What Is Known
- Definitive diagnosis of biliary atresia is confirmed by surgical cholangiogram.
- Duodenal tube test is a preoperative procedure that is used to examine extrahepatic bile duct patency based on duodenal fluid color. It has high sensitivity but relatively low specificity.
What Is New
- Maximum total bile acid concentration in duodenal fluid, duodenal total bile acid/serum total bile acid ratio, and duodenal total bile acid/serum gamma-glutamyl transpeptidase ratio demonstrated high sensitivity and specificity to rule out biliary atresia.
- Duodenal tube test evaluated by duodenal total bile acid was a useful method to rule out biliary atresia owing to its high accuracy. It could be performed safely and rapidly.
The incidence of neonatal cholestasis (NC) is almost 1 in 2500 newborns and poses a complicated diagnostic challenge (1,2). Biliary atresia (BA) is the main cause of NC. It is characterized by destructive inflammatory obliterating cholangiopathy that affects varying lengths of both extrahepatic and intrahepatic bile ducts of unknown etiopathogenesis. If untreated, progressive liver fibrosis and portal hypertension can lead to death by the age of 2 years (3). An early and rapid diagnosis and surgical treatment (portoenterostomy) lead to better prognosis (4).
Definitive diagnosis of BA is made by surgical cholangiogram. Given the invasive nature of surgical cholangiogram, several diagnostic procedures such as abdominal ultrasound, hepatobiliary scintigraphy, magnetic resonance cholangiopancreatography, endoscopic retrograde cholangiopancreatography, and duodenal tube test (DTT) are, however, employed as preoperative screening techniques to rule out BA and prevent unnecessary surgery.
DTT is a diagnostic procedure often performed in Japanese and Chinese centers. In DTT, a nasoduodenal tube is passed into the third portion of the duodenum, and duodenal fluid is collected to examine the presence of bile (3). Although several studies have evaluated the accuracy of DTT in the diagnosis of BA, the sensitivity and specificity of bile-negative or bile-positive DTT have been assessed by the color of duodenal fluid (5–10), and quantitative and objective values were not present. In this study, we assessed bile-negative or bile-positive DTT using duodenal total bile acid (TBA) concentrations in combination with serum markers and examined its efficacy as a screening method to rule out BA.
In the Osaka University Hospital, when infants with cholestasis are admitted, laboratory tests and abdominal ultrasounds are performed after their clinical history and physical examinations have been taken. If BA cannot be ruled out by their clinical history, presentation, laboratory findings, and abdominal ultrasound, DTT is performed. When results of DTT cannot rule out BA, we perform intraoperative cholangiogram and liver biopsy and then portoenterostomy in cases with proven obstruction of the extrahepatic bile duct. If cholestasis persists in infants in whom BA has been denied, we perform further investigations including genetic tests.
This study included infants with cholestasis who were admitted to our hospital over a period of 9 years (2008–2016) and underwent DTT to rule out BA. Their clinical history, clinical presentation, laboratory findings, outcome of DTT (evaluated by the color of duodenal fluid and TBA concentration in duodenal fluid), and complications of DTT, and final diagnosis were reviewed retrospectively. Stool color was evaluated by clinician's subjective judgment until 2012 and was evaluated based on stool color card since 2012.
For DTT, a 5-Fr nasoduodenal tube was placed at the third portion of the duodenum using fluoroscopy, and its position was confirmed by abdominal radiography 3 to 5 hours after insertion of the tube. Oral feeding was stopped under intravenous support with 4.2% dextrose-saline solution during the procedure. At least 0.5 mL of duodenal fluid was collected by gravity every 3 hours, and the color of the duodenal fluid and TBA concentration in the fluid were examined. The duodenal fluid obtained during night time was stored at 4°C until next morning. If the color of the duodenal fluid was not bile-pigmented yellow by 6 hours, 5 mL of 20% magnesium sulfate was introduced into the tube.
TBA concentration in duodenal fluid was measured with enzymatic cycling method using commercially available test reagents (Kainos Laboratories Inc, Tokyo, Japan) and was performed on the autoanalyzer (BioMajesty JCA-BM8040, JEOL, Tokyo, Japan) according to the manufacturer's protocols. Four microliters of supernatant of duodenal fluid was required for each measurement.
In DTT assessed by the color of duodenal fluid, the test was considered bile positive, if bile-pigmented yellow biliary fluid was observed, and the test was considered bile negative, when colored duodenal fluid was not collected within 24 hours.
For DTT evaluated by the TBA concentration in duodenal fluid, maximum TBA in duodenal fluid (dTBA), dTBA/serum TBA (sTBA) ratio, and dTBA/serum gamma-glutamyl transpeptidase (sGGT) ratio were examined. sTBA and sGGT were obtained within 6 days of DTT.
All statistical analyses were performed with EZR (Saitama Medical Center, Jichi Medical University, Saitama, Japan), which is a graphical user interface for R (The R Foundation for Statistical Computing, Vienna, