Author Affiliations: Department of Ultrasound, Tongji Hospital, Tongji Medical College, Huazhong University of Science and Technology, Wuhan 430030, China (Jiang LP, Chen YC, Ding L, Liu XL, Li KY, Huang DZ and Zhang QP); Department of Ultrasound, the First Affiliated Hospital of Nanchang University, Nanchang 330006, China (Jiang LP and Zhou AY)

Corresponding Author: Yun-Chao Chen, MD, Department of Ultrasound, Tongji Hospital, Tongji Medical College, Huazhong University of Science and Technology, No.1095 Jiefang Road, Wuhan 430030, China (Tel: 86-27- 83665457; Email: chen_yunchao@yahoo.com.cn)

 

© 2013, Hepatobiliary Pancreat Dis Int. All rights reserved.

doi: 10.1016/S1499-3872(13)60065-X

 

 

Acknowledgements: The authors would like to thank Anand Rattansingh, Medical Imaging Department of Toronto General Hospital, for critically reviewing the manuscript and Prof. Yi-Zhen Weng, Department of Pediatric Surgery, Tongji Hospital, Tongji Medical College, Huazhong University of Science and Technology, for supporting surgical data.

Contributors: CYC conceived the topic and contributed the final manuscript. JLP and CYC performed the research, reviewed the literature, analyzed the data and prepared the initial manuscript. All authors contributed to the design and interpretation of the study and to further drafts. CYC is the guarantor.

Funding: This work was supported by a grant from the New Technology and Service Project of Tongji Hospital (2008057).

Ethical approval: This study was approved by Tongji Hospital Medical Ethical Review Board.

Competing interest: No benefits in any form have been received or will be received from a commercial party related directly or indirectly to the subject of this article.

 

 

BACKGROUND: It is a globally challenging problem to differentially diagnose biliary atresia (BA) from other disease processes causing infantile cholestatic jaundice. The high-frequency ultrasonography (HUS) yields much improved spatial resolution and therefore, might show better image in BA diagnostic examination. The present study was to evaluate the HUS on the diagnosis of BA in infants with jaundice.

 

METHODS: Fifty-one infants with neonatal jaundice were scanned with ultrasonography. Images included gallbladder, bile duct, right hepatic artery (RHA), portal vein (PV) and triangular cord (TC) sign, magnetic resonance imaging and additionally, laboratory tests and histopathology reports were assessed.

 

RESULTS:Twenty-three BA and 28 non-BA cases were con-firmed. The sensitivity, specificity, and accuracy of HUS were 91.3%, 92.9%, and 92.2%, respectively. All of these indices were significantly higher than those of conventional ultrasonography (P<0.01) and MR cholangiopancreatography (P<0.05). The HUS features, included a positive TC sign, an increased RHA diameter and RHA-diameter to portal-vein-diameter ratio (RHA/PV) and abnormal gallbladder, were important in the diagnosis of BA.

 

CONCLUSION:HUS provided better imaging of BA and should be considered as a primary modality in the differential diagnosis of infantile jaundice.

 

(Hepatobiliary Pancreat Dis Int 2013;12:415-422)

 

KEY WORDS: neonatal jaundice; biliary atresia; high-frequency ultrasonography

 

 

Biliary atresia (BA) requires urgent attention. If unrecognized, BA leads to liver failure. BA is relatively common in eastern Asia, with an incidence of around 1:5000 live births in Taiwan. BA is 1.4 times more common in males than that in females.^[1] BA is caused by perinatal inflammation originated from the biliary ducts. The inflammation induces progressive sclerosis and obstruction of the intrahepatic biliary tree.^[2] BA is categorized into 3 types:^[3] type 1 (about 5% of cases) has luminal patency down to the common bile duct (often associated with a proximal cystic element) and proximal cystic biliary duct; type 2 (about 2% of cases) has patency to the level of the common hepatic duct; and type 3 (>90% of cases) displays as a solid structure in the most proximal part of the extrahepatic biliary tract within the porta hepatis. The gallbladder is absent or atretic in 25% of cases.^[4] The clinical manifestations of BA include jaundice, pale stool, dark urine, coagulopathy, hepatosplenomegaly, and ascites. Up to 60% of children will achieve biliary drainage after Kasai portoenterostomy and will have serum bilirubin within the normal range in 6 months.^[3] Additionally, where non-transplantation surgery is successful, more than 80% of infants with BA will survive over 10 years with good quality of life.^[5] Unfortunately, BA is difficult to diagnose in the clinical settings, where common and easily treated causes of jaundice such as infantile hepatitis syndrome and breast milk jaundice are encountered. As such, the need for screening is recognized, and early referral for investigations to exclude BA is important.^{[1, 6]}

 

Numerous ultrasound features have been described in diagnosing BA. Relevant signs seen on an abdominal scan using high-frequency ultrasonography (HUS)^[7] include the triangular cord (TC) sign, absence of biliary dilation, an abnormal gallbladder and hypertrophy of the right hepatic artery (RHA). A normal gallbladder has a regular outline and a uniformly thin wall. However, patients with BA usually have abnormal gallbladders which show an irregular outline, irregularly thick walls, or sometimes a small lumen. Farrant et al^[8] reported that the shape and the wall of the gallbladder might be abnormal in infants with BA. As described by Park et al,^[9] the TC sign was defined as a triangular- or tubular-shaped echogenic density measuring 3 mm or greater, and located immediately cranial to the portal vein bifurcation. This was found on surgery and displayed as a fibrous plate at the porta hepatis. Hypertrophy of the RHA has also been reported in infants with BA.^[10-12] The present study was to evaluate the HUS in BA diagnosis by its prominent ability to identify these features and compare HUS with conventional ultrasonography (CUS) and the other imaging methods, such as magnetic resonance cholangiopancreatography (MRCP) in BA diagnosis.

 

 

Sixty-three patients initially enrolled in this study from May 2009 to January 2011 were infants referred to the Departments of Pediatric Medicine and Pediatric Surgery of our hospital with persistent jaundice of unknown causes. In 12 patients excluded from the study, 10 were either untreated or had no follow-up details, and 2 were diagnosed with choledochal cysts. In 51 infants included in this study, 36 were male and 15 female, and their age ranged from 1 to 6 months. All infants presented with persistent jaundice, dark urine, and pale stools shortly after birth. This study was approved by our Institutional Medical Ethical Review Board. Informed consent was obtained from their parents before the examination.

 

Other data included a detailed laboratory report about biochemical and metabolic workup, magnetic resonance imaging, surgical, and pathological findings after surgery or biopsy. In most cases, the diagnosis was confirmed histologically by laparotomy or liver biopsy. In the cases of non-BA, the diagnosis was confirmed with the resolution of symptoms or the results of liver biopsy with a follow-up of at least 6 months.

 

One doctor, who was blinded to the results of other investigations, performed the ultrasonography (US) examination using a 3.5 MHz curve transducer first (CUS) and then a 7-12 MHz HUS linear transducer (GE LOGIQ9, GE, Milwaukee, USA, and SEQUOIA512, Siemens, Erlangen, Germany). After the gallbladder length was measured, the infants were pacified by feeding before proceeding the other examinations; none of the infants were sedated. The TC sign was identified anterior to the bifurcation of the portal vein (PV) which was seen as a focal area of increased echogenicity. Observations on the gallbladder included the shape, the regularity and thickness of the wall, and the length. If the common bile duct was visible or an intrahepatic bile duct dilated, the diameter was recorded. The diameter of PV and RHA was measured at the level of the proximal portion of the main PV and the proximal end of the RHA which runs parallel to the right PV. Measurement was made by using an electronic caliper from the midportion of the anterior wall to the midportion of the posterior wall of PV and RHA more than twice and chose the largest ones. After measurement, all patients first underwent color Doppler US with medium flow velocity and medium wall filter and then pulse Doppler for spectral analysis of the RHA. The sampling cursor was placed in the middle and paralleled with the RHA, the sampling volume was about the width of the diameter of the RHA, and the insonation angle was adjusted below 60°. The measurements, which included the peak systolic velocity (PSV) and resistance index (RI) of the RHA, were taken after the optimal Doppler signal was recognized. Each measurement was recorded three times from three different cardiac cycles and the average was used for statistical analysis.

 

Statistical analysis was performed using software (SPSS for Windows, version 17.0; SPSS, Chicago, IL., USA). An unpaired t test was used to compare normally distributed continuous variables, such as hepatic size, lengths of the gallbladder, PSV, RI, and diameters of the PV and RHA. The Chi-square test was used to assess categorical data (presence or absence of TC sign, small gallbladder, abnormal gallbladder, and RHA enlargement). To determine the diagnostic performance of the length of the gallbladder, PSV of the RHA, RHA diameter and RHA/PV, receiver operating characteristic (ROC) analysis was performed. Cutoff values were obtained by balance between sensitivity and specificity, and the values higher than the cutoff values indicated a positive diagnosis of BA. Significant differences were defined as P values less than 0.05.

 

 

In the 51 infants enrolled in the study, 16 BA patients were proven by Kasai procedure and histopathology, 4 non-BA and 4 BA cases were confirmed by the results of liver biopsy, the other 24 non-BA and 3 BA cases were diagnosed by a series of clinical assessments and detailed follow-up. BA was diagnosed when the patient's condition was deteriorated, and non-BA was diagnosed if the patient's condition was improved with treatment. The mean age between BA (2.91±1.08 mon) and non-BA groups (2.89±1.15 mon) was not different significantly (P>0.05). In the non-BA group, 22 patients had infantile hepatitis syndrome, 3 breast milk jaundice, 2 fat storage diseases, and one progressive familial intrahepatic cholestasis (PFIC). Two BA patients were misdiagnosed because of the absence of TC sign at the porta hepatis, even though both of them had an enlarging RHA and one displayed an abnormal gallbladder. The detailed specific features of 23 BA cases are listed in Table 1.

 

Twenty-one patients showed TC sign, which was described as the fibrous plate on surgery (Table 1). Among the 16 BA patients who had Kasai procedure, 14 with TC sign and 1 false negative case without TC sign had a fibrous plate at the porta hepatis on surgery. However, in a false negative patient without TC sign, no obvious fibrous plate was seen at the time of surgery. The thickness of TC sign ranged from 2.3 to 11 mm (average 5.1). In 2 of 28 non-BA patients, a TC sign with the thickness of 4 mm was seen. The sensitivity, specificity and accuracy of TC sign in the diagnosis of BA were 91.3% (21/23), 92.9% (26/28) and 92.2% (47/51), respectively (Table 2).

 

The length of gallbladder was significantly shorter in the BA group compared with the non-BA group (1.78±0.92 vs 2.61±0.98 cm, P<0.01, Table 3). A gallbladder length of less than 1.6 cm was considered a cutoff value for a small gallbladder according to ROC analysis (area under the ROC curve, 0.76). This criterion gave a sensitivity of 52.2% (12/23), a specificity of 89.3% (25/28), and an accuracy of 72.5% (37/51) in BA diagnosis (Fig. 4A). In this study, a non-visible gallbladder and a gallbladder seen as a streak were classified as an abnormal gallbladder. A gallbladder with an irregular and thickened wall and a distortion was also considered as abnormal (even if the length or capacity was normal). Twenty-two BA and 3 non-BA cases displayed abnormal gallbladders, and the difference was statistically significant between the two groups (P<0.01, Table 1). The sensitivity, specificity, and accuracy for diagnosis BA of abnormal gallbladder were 95.7% (22/23), 89.3% (25/28), and 92.2% (47/51), respectively (Table 2).

 

The diameter of the RHA in the BA group was significantly longer than that in the non-BA group (0.24±0.04 vs 0.14±0.04 cm, P<0.01), but the diameter of the PV was not significantly different between the BA and non-BA groups (0.46±0.08 vs 0.45±0.08 cm, P>0.05). The RHA/PV in the BA group was significantly larger than that in the non-BA group (0.56±0.16 vs 0.34±0.09, P<0.001). The optimal cutoff value of the RHA was 0.19 cm (area under the ROC curve, 0.939), that indicates a sensitivity of 100% (23/23), a specificity of 89.3% (25/28), and an accuracy of 94.1% (48/51) for the diagnosis of BA (Fig. 4B). Whereas the optimal cutoff value of the RHA/PV was 0.34 (area under the ROC curve, 0.896), showing a sensitivity, a specificity, and an accuracy of 100% (23/23), 60.7% (17/28), and 78.4% (40/51), respectively in the diagnosis of BA (Fig. 4C). The PSV of the RHA in patients with BA was significantly higher than that in patients with non-BA (66.9±25.0 vs 44.3±25.5 cm/s, P<0.001), and the optimal cutoff value of 55.7 cm/s (area under the ROC curve, 0.776) indicated a sensitivity of 78.3% (18/23), a specificity of 85.7% (24/28), and an accuracy of 82.4% (42/51) for the diagnosis of BA (Fig. 4D). There was no significant difference of RI between patients with BA and those with non-BA (0.79±0.08 vs 0.77±0.07, P>0.05) (Tables 2 and 3).

 

Only 2 of 23 BA patients were examined with CUS. If a small or absent gallbladder was considered as a sign of BA, CUS would indicate a provisional diagnosis of BA with a sensitivity of 43.5% (10/23), a specificity of 78.6% (22/28), and an accuracy of 62.7% (32/51), which was devoid of confidence as the diagnosis given. CUS displayed 12 (52.2%, 12/23) gallbladders with abnormalities including 7 non-visible, 3 small, and 2 irregular ones, but the other 10 gallbladders with abnormal wall or shape displayed by HUS were not shown by CUS. CUS could not find the gallbladder in 7 infants, whereas, in these infants, HUS showed a small gallbladder (3 infants), a hyperechogenic streak structure or atrophic gallbladder (2), and invisible structure (2). There was a significant difference in the diagnosis of abnormal gallbladder between CUS and HUS (P<0.05, Tables 4, 5). CUS displayed only 2 TC signs and 5 RHA enlargements in the 23 BA patients, which were significantly different from those shown by HUS (P<0.001, Table 4).

 

Besides, 19 of the 23 BA patients and 4 of 28 non-BA patients were selected for MRCP, and among them, 11 BA and 3 non-BA patients were correctly diagnosed, showing a sensitivity of 57.9% (11/19), a specificity of 75.0% (3/4), and an accuracy of 60.9% (14/23) in the diagnosis of BA. There was a significant difference in the diagnosis of BA between HUS and MRCP (P<0.05, Table 5).

 

 

It is a global challenge to differentiate BA from other diseases causing infantile cholestatic jaundice. Early diagnosis is of critical importance because survival depends on prompt surgery, which is recommended within 100 days of birth.^[3] There are many reasons to suspect BA, but no single clinical feature, laboratory finding, or imaging has been found to show sufficient sensitivity and specificity to differentiate BA from other causes of infantile cholestatic jaundice. MRCP is not accurate because of the small bile duct and the low rate of bile excretion in neonates and infants with jaundice.^[13-15] Norton et al^[14] reported a sensitivity of 90% and a specificity of 77% in MRCP, in which MRCP also showed a sensitivity of 57.9%, a specificity of 75%, and an accuracy of 60.9% in the diagnosis of BA. On the other hand, neonates and infants were thought too small to undergo complete sedation for MRI.^[16] A positive result on SPECT is not specific for BA.^[17] Ultrasonography is a preferred tool for evaluation of hepatobiliary tree, and has several advantages over other modalities: non-ionizing, noninvasive and convenient,^[4] but general abdominal ultrasound with a frequency less than 5 MHz is not applicable to evaluate the hepatobiliary tree in infants and neonates. With the updating of ultrasonic apparatus, especially the application of the high-frequency detecting probe in the 7 to 12 MHz range, HUS has been considered a superior method for differentiating non-BA from BA, based on the fact that it could display many specific features of BA such as TC sign, abnormal gallbladder, RHA enlargement, and so on even though it also could not adequately show bile duct dilatation.^[18-21] The data of the present study showed that HUS is superior to CUS in detecting TC sign, abnormal gallbladder, gallbladder length, RHA diameter, RHA/PV, and PSV. All of these indices were relatively accurate in differentiating BA from non-BA. Take TC sign for example, the diagnostic sensitivity, specificity and accuracy were 91.3%, 92.9% and 92.2%, respectively. Compared with HUS, CUS showed a sensitivity of 43.5%, a specificity of 78.6%, and an accuracy of 62.9%, but MRCP demonstrated a sensitivity of 57.9%, a specificity of 75.0%, and an accuracy of 60.7%.

 

TC sign is an abnormal hyperechogenic triangular area in the porta hepatis. Since reported by Choi et al in 1996,^[22] TC sign has been recognized as a specific marker for the diagnosis of BA, typically type 3.^[20] Park et al^[23] found that TC sign had a sensitivity of 84% and a specificity of 98% in a series of 79 infants. Imanieh et al^[24] demonstrated that TC sign had a sensitivity of 70%, a specificity of 95.8%, and an accuracy of 91.3% in detecting BA. When used as the only criterion for the diagnosis of BA,^{[7, 19, 25]} TC sign had a sensitivity of 49% to 73% and a specificity of 100% in the diagnosis of BA. Hence, TC sign should have a much higher specificity but a relative low sensitivity in the diagnosis of BA. However, our data showed that both sensitivity and specificity are relatively high in the diagnosis of BA.

 

Identification of hyperechogenic liver hilum, known as TC sign, is a special finding in almost all cases, but is operator-dependent and difficult to achieve.^[25] A small TC sign, an enlarged liver and sonographer's experience all affect TC sign image,^{[15, 25, 26]} sometimes an uncooperative infant also causes a false negative diagnosis. The factors above might explain the relatively low sensitivity in some studies. To improve the sensitivity of TC sign in the diagnosis of BA, adequate training of the ultrasound team involved in examining children is required. In the present study, two false negative cases did not show TC sign. One case showed a fibrous plate of 4 mm thickness at the porta hepatis on surgery. However, no obvious fibrous plate was seen at surgery in another BA case. This implied that TC sign was missed by HUS.

 

Although HUS increased the diagnostic sensitivity, there are still some false negative cases. False positive might lead to unnecessary laparotomy or liver biopsy. Our study indicated that TC sign has a relatively high rate of diagnostic accuracy and thus making other diagnostic tests unnecessary.^[25] Unfortunately, there are reports on false positive findings of TC sign. Visrutaratna et al^[27] found that false positive TC sign may be due to intrahepatic cholestasis in patients with periportal edema or thickening. Another possibility is hepatic fibrosis with severe periportal fibrosis. Neonatal hepatitis advanced to cirrhosis with periportal fibrosis also cause TC sign. In this study, two non-BA patients showed TC sign. One patient suffered from infantile hepatitis with a 1.0×0.4 cm hyperechogenic area at the hepatic hilum, which disappeared after treatment. We assumed that the hyperechogenic area might be periportal edema. Another patient with several episodes of jaundice did not show TC sign at age of 47 days, but a 1.3×0.4 cm TC sign was found one and half month later when the infant was readmitted. This infant was diagnosed with progressive familial intrahepatic cholestasis (PFIC) type 3. PFIC^{[28, 29]} is not a benign cholestatic jaundice but an autosomal recessive inherited disease, and is characterized by impaired bile acid secretion resulting in cholestasis, fibrosis, and cirrhosis leading to progressive liver failure and death in childhood. PFIC and BA have many similarities, particularly with pathologic progress and therapeutic approaches, and are impossible to differentiate between the two on imaging, such as with US. TC sign might also be a diagnostic feature of PFIC. Overall, there was greater accuracy of TC sign in the diagnosis of BA. However, physicians should be aware of potential false negative and positive cases.

 

Abnormal gallbladder findings, such as no gallbladder, gallbladder with irregular wall or abnormal shape, are other specific US markers in the diagnosis of BA. Because the gallbladder is absent or atretic in 25% of cases of BA,^[18] a small or an absent gallbladder is always regarded as a characteristic of BA. However, the appearance of the gallbladder is influenced by certain factors, such as meal ingestion and therefore, the length of the gallbladder has limitations in the diagnosis of BA. A small gallbladder might be seen both in BA and in neonatal hepatitis.^[30] Takamizawa et al^[21] reported when the length of the gallbladder was less than 1.5 cm, which was regarded as abnormal, the sensitivity was just 77% and the specificity was 73% in the diagnosis of BA. We found that if the cutoff value was defined 1.6 cm, the diagnostic sensitivity was 52.2%, the specificity 89.3%, and the overall accuracy 72.5%.

 

Using high frequencies in the 7 to 12 MHz range in HUS enables detailed, high resolution imaging. HUS can detect the gallbladder with irregular wall thickeness or an abnormal shape in the diagnosis of BA.^[19] In our study, 5 gallbladders which were not detected with CUS were shown as atrophic irregular small gallbladders or hyperechogenic streaks in the gallbladder fossa on HUS. Tan Kendrick et al^[31] named the "gallbladder ghost triad" as gallbladder length less than 1.9 cm, lack of smooth or complete echogenic mucosal lining with an indistinct wall and irregular or lobular contour. They reported that the gallbladder ghost triad in diagnosing BA had a sensitivity of 97% and a specificity of 100%. Farrant et al^[8] reported that an absent gallbladder or one with an irregular wall or abnormal shape had a sensitivity, specificity, and accuracy of 90%, 92.4%, and 91.9%, respectively, in the diagnosis of BA. Using the criteria proposed by Farrant et al, another group^[7] obtained an overall diagnostic accuracy of 95%. In the current study the sensitivity, specificity, and accuracy was 95.7% (22/23), 89.3% (25/28), and 92.2% (49/51), respectively, which were comparable to other studies. An atrophic gallbladder with an irregular wall is the diagnostic sign of BA, whereas a gallbladder that is simply small or contracted will have a smooth wall. This difference can be easily detected with HUS compared with CUS. These comparisons implied that HUS is more reliable in making a diagnosis of BA.

 

Hyperplastic and hypertrophic changes in the branches of the intrahepatic portion of the hepatic artery near the portal branches in patients with BA were initially described by Stowens,^[32] and reproduced elsewhere.^{[10,?11,?19,?33]} Although the definitive pathogenesis of RHA enlargement remains uncertain, it has been speculated that this may be due to compensatory change to improve the blood supply in the hepatobiliary tree, a secondary change due to liver cirrhosis, or an essential vascular formation.^{[10, 11]} Lee et al^[19] suggested that the presence of hepatic subcapsular flow was useful for differentiating BA from other causes of neonatal jaundice with sensitivity, specificity, positive and negative predictive values of 100%, 86%, 85%, and 100%, respectively, on the basis of consensus reading. Kim et al^[10] suggested that optimum cutoff values for diagnosis of BA were 1.6 mm (sensitivity, 92%; specificity, 87%; accuracy, 89%) for RHA and 0.45 for RHA/PV ratio (sensitivity, 76%; specificity, 79%; accuracy, 78%). Our study showed measurements of RHA>0.19 cm and RHA/PV>0.34 as the best cutoff value in the diagnosis of BA, with a sensitivity of 100%, a specificity of 89.3%, an accuracy of 94.1% and a sensitivity of 100%, a specificity of 61%, and an accuracy of 78.4%, respectively. Although the current study did not evaluate hepatic subcapsular flow, we found that the PSV of the RHA was higher in BA than in non-BA patients, and the optimal cutoff value of 55.7 cm/s had a sensitivity, a specificity and an accuracy of 78.3%, 85.7% and 82.4%, respectively. The RHA/PV ratio and PSV could potentially be used to differentiate BA from non-BA but these displayed a relative higher SD and lower accuracy than RHA alone, suggesting that the latter is better in the diagnosis of BA.

 

This study showed promising potential for the diagnosis of BA using HUS. However, the data discussed were not isolated. Other indices, such as an enlarged liver, increased liver parenchymal echogenicity or coarse echotexture of liver, splenomegaly, and gallbladder contractility were also evaluated in the study. These features were also helpful in the diagnosis of BA.

 

Additionally, the current study potentially under- estimated the overall sensitivity. For example, in the 2 false negative cases mentioned, although neither of them showed TC sign, one displayed a RHA diameter of 0.25 cm and a hyperechogenic and thickness of hepatic Glisson's capsule, and the other displayed a RHA of 0.29 cm with an abnormal gallbladder, features suggestive of a diagnosis of BA. If these features were all considered, there would be a higher sensitivity to diagnose BA.

 

Finally, there are several limitations in the current study. US examination was performed by a single sono-grapher and it is well known that operator experience is a critical component in the detection and interpretation of US findings. This study comprised a small number of cases (51 infants), and not all cases were diagnosed histopathologically. Therefore, a multicenter, multi-operator and larger sample size would potentially standardize the diagnostic process in clinical practice.

 

In conclusion, BA can be diagnosed by HUS using the specific features of TC sign with a sensitivity, specificity and accuracy of 91.3%, 92.9% and 92.2%, respectively. If all the HUS characteristics are considered, it would improve the overall sensitivity, specificity and accuracy in the diagnosis of BA. Since HUS is superior to CUS or MRCP, it should be a primary modality in evaluating infantile jaundice.

 

 

1 Hsiao CH, Chang MH, Chen HL, Lee HC, Wu TC, Lin CC, et al. Universal screening for biliary atresia using an infant stool color card in Taiwan. Hepatology 2008;47:1233-1240. PMID: 18306391

2 Cauduro SM. Extrahepatic biliary atresia: diagnostic methods. J Pediatr (Rio J) 2003;79:107-114. PMID: 14502330

3 Hartley JL, Davenport M, Kelly DA. Biliary atresia. Lancet 2009;374:1704-1713. PMID: 19914515

4 Goldman M, Pranikoff T. Biliary disease in children. Curr Gastroenterol Rep 2011;13:193-201. PMID: 21225381

5 Kuroda T, Saeki M, Morikawa N, Fuchimoto Y. Biliary atresia and pregnancy: puberty may be an important point for predicting the outcome. J Pediatr Surg 2005;40:1852-1855. PMID: 16338304

6 Powell JE, Keffler S, Kelly DA, Green A. Population screening for neonatal liver disease: potential for a community-based programme. J Med Screen 2003;10:112-116. PMID: 14561261

7 Humphrey TM, Stringer MD. Biliary atresia: US diagnosis. Radiology 2007;244:845-851. PMID: 17709832

8 Farrant P, Meire HB, Mieli-Vergani G. Ultrasound features of the gall bladder in infants presenting with conjugated hyperbilirubinaemia. Br J Radiol 2000;73:1154-1158. PMID: 11144791

9 Park WH, Choi SO, Lee HJ. Technical innovation for noninvasive and early diagnosis of biliary atresia: the ultraso-nographic "triangular cord" sign. J Hepatobiliary Pancreat Surg 2001;8:337-341. PMID: 11521178

10 Kim WS, Cheon JE, Youn BJ, Yoo SY, Kim WY, Kim IO, et al. Hepatic arterial diameter measured with US: adjunct for US diagnosis of biliary atresia. Radiology 2007;245:549-555. PMID: 17890351

11 dos Santos JL, da Silveira TR, da Silva VD, Cerski CT, Wagner MB. Medial thickening of hepatic artery branches in biliary atresia. A morphometric study. J Pediatr Surg 2005;40:637-642. PMID: 15852270

12 Ho CW, Shioda K, Shirasaki K, Takahashi S, Tokimatsu S, Maeda K. The pathogenesis of biliary atresia: a morphological study of the hepatobiliary system and the hepatic artery. J Pediatr Gastroenterol Nutr 1993;16:53-60. PMID: 8433241

13 Kim MJ, Park YN, Han SJ, Yoon CS, Yoo HS, Hwang EH, et al. Biliary atresia in neonates and infants: triangular area of high signal intensity in the porta hepatis at T2-weighted MR cholangiography with US and histopathologic correlation. Radiology 2000;215:395-401. PMID: 10796915

14 Norton KI, Glass RB, Kogan D, Lee JS, Emre S, Shneider BL. MR cholangiography in the evaluation of neonatal cholestasis: initial results. Radiology 2002;222:687-691. PMID: 00000

15 Takaya J, Nakano S, Imai Y, Fujii Y, Kaneko K. Usefulness of magnetic resonance cholangiopancreatography in biliary structures in infants: a four-case report. Eur J Pediatr 2007; 166:211-214. PMID: 16960695

