Latest diagnostic performance of different ultrasonic features for biliary atresia

Abstract

Background

Early diagnosis of biliary atresia (BA) is an important clinical challenge.

Purpose

To summarize the latest diagnostic performance of different ultrasonic (US) features for BA.

Material and Methods

MeSH terms “biliary atresia” and “ultrasonography” and related hyponyms were used to search PubMed, EMBASE, and the Cochrane Central Register of Controlled Trials. Eligible articles were included and data were retrieved. The methodologic quality was assessed by version 2 of the Quality Assessment of Diagnostic Accuracy Studies tool. Estimated sensitivity and specificity of each US feature were calculated by Stata 14.0.

Results

Fifty eligible studies on 5622 patients were included. Respective summary sensitivity and specificity were 77% (95% CI=69–84) and 98% (95% CI=96–99) for triangular cord sign (TCS) in 32 studies, 86% (95% CI=78–92) and 86% (95% CI=72–94) for shear wave elastography (SWE) in seven studies, 75% (95% CI=65–83) and 92% (95% CI=86–95) for gallbladder and biliary system abnormality (GBA) in 25 studies, and 81% (95% CI=69–90) and 79% (95% CI=67–87) for hepatic artery (HA) enlargement in seven studies. The overall US features from 11 studies yielded a summary sensitivity of 84% (95% CI=72–92) and specificity of 86% (95% CI=77–92).

Conclusion

TCS and GBA were the two most widely accepted US features currently used for differential diagnosis of BA. The newly developed SWE was an objective and convenient method with good diagnostic performance. HA enlargement can be used as an auxiliary sign.

Keywords

Biliary atresia ultrasonics diagnostic performance triangular cord sign shear wave elastography elasticity imaging techniques

Introduction

Biliary atresia (BA) is a disease characterized by progressive inflammation and fibrosis of the intrahepatic and extrahepatic bile ducts with unknown etiology, which can lead to cholestasis, progressive liver fibrosis, and even cirrhosis, and endangers the lives of children (1). It is one of the common serious hepatobiliary diseases in infancy (2). If it is not treated in time, biliary cirrhosis, portal hypertension, and liver failure will occur.

Early diagnosis and intervention of BA are critical to the prognosis of the disease. The current auxiliary examination used for this disease include ultrasound, hepatobiliary scintigraphy, magnetic resonance (MR) cholangiography and liver biopsy (3). Among these methods, ultrasound is recommended as the initial screening method because it is affordable, convenient, and radiation-free (4). Previous studies have found that triangular cord sign (TCS), gallbladder and biliary system abnormality (GBA), hepatic artery (HA) enlargement, and the newly developed liver shear wave elastography (SWE) provide good reference for the diagnosis of BA under ultrasonic (US) examination. Some meta-analyses on the diagnostic performance of ultrasound for BA have been carried out in the past, but in recent years, there have been many new studies exploring the diagnostic performance of different US features for BA; there is no new research about them in meta-analyses. In addition, the newly developed SWE has shown good diagnostic performance for BA, and there is currently no research to systematically review. Therefore, the aim of the present study was to conduct a meta-analysis of the latest diagnostic performance of different US features and SWE in diagnosing BA, and to evaluate their roles in identifying BA in infantile cholestatic patients.

Material and Methods

The study was a systematic review and meta-analysis, and institutional review board approval was not required.

Search strategy

MeSH terms “biliary atresia” and “ultrasonography” and related hyponyms were used to search PubMed, EMBASE, and the Cochrane Central Register of Controlled Trials (CENTRAL). The search was updated to March 2021. The search strategy in PubMed was as follows: (sensitiv*[Title/Abstract] OR sensitivity and specificity[MeSH Terms] OR (predictive[Title/Abstract] AND value*[Title/Abstract]) OR predictive value of tests[MeSH Term] OR accuracy*[Title/Abstract]) AND (((((((((((((((((((((((((((“Ultrasonography"[Mesh]) OR (Diagnostic Ultrasound)) OR (Diagnostic Ultrasounds)) OR (Ultrasound, Diagnostic)) OR (Ultrasounds, Diagnostic)) OR (Ultrasound Imaging)) OR (Imaging, Ultrasound)) OR (Imagings, Ultrasound)) OR (Echotomography)) OR (Ultrasonic Imaging)) OR (Imaging, Ultrasonic)) OR (Sonography, Medical)) OR (Medical Sonography)) OR (Ultrasonographic Imaging)) OR (Imaging, Ultrasonographic)) OR (Imagings, Ultrasonographic)) OR (Ultrasonographic Imagings)) OR (Echography)) OR (Diagnosis, Ultrasonic)) OR (Diagnoses, Ultrasonic)) OR (Ultrasonic Diagnoses)) OR (Ultrasonic Diagnosis)) OR (Echotomography, Computer)) OR (Computer Echotomography)) OR (Tomography, Ultrasonic)) OR (Ultrasonic Tomography)) AND (((((((((“Biliary Atresia"[Mesh]) OR (Atresia, Biliary)) OR (Intrahepatic Biliary Atresia)) OR (Biliary Atresia, Intrahepatic)) OR (Biliary Atresia, Extrahepatic)) OR (Atresia, Extrahepatic Biliary)) OR (Extrahepatic Biliary Atresia)) OR (Idiopathic Extrahepatic Biliary Atresia)) OR (Familial Extrahepatic Biliary Atresia))).

Study eligibility

First, duplicate articles were deleted by literature manager Endnote. An article was then considered potentially eligible if the abstract described the diagnostic efficiency of US features in the diagnosis of BA. Thereafter, full texts were obtained for further evaluation.

The inclusion criteria included the following: (i) explicit criteria defining US features; (ii) reference standard confirming BA, including liver biopsy, intraoperative cholangiography (IOC), MR cholangiography, 99mTc DISIDA imaging, percutaneous transhepatic, or endoscopic retrograde cholangiography; and (iii) sufficient data to extract the diagnostic performance of US features. The exclusion criteria included the following: (i) case reports, case series with a sample size of <10 patients, editorials, comments, letters, review articles, animal experiment studies, and conference proceedings; (ii) studies with insufficient data on the diagnostic performance of US features in patients with BA; and (iii) studies with overlapping patients and data.

Two investigators independently assessed the eligibility of each article, while the third investigator determined the disagreement.

Data extraction

We extracted the following data from the selected studies: (i) study characteristics—author(s), year of publication, affiliation, sample size, study design, and US features; (ii) diagnostic performance of US features in patients with BA, including the number of true-positive (TP), true-negative (TN), false-positive (FP) and false-negative (FN) findings; and (iii) detailed US features of BA. Two authors of this paper independently extracted the data. Any disagreement was resolved by discussion until consensus or by consulting a third author.

Quality assessment

The methodological quality of the studies was assessed by using version 2 of the Quality Assessment of Diagnostic Test Accuracy Studies (QUADAS-2) tool. Both reviewers scored the tool independently and any disagreement was resolved by discussion until consensus or by consulting a third author.

Statistical analysis

Meta-disc 1.4 software (Cochrane Colloquium, Barcelona, Spain) and Stata 14.0 (Stata Corporation, College Station, TX, USA) software were used for statistical analysis. We extracted or reconstructed 2 × 2 contingency tables for all the US features reported in the included studies. The TP, TN, FP, and FN were extracted or calculated from the original articles.

The overall sensitivity and specificity were calculated from the extracted data. A summary receiver operating characteristic (SROC) curve with a 95% confidence region and prediction region was also plotted to graphically present the results. The diagnostic odds ratio (DOR) and area under the curve (AUC) were calculated for evaluating the diagnostic performance. Heterogeneity among studies was determined by I² statistic. When heterogeneity was noted, Spearman correlation coefficient between the sensitivity and FP rate was calculated to evaluate whether a “threshold effect” existed. We used Deeks’ funnel plot to assess potential publication bias and used Deeks’ asymmetry test to test the statistical significance. Slope coefficients with P < 0.1 were regarded as statistically significant asymmetry, indicating that the results have a publication bias (5).

Results

Literature search

The flow diagram of the study selection process is showed in Fig. 1. The literature search of PubMed, EMBASE, and the CENTRAL databases identified a total of 332 unique references; of them, 123 were deemed potentially relevant based on the titles and abstracts. Of these articles, 73 were further excluded after full-text reading (including 37 review articles, 21 case reports, 1 letter, 3 animal studies, and 11 conference proceedings). Finally, a total of 50 articles satisfied the inclusion criteria and were selected for data extraction and analysis (6 –55) (Fig. 1).

Fig. 1.

Flow diagram of the study selection process.

Study and design characteristics

The characteristics of the 50 included studies are demonstrated in Table 1. Of these 50 articles, 26 were prospective and 24 were retrospective. A total of 32 studies reported the diagnostic performance of TCS, 7 described the diagnostic performance of SWE, 25 described the performance of GBA, 7 described the performance of HA enlargement, and 11 described the performance of overall US features.

Table 1.

Characteristics of included studies.

No.	First author	Year	Affiliation	Total No. of patients	Patients with BA	Prospective or retrospective	US features
1	Humphrey (6)	2007	Department of Radiology	90	30	Prospective	TCS, GBA
2	Lee (7)	2009	Department of Radiology	64	29	Retrospective	TCS, GBA
3	Tan Kendrick (8)	2003	Department of Diagnostic Imaging	217	31	Prospective	TCS, GBA
4	Woo Sun Kim (9)	2007	Department of Radiology	85	38	Retrospective	TCS, GBA, HA
5	Aziz (10)	2011	Department of Radiology and Biomedical Imaging	35	15	Retrospective	TCS, GBA
6	Kanegawa (11)	2003	Department of Radiology	55	29	Prospective	TCS, GBA
7	Takamizawa (12)	2007	Department of Surgery	85	48	Retrospective	TCS, GBA
8	Kotb (13)	2001	Departments of Pediatrics and Radiology	65	25	Prospective	TCS, GBA
9	Lee (14)	2003	Department of Diagnostic Radiology	86	20	Retrospective	TCS
10	Imanieh (15)	2010	Gastroenterohepatology Research Center	58	10	Prospective	TCS
11	Mittal (16)	2010	Department of Radio Diagnosis and Imaging	99	30	Prospective	TCS, GBA, HA
12	El-Guindi (17)	2013	Department of Pediatric Hepatology	54	27	Prospective	TCS, GBA, HA
13	El-Guindi (18)	2014	Department of Pediatric Hepatology	60	30	Prospective	TCS, GBA, HA
14	Visrutaratna (19)	2003	Department of Radiology	46	23	Retrospective	TCS, GBA
15	Park (20)	2001	Department of Pediatric Surgery	87	30	Retrospective	TCS
16	Tan Kendrick (21)	2000	Department of Diagnostic Imaging	60	12	Prospective	TCS, GBA
17	Hanquinet (22)	2015	Department of Pediatric Radiology	20	10	Retrospective	TCS, GBA
18	Zhou (23)	2015	Department of Medical Ultrasonics	273	129	Retrospective	TCS, HA
19	Park (24)	1997	Department of Pediatric Surgery	63	20	Prospective	TCS, US
20	Kim (25)	2000	Department of Diagnostic Radiology	21	12	Prospective	TCS
21	Ryeom (26)	2005	Department of Radiology	23	4	Prospective	TCS
22	Jiang (27)	2013	Department of Ultrasound	51	23	Prospective	TCS, GBA, HA
23	Lee (28)	2015	Department of Radiology	100	46	Retrospective	TCS, GBA
24	Wongsawasdi (29)	2008	Department of Pediatrics	61	31	Retrospective	TCS
25	Choochuen (30)	2019	Department of Radiology	100	51	Retrospective	TCS, GBA, HA
26	Weng (31)	2019	Department of Medical Ultrasonics	80	45	Prospective	TCS
27	Duan (32)	2019	Department of Ultrasound	138	51	Retrospective	TCS, GBA, SWE
28	Dong (33)	2018	Department of Pediatrics	396	178	Retrospective	TCS
29	Zhou (34)	2017	Department of Medical Ultrasonics	172	97	Prospective	TCS, SWE
30	Roquete (35)	2008	Departamento de Pediatria	96	51	Retrospective	TCS
31	Park (36)	1999	Department of Pediatric Surgery	79	25	Prospective	TCS
32	Dillman (37)	2019	Department of Radiology	41	13	Prospective	SWE
33	Wu (38)	2018	Department of Pediatrics	48	15	Prospective	SWE
34	Wang (39)	2016	Department of Ultrasound	69	38	Retrospective	SWE
35	Shen (40)	2020	Department of General Surgery	282	135	Retrospective	SWE
36	Sun (41)	2011	Department of Ultrasound	182	151	Retrospective	US, GBA
37	Yang (42)	2009	Nuclear Medicine Department	69	34	Retrospective	US
38	Lai (43)	1994	Department of Pediatrics	126	42	Prospective	US
39	Dehghani (44)	2006	Department of Pediatric Gastroenterology	65	19	Prospective	US
40	Azuma (45)	2003	Department of Pediatric Surgery	30	23	Prospective	US
41	Boskovic (46)	2014	Department of Gastroenterology	109	71	Prospective	US
42	Farrant (47)	2000	Variety Club Children's Hospital	346	71	Prospective	US
43	Lin (48)	1997	Department of Nuclear Medicine	63	15	Retrospective	US
44	Sung (49)	2016	Department of Radiology and Center for Imaging Science	64	41	Retrospective	US
45	Koob (50)	2017	Pediatric Radiology	321	193	Retrospective	TCS, GBA
46	Jancelewicz (51)	2015	Division of Pediatric Surgery	212	45	Retrospective	GBA
47	Lee (52)	2000	Department of Nuclear Medicine	152	49	Prospective	GBA
48	Wang (53)	2020	Department of Ultrasound	294	89	Prospective	SWE, GBA
49	Farrant (54)	2001	Department of Radiology	158	37	Prospective	US, GBA
50	Ikeda (55)	1998	Department of Paediatric Surgery	72	34	Retrospective	GBA

BA, biliary atresia; GBA, gallbladder and biliary system abnormality; HA, hepatic artery; SWE, shear wave elastography; TCS, triangular cord sign; US, ultrasonic.

QUADAS-2 was used for the overall quality of the included studies. The quality assessment chart is showed in Fig. 2. The included studies basically met the criteria for quality assessment of diagnostic accuracy.

Fig. 2.

Quality assessment of the studies selected for the meta-analysis (QUADAS-2).

The diagnostic performance of different US features

Diagnostic performance of TCS

Data on the diagnostic performance of TCS were collected from 32 studies on 3240 patients. A coupled forest plot of the sensitivity and specificity of the diagnostic performance of TCS in the 32 included studies (1 –31,45) is shown in Fig. 3a. The sensitivities and specificities of individual studies were in the range of 23%–100% and 74%–100%, respectively. The TCS showed summary sensitivity of 77% (95% confidence interval [CI] = 69%–84%), specificity of 98% (95% CI = 96%–99%), DOR of 140 (95% CI = 73–272), and AUC of 0.97 (95% CI = 0.95–0.98) for the detection of BA (Fig. 3b). The between-study heterogeneity was high for both sensitivity (I² = 94.4%, P < 0.001) and specificity (I² = 87.3%, P < 0.001). The Spearman correlation coefficient between sensitivity and FP rate was 0.22 (P = 0.22), indicating no threshold effect. No significant publication bias existed among the studies (P = 0.47) (Fig. 3c).

Fig. 3.

The diagnostic performance of the TCS from the 32 included studies (1 –31,45). (a) Forest plots of sensitivity and specificity of TCS for diagnosis. (b) Summary ROC curve. (c) Funnel plot for evaluating publication bias. TCS, triangular cord sign. ROC, receiver operating characteristic; TCS, triangular cord sign.

Diagnostic performance of SWE

Data on the diagnostic performance of SWE were collected from seven studies (27,29,32 –35,48) on 1044 patients. The sensitivities and specificities of individual studies were in the range of 73%–97% and 67%–100%, respectively. A coupled forest plot of the sensitivity and specificity of the diagnostic performance of SWE in the seven included studies is shown in Fig. 4a. The SWE showed summary sensitivity of 86% (95% CI = 78%–92%), specificity of 86% (95% CI = 72%–94%), DOR of 39 (95% CI = 11–142), and AUC of 0.92 (95% CI = 0.89–0.94) (Fig. 4b). Substantial heterogeneity in both sensitivity (I² = 83.0%) and specificity (I² = 89.3%) existed. The Spearman correlation coefficient between sensitivity and FP rate was −0.46 (P = 0.29), indicating no threshold effect. No significant publication bias existed among the studies (P = 0.11) (Fig. 4c).

Fig. 4.

The diagnostic performance of the SWE from the seven studies (27,29,32 –35,48). (a) Forest plots of sensitivity and specificity of SWE for diagnosis. (b) Summary ROC curve. (c) Funnel plot for evaluating publication bias. SWE, shear wave elastography. ROC, receiver operating characteristic; SWE, shear wave elastography; TCS, triangular cord sign.

Diagnostic performance of GBA

Data on the diagnostic performance of GBA were collected from 25 studies (1 –8, 11 –14,16,17,22,23,25,27,36,45 –50) on 2803 patients. Some studies conducted diagnostic performance analysis on various types of GBA, including absence of gallbladder, small gallbladder, irregular gallbladder wall, absence of gallbladder constriction, and absence of common bile duct (CBD). All these different types of GBA were collected and analyzed, and the results showed summary sensitivity of 75% (95% CI = 65%–83%), specificity of 92% (95% CI = 86%–95%), DOR of 33 (95% CI = 18–58), and AUC of 0.91 (95% CI = 0.89–0.93) for the detection of BA (Table 2). The between-study heterogeneity was high for both sensitivity (I² = 95.2%, P < 0.001) and specificity (I² = 94.5%, P < 0.001). The Spearman correlation coefficient between sensitivity and FP rate was 0.34 (P = 0.01), and the coupled forest plot of sensitivity and specificity did not reveal a threshold effect. No significant publication bias existed among the studies (P = 0.68).

Table 2.

The diagnostic performance of different US features used for the diagnosis of BA.

Ultrasound feature	No.	Sensitivity (%)	Specificity (%)	DOR	AUC
TCS	32	77 (69–84)	98 (96–99)	140 (73–272)	0.97 (0.95–0.98)
SWE	7	86 (78–92)	86 (72–94)	39 (11–142)	0.92 (0.89–0.94)
GBA	25	75 (65–83)	92 (86–95)	33 (18–58)	0.91 (0.89–0.93)
HA enlargement	7	81 (69–90)	79 (67–87)	16 (6–43)	0.87 (0.84–0.90)
Overall US	11	84 (72–92)	86 (77–92)	39 (12–94)	0.92 (0.89–0.94)

AUC, area under the curve; BA, biliary atresia; DOR, diagnostic odds ratio; GBA, gallbladder and biliary system abnormality; HA, hepatic artery; SWE, shear wave elastography; TCS, triangular cord sign; US, ultrasonic.

