01
ALT > 35 12 30.98
Akaline phosphatase (U/L)
ALP < 200 3 5.77 0.04
ALP > 200 17 24.21
Prothrombin time-INR**
INR < 1.2 10 13.41 0.04
INR > 1.2 9 32.06
Platelet count (x 10
9
cells/L)**
< 150 9 27.8 0.047
> 150 9 9.96
*Mann Whitney U test, **presence of missing values
Fig 1. Diagnostic performance of (A) liver stiness measure and (B) aspartate transaminase to platelet ratio index in predicting esophageal
varices in biliary atresia patients.
Volume 73, No.1: 2021 Siriraj Medical Journal
https://he02.tci-thaijo.org/index.php/sirirajmedj/index
37
Original Article
SMJ
ACKNOWLEDGEMENTS
e authors thank the nurses in the NKC Institute
of Gastroenterology and Hepatology, Songklanagarind
Hospital for their cooperation in performing the transient
elastography. Dave Patterson edited the English Language
in the manuscript.
REFERENCES
1. Hartley JL, Davenport M, Kelly DA. Biliary atresia. Lancet
2009;374:1704-13.
2. Lee KJ, Kim JW, Moon JS, Ko JS. Epidemiology of Biliary
Atresia in Korea. J Korean Med Sci 2017;32:656-60.
3. Sanchez-Valle A, Kassira N, Varela VC, Radu SC, Paidas C,
Kirby RS. Biliary Atresia: Epidemiology, Genetics, Clinical
Update, and Public Health Perspective. Adv Pediatr 2017;64:285-
305.
4. Sangkhathat S, Patrapinyokul S, Tadtayathikom K, Osatakul
S. Peri-operative factors predicting the outcome of hepatic
porto-enterostomy in infants with biliary atresia. J Med Assoc
ai 2003;86:224-31.
5. Wai CT, Greenson JK, Fontana RJ, Kalbeisch JD, Marrero JA,
Conjeevaram HS, et al. A simple noninvasive index can predict
both signicant brosis and cirrhosis in patients with chronic
hepatitis C. Hepatology 2003;38:518-26.
6. Kim SY, Seok JY, Han SJ, Koh H. Assessment of liver brosis
and cirrhosis by aspartate aminotransferase-to-platelet ratio
index in children with biliary atresia. J Pediatr Gastroenterol
Nutr 2010;51:198-202.
7. Wiesner RH, McDiarmid SV, Kamath PS, Edwards EB, Malinchoc
M, Kremers WK, et al. MELD and PELD: application of survival
models to liver allocation. Liver Transpl 2001;7:567-80.
8. Rockey DC, Bissell DM. Noninvasive measures of liver brosis.
Hepatology 2006;43:S113-20.
9. Sandrin L, Fourquet B, Hasquenoph JM, Yon S, Fournier C,
Mal F, et al. Transient elastography: a new noninvasive method
for assessment of hepatic fibrosis. Ultrasound Med Biol
2003;29:1705-13.
10. Yoneda M, Yoneda M, Fujita K, Inamori M, Tamano M, Hiriishi
H, et al. Transient elastography in patients with non-alcoholic
fatty liver disease (NAFLD). Gut 2007;56:1330-1.
11. Castera L, Le Bail B, Roudot-oraval F, Bernard PH, Foucher
J, Merrouche W, et al. Early detection in routine clinical
practice of cirrhosis and oesophageal varices in chronic hepatitis
C: comparison of transient elastography (FibroScan) with standard
laboratory tests and non-invasive scores. J Hepatol 2009;50:
59-68.
12. Chongsrisawat V, Vejapipat P, Siripon N, Poovorawan Y.
Transient elastography for predicting esophageal/gastric varices
in children with biliary atresia. BMC Gastroenterol 2011;11:41.
13. Hahn SM, Kim S, Park KI, Han SJ, Koh H. Clinical benet
of liver stiffness measurement at 3 months after Kasai
hepatoportoenterostomy to predict the liver related events in
biliary atresia. PLoS One 2013;8:e80652.
14. Shen QL, Chen YJ, Wang ZM, Zhang TC, Pang WB, Shu J,
et al. Assessment of liver brosis by Fibroscan as compared to
liver biopsy in biliary atresia. World J Gastroenterol 2015;21:
6931-6.
15. Wu JF, Lee CS, Lin WH, Jeng YM, Chen HL, Ni YH, et al.
Transient elastography is useful in diagnosing biliary atresia
and predicting prognosis after hepatoportoenterostomy.
Hepatology 2018;68:616-24.
16. Idezuki Y. General rules for recording endoscopic ndings of
esophagogastric varices (1991). Japanese Society for Portal
Hypertension. World J Surg 1995;19:420-2.
17. Ziol M, Handra-Luca A, Kettaneh A, Christidis C, Mal F, Kazemi
F, et al. Noninvasive assessment of liver brosis by measurement
of stiness in patients with chronic hepatitis C. Hepatology
2005;41:48-54.
18. Chang HK, Park YJ, Koh H, Kim SM, Chung KS, Oh JT, et al.
Hepatic brosis scan for liver stiness score measurement: a
useful preendoscopic screening test for the detection of varices
in postoperative patients with biliary atresia. J Pediatr Gastroenterol
Nutr 2009;49:323-8.
19. Colecchia A, Festi D, di Biase AR. Noninvasive parameters
for predicting esophageal varices in children: their sequential
use provides the best accuracy. Gastroenterology. 2012;142:e32.