16 Azuma T, Nakamura T, Nakahira M, Harumoto K, Nakaoka T, Moriuchi T. Pre-operative ultrasonographic diagnosis of biliary atresia--with reference to the presence or absence of the extrahepatic bile duct. Pediatr Surg Int 2003;19:475-477. PMID: 12750934

17 Gilmour SM, Hershkop M, Reifen R, Gilday D, Roberts EA. Outcome of hepatobiliary scanning in neonatal hepatitis syndrome. J Nucl Med 1997;38:1279-1282. PMID: 9255166

18 Rozel C, Garel L, Rypens F, Viremouneix L, Lapierre C, Décarie JC, et al. Imaging of biliary disorders in children. Pediatr Radiol 2011;41:208-220. PMID: 20865413

19 Lee MS, Kim MJ, Lee MJ, Yoon CS, Han SJ, Oh JT, et al. Biliary atresia: color doppler US findings in neonates and infants. Radiology 2009;252:282-289. PMID: 19561262

20 Li SX, Zhang Y, Sun M, Shi B, Xu ZY, Huang Y, et al. Ultrasonic diagnosis of biliary atresia: a retrospective analysis of 20 patients. World J Gastroenterol 2008;14:3579-3582. PMID: 18567090

21 Takamizawa S, Zaima A, Muraji T, Kanegawa K, Akasaka Y, Satoh S, et al. Can biliary atresia be diagnosed by ultrasonography alone? J Pediatr Surg 2007;42:2093-2096. PMID: 18082715

22 Choi SO, Park WH, Lee HJ, Woo SK. 'Triangular cord': a sonographic finding applicable in the diagnosis of biliary atresia. J Pediatr Surg 1996;31:363-366. PMID: 8708904

23 Park WH, Choi SO, Lee HJ. The ultrasonographic 'triangular cord' coupled with gallbladder images in the diagnostic prediction of biliary atresia from infantile intrahepatic cholestasis. J Pediatr Surg 1999;34:1706-1710. PMID: 10591576

24 Imanieh MH, Dehghani SM, Bagheri MH, Emad V, Haghighat M, Zahmatkeshan M, et al. Triangular cord sign in detection of biliary atresia: is it a valuable sign? Dig Dis Sci 2010;55:172-175. PMID: 19229615

25 Roquete ML, Ferreira AR, Fagundes ED, Castro LP, Silva RA, Penna FJ. Accuracy of echogenic periportal enlargement image in ultrasonographic exams and histopathology in differential diagnosis of biliary atresia. J Pediatr (Rio J) 2008; 84:331-336. PMID: 18688554

26 Tan Kendrick AP, Phua KB, Ooi BC, Subramaniam R, Tan CE, Goh AS. Making the diagnosis of biliary atresia using the triangular cord sign and gallbladder length. Pediatr Radiol 2000;30:69-73. PMID: 10663514

27 Visrutaratna P, Wongsawasdi L, Lerttumnongtum P, Sing-havejsakul J, Kattipattanapong V, Ukarapol N. Triangular cord sign and ultrasound features of the gall bladder in infants with biliary atresia. Australas Radiol 2003;47:252-256. PMID: 12890244

28 Shneider BL. Progressive intrahepatic cholestasis: mechanisms, diagnosis and therapy. Pediatr Transplant 2004;8:609-612. PMID: 15598335

29 Clifton MS, Romero R, Ricketts RR. Button cholecystostomy for management of progressive familial intrahepatic cholestasis syndromes. J Pediatr Surg 2011;46:304-307. PMID: 21292078

30 Kirks DR, Coleman RE, Filston HC, Rosenberg ER, Merten DF. An imaging approach to persistent neonatal jaundice. AJR Am J Roentgenol 1984;142:461-465. PMID: 6607624

31 Tan Kendrick AP, Phua KB, Ooi BC, Tan CE. Biliary atresia: making the diagnosis by the gallbladder ghost triad. Pediatr Radiol 2003;33:311-315. PMID: 12695863

32 Stowens D. Congenital Biliary Atresia. Ann N Y Acad Sci 1963;111:337-357. PMID: 14085860

33 Uflacker R, Pariente DM. Angiographic findings in biliary atresia. Cardiovasc Intervent Radiol 2004;27:486-490. PMID: 15383852