Subgroup analysis of GBA was performed according to different types (Table 3). Absence of gallbladder had the highest specificity (98%), but its sensitivity (29%) was very low. In addition, irregular gallbladder wall also had a very high specificity (98%), and the sensitivity (59%) was higher than that of absence of gallbladder. The absence of gallbladder constriction and absence of CBD showed high sensitivity and low specificity. Irregular gallbladder wall, absence of gallbladder contraction, and absence of CBD had good diagnostic performance (AUC = 0.97, 0.92, and 0.94, respectively). A total of 14 articles studied the diagnostic performance of overall abnormal gallbladder on BA, with an AUC of 0.95 (95% CI = 0.92–0.96), which is similar to the results obtained by our present meta-analysis of various GBA in 29 different studies.

Table 3.

Subgroup analysis of different types of GBA.

Ultrasonic features	No.	Sensitivity (%)	Specificity (%)	DOR	AUC
GBA	25
Absence	15	29 (19–40)	98 (94–100)	23 (8–71)	0.71(0.67–0.75)
Small	8	80 (69–88)	82 (71–89)	18 (7–47)	0.88 (0.84–0.90)
Irregular wall	4	59 (23–88)	98 (92–100)	76 (6–987)	0.97 (0.95–0.98)
Absence of contraction	6	90 (75–97)	77 (56–90)	32 (5–215)	0.92 (0.89–0.94)
Absence of CBD	4	94 (88–97)	68 (41–87)	33 (9–123)	0.94 (0.91–0.96)
Abnormal gallbladder	14	92 (87–95)	85 (69–94)	63 (19–222)	0.95 (0.92–0.96)

AUC, area under the curve; CBD, common bile duct; DOR, diagnostic odds ratio; GBA, gallbladder and biliary system abnormality.

Diagnostic performance of HA enlargement

Data on the diagnostic performance of HA enlargement were collected from seven studies (4,11 –13,18,22,25) on 722 patients. A coupled forest plot of sensitivity and specificity of the diagnostic performance of HA enlargement in the 7 included studies was shown in Fig. 5a. HA enlargement showed summary sensitivity of 81% (95% CI = 69%–90%) and specificity of 79% (95% CI = 67%–87%). The DOR was 16 (95% CI = 6–43) and the hierarchical summary ROC curves showed the AUC was 0.87 (95% CI = 0.84–0.90) (Fig. 5b). The Higgins I² statistics showed substantial heterogeneity in both the sensitivity (I² = 72.3%) and specificity (I² = 88.2%). The coupled forest plot of sensitivity and specificity did not reveal a threshold effect. The Spearman correlation coefficient between sensitivity and FP rate was −0.36 (P = 0.43), indicating no threshold effect. No significant publication bias existed among the studies (P = 0.79) (Fig. 5c).

Fig. 5.

The diagnostic performance of the HA enlargement from the seven studies (4,11 –13,18,22,25). (a) Forest plots of sensitivity and specificity of HA enlargement for diagnosis. (b) Summary ROC curve. (c) Funnel plot for evaluating publication bias. HA, hepatic artery; ROC, receiver operating characteristic.

Diagnostic performance of overall US features

A total of 14 studies reported the diagnostic performance of overall US features on BA. Overall US features showed summary sensitivity of 84% (95% CI = 72%–92%) and specificity of 86% (95% CI = 77%–92%). The DOR was 39 (95% CI = 12–94). The hierarchical summary ROC curve was symmetric, and the AUC was 0.92 (95% CI = 0.89–0.94).

Discussion

Among the 50 retrieved articles, 32 articles studied the diagnostic performance of TCS, the most studied US feature. Our meta-analysis showed that it had the highest diagnostic accuracy for BA (AUC = 0.97), which is consistent with the results of Zhou's study in 2015 from 20 studies (56), where the summary sensitivity of TCS was as low as 74% and the summary specificity was as high as 97%. Compared with Zhou's study, the results obtained after joining the new 12 studies carried out after 2015 are similar, which further confirms that TCS is the most commonly used US feature for BA and has excellent sensitivity and specificity.

SWE technology is an advanced US elasticity technology with advantages of quantitative evaluation, simple results for reading, and convenient operation for measuring liver tissue hardness (57). The Young's modulus value of liver elasticity is related to the pathological stage of liver fibrosis (58). The larger the Young's modulus value is, the greater the stiffness and elasticity value. Fibrous tissue proliferation of the periportal vein region is the main pathological characteristics of BA, which can be manifested as increasing liver stiffness. A total of 7 SWE studies were included in our meta-analysis and the results demonstrated that despite its inconspicuous overall sensitivity and total specificity, it has excellent diagnostic accuracy (AUC = 0.92). It is worth noting that the cutoff values used in these 7 studies are not completely consistent. This may be due to the difference between machines and the operating distinction between ultrasound performers. Because the current studies on SWE for BA is relatively few, with only 7 items, subgroup analysis cannot be performed. We only made a rough evaluation on SWE's BA diagnostic performance. In the future, more relevant research is needed and subgroup analysis could be carried out.

Among the 50 included studies, 25 mentioned GBA, second only to TCS sign, which has been regarded as a reliable and widely used US feature. Although the diagnostic performance of GBA is lower than that of TCS, its overall diagnostic performance is still very good (AUC = 0.92). Compared with the meta-analysis on GBA in 2015 (56), our study made a detailed subgroup analysis of it, with more included studies and more detailed grouping category. GBA includes many conditions, covering absence of gallbladder, small gallbladder, irregular gallbladder wall, absence of contraction, absence of CBD, and overall abnormal gallbladder. A subgroup analysis on different types of GBA showed that absence of gallbladder had the highest specificity, but its sensitivity was very low. In addition, irregular gallbladder wall also had a very high specificity. Absence of contraction and CBD had high sensitivity, but low specificity. Irregular gallbladder wall, absence of contraction, and CBD all had good diagnostic performance. However, it is worth noting that recognizing GBA during ultrasound examination is a subjective task. For example, it is also difficult to identify CBD in healthy infants. Judging whether the gallbladder wall is irregular mainly depends on personal experience and subjective feelings of the sonographer. Unifying and developing an objective gallbladder classification system may improve the diagnostic accuracy of BA.

In addition to the commonly used TCS, GBA, and the newly developed SWE for differential diagnosis of BA, HA is also used as one of the indicators by researchers. Its sensitivity, specificity, and diagnostic performance are not particularly high, but it can still be used as an auxiliary sign to help identify BA. In addition, there are some studies that simply describe the diagnostic performance of overall US features for BA, without specifying which feature or combined features. Therefore, the sensitivity and specificity obtained by using the vague concept of ultrasound signs to diagnose BA is at a moderate level, but its diagnostic performance is still very high. In the future, studies on specified combined US features would assist the differential diagnosis of BA.

Most of the studies in this meta-analysis used TCS and GBA as diagnostic features, and surgery and biopsy as reference standards, with broad sensitivity and specificity. These differences may be caused by a variety of factors. First, the research we included had a large time span, and technological advances in US equipment can explain some of these changes. Second, there were some differences in the criteria for abnormal US features between studies. For example, some studies used a cutoff value of 3 mm for TCS, while some used 4 mm. Small differences in the cutoff values of SWE between studies existed, which may be due to differences between machines. There were some subtle differences in the definition of GBA in different studies. In addition to the US features mentioned in this paper, researchers have also studied the diagnostic performance of other features, such as hilar lymph nodes, subhepatic blood flow, and portal vein/hepatic artery ratio (17,31). In view of the small number of studies on these features, systematic review and evaluation has not been performed at this time. It is likely that these US features also show good diagnostic accuracy for BA.

In conclusion, TCS and GBA are the two most accurate and widely accepted ultrasound features currently used for differential diagnosis of BA. The newly developed SWE is an objective and convenient method with good diagnostic sensitivity, specificity, and diagnostic performance, and can be used as an excellent means in clinical applications. HA enlargement can be used as an auxiliary feature to help the diagnosis of BA. More research on the diagnostic value of combined ultrasound signs for BA is needed in the future.

Footnotes

Authors' note

Yajie Tang and Pan Yang have made substantial contributions to conception and design, including literature retrieval, acquisition of data, analysis and interpretation of data, drafting the manuscript.

Declaration of conflicting interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

ORCID iD

Pan Yang

References

Vij

Rela

. Biliary atresia: pathology, etiology and pathogenesis. Future Sci OA 2020;6:FSO466.

Wehrman

Waisbourd-Zinman

Wells

. Recent advances in understanding biliary atresia. F1000Res 2019;8:F1000 Faculty Rev-218.

Bezerra

Wells

Mack

, et al. Biliary atresia: clinical and research challenges for the twenty-first century. Hepatology 2018;68:1163–1173.

Lakshminarayanan

Davenport

. Biliary atresia: a comprehensive review. J Autoimmun 2016;73:1–9.

Devillé

Buntinx

Bouter

, et al. Conducting systematic reviews of diagnostic studies: didactic guidelines. BMC Med Res Methodol 2002;2:9.

Humphrey

Stringer

. Biliary atresia: US diagnosis. Radiology 2007;244:845–851.

Lee

Kim

Lee

, et al. Biliary atresia: color Doppler US findings in neonates and infants. Radiology 2009;252:282–289.

Tan Kendrick

Phua

Ooi

, et al. Biliary atresia: making the diagnosis by the gallbladder ghost triad. Pediatr Radiol 2003;33:311–315.

Kim

Cheon

Youn

, et al. Hepatic arterial diameter measured with US: adjunct for US diagnosis of biliary atresia. Radiology 2007;245:549–555.

10.

Aziz

Wild

Rosenthal

, et al. Pseudo gallbladder sign in biliary atresia--an imaging pitfall. Pediatr Radiol 2011;41:620–626.

11.

Kanegawa

Akasaka

Kitamura

, et al. Sonographic diagnosis of biliary atresia in pediatric patients using the “triangular cord” sign versus gallbladder length and contraction. AJR Am J Roentgenol 2003;181:1387–1390.

12.

Takamizawa

Zaima

Muraji

, et al. Can biliary atresia be diagnosed by ultrasonography alone? J Pediatr Surg 2007;42:2093–2096.

13.

Kotb

Sheba

, et al. Evaluation of the triangular cord sign in the diagnosis of biliary atresia. Pediatrics 2001;108:416–420.

14.

Lee

Park

, et al. Objective criteria of triangular cord sign in biliary atresia on US scans. Radiology 2003;229:395–400.

15.

Imanieh

Dehghani

Bagheri

, et al. Triangular cord sign in detection of biliary atresia: is it a valuable sign? Dig Dis Sci 2010;55:172–175.

16.

Mittal

Saxena

Sodhi

, et al. Role of abdominal sonography in the preoperative diagnosis of extrahepatic biliary atresia in infants younger than 90 days. AJR Am J Roentgenol 2011;196:W438–W445.

17.

El-Guindi

Sira

Konsowa

, et al. Value of hepatic subcapsular flow by color Doppler ultrasonography in the diagnosis of biliary atresia. J Gastroenterol Hepatol 2013;28:867–872.

18.

El-Guindi

Sira

, et al. Design and validation of a diagnostic score for biliary atresia. J Hepatol 2014;61:116–123.

19.

Visrutaratna

Wongsawasdi

Lerttumnongtum

, et al. Triangular cord sign and ultrasound features of the gall bladder in infants with biliary atresia. Australas Radiol 2003;47:252–256.

20.

Park

Choi

Lee

. Technical innovation for noninvasive and early diagnosis of biliary atresia: the ultrasonographic “triangular cord” sign. J Hepatobiliary Pancreat Surg 2001;8:337–441.

21.

Tan Kendrick

Phua

Ooi

, et al. Making the diagnosis of biliary atresia using the triangular cord sign and gallbladder length. Pediatr Radiol 2000;30:69–73.

22.

Hanquinet

Courvoisier

Rougemont

, et al. Contribution of acoustic radiation force impulse (ARFI) elastography to the ultrasound diagnosis of biliary atresia. Pediatr Radiol 2015;45:1489–1495.

23.

Zhou

Wang

Shan

, et al. Optimizing the US diagnosis of biliary atresia with a modified triangular cord thickness and gallbladder classification. Radiology 2015;277:181–191.

24.

Park

Choi

Lee

, et al. 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. J Pediatr Surg 1997;32:1555–1559.

25.

Kim

Park

Han

, 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.

26.

Ryeom

Choe

Kim

, et al. Biliary atresia: feasibility of mangafodipir trisodium-enhanced MR cholangiography for evaluation. Radiology 2005;235:250–258.

27.

Jiang

Chen

Ding

, et al. The diagnostic value of high-frequency ultrasonography in biliary atresia. Hepatobiliary Pancreat Dis Int 2013;12:415–422.

28.

Lee

Cheon

Choi

, et al. Ultrasonographic diagnosis of biliary atresia based on a decision-making tree model. Korean J Radiol 2015;16:1364–1372.

29.

Wongsawasdi

Ukarapol

Visrutaratna

, et al. Diagnostic evaluation of infantile cholestasis. J Med Assoc Thai 2008;91:345–349.

30.

Choochuen

Kritsaneepaiboon

Charoonratana

, et al. Is “gallbladder length-to-width ratio” useful in diagnosing biliary atresia? J Pediatr Surg 2019;54:1946–1952.

31.

Weng

Zhou

, et al. Enlarged hepatic hilar lymph node: an additional ultrasonographic feature that may be helpful in the diagnosis of biliary atresia. Eur Radiol 2019;29:6699–6707.

32.

Duan

Peng

Liu

, et al. Does supersonic shear wave elastography help differentiate biliary atresia from other causes of cholestatic hepatitis in infants less than 90 days old? Compared with grey-scale US. Biomed Res Int 2019;2019:9036362.

33.

Dong

Zhu

Chen

, et al. Clinical assessment of differential diagnostic methods in infants with cholestasis due to biliary atresia or non-biliary atresia. Curr Med Sci 2018;38:137–143.

34.

Zhou

Jiang

Shan

, et al. Liver stiffness measurements with supersonic shear wave elastography in the diagnosis of biliary atresia: a comparative study with grey-scale US. Eur Radiol 2017;27:3474–3484.

35.

Roquete

Ferreira

Fagundes

, et al. Accuracy of echogenic periportal enlargement image in ultrasonographic exams and histopathology in differential diagnosis of biliary atresia. J Pediatr 2008;84:331–336.

36.

Park

Choi

Lee

. 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.

37.

Dillman

DiPaola

Smith

, et al. Prospective assessment of ultrasound shear wave elastography for discriminating biliary atresia from other causes of neonatal cholestasis. J Pediatr 2019;212:60–65.e3.

38.

Lee

Lin

, et al. Transient elastography is useful in diagnosing biliary atresia and predicting prognosis after hepatoportoenterostomy. Hepatology 2018;68:616–624.

39.

Wang

Qian

Jia

, et al. Utility of shear wave elastography for differentiating biliary atresia from infantile hepatitis syndrome. J Ultrasound Med 2016;35:1475–1479.

40.

Shen

Tan

Wang

, et al. Combination of gamma-glutamyl transferase and liver stiffness measurement for biliary atresia screening at different ages: a retrospective analysis of 282 infants. BMC Pediatr 2020;20:276.

41.

Sun

Zheng

Qian

. Ultrasonographic evaluation in the differential diagnosis of biliary atresia and infantile hepatitis syndrome. Pediatr Surg Int 2011;27:675–679.

42.

Yang

Peng

, et al. Comparison of different diagnostic methods for differentiating biliary atresia from idiopathic neonatal hepatitis. Clin Imaging 2009;33:439–446.

43.

Lai

Chang

Hsu

, et al. Differential diagnosis of extrahepatic biliary atresia from neonatal hepatitis: a prospective study. J Pediatr Gastroenterol Nutr 1994;18:121–127.

44.

Dehghani

Haghighat

Imanieh

, et al. Comparison of different diagnostic methods in infants with cholestasis. World J Gastroenterol 2006;12:5893–5896.

45.

Azuma

Nakamura

Nakahira

, et al. 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.

46.

Boskovic

Kitic

Prokic

, et al. Predictive value of hepatic ultrasound, liver biopsy, and duodenal tube test in the diagnosis of extrahepatic biliary atresia in Serbian infants. Turk J Gastroenterol 2014;25:170–174.

47.

Farrant

Meire

Mieli-Vergani

. Ultrasound features of the gall bladder in infants presenting with conjugated hyperbilirubinaemia. Br J Radiol 2000;73:1154–1158.

48.

Lin

Changlai

, et al. Comparison technetium of Tc-99 m disofenin cholescintigraphy with ultrasonography in the differentiation of biliary atresia from other forms of neonatal jaundice. Pediatr Surg Int 1997;12:30–33.

49.

Sung

Jeon

Yoo

, et al. Incremental value of MR cholangiopancreatography in diagnosis of biliary atresia. PLoS One 2016;11:e0158132.

50.

Koob

Pariente

Habes

, et al. The porta hepatis microcyst: an additional sonographic sign for the diagnosis of biliary atresia. Eur Radiol 2017;27:1812–1821.

51.

Jancelewicz

Barmherzig

Chung

, et al. A screening algorithm for the efficient exclusion of biliary atresia in infants with cholestatic jaundice. J Pediatr Surg 2015;50:363–370.

52.

Lee

Wang

Lee

, et al. The significance of functioning gallbladder visualization on hepatobiliary scintigraphy in infants with persistent jaundice. J Nucl Med 2000;41:1209–1213.

53.

Wang

Jia

, et al. Development and validation of a nomogram incorporating ultrasonic and elastic findings for the preoperative diagnosis of biliary atresia. Acad Radiol 2020. DOI: 10.1016/j.acra.2020.08.035.

54.

Farrant

Meire

Mieli-Vergani

. Improved diagnosis of extraheptic biliary atresia by high frequency ultrasound of the gall bladder. Br J Radiol 2001;74:952–954.

55.

Ikeda

Sera

Ohshiro

, et al. Gallbladder contraction in biliary atresia: a pitfall of ultrasound diagnosis. Pediatr Radiol 1998;28:451–453.

56.

Zhou

Shan

Tian

, et al. Ultrasound for the diagnosis of biliary atresia: a meta-analysis. AJR Am J Roentgenol 2016;206:W73–W82.

57.

Taljanovic

Gimber

Becker

, et al. Shear-wave elastography: basic physics and musculoskeletal applications. Radiographics 2017;37:855–870.

58.

Zemanová

. New diagnostic imaging technique - shear wave elastography. Cesk Slov Oftalmol 2016;72:103–110.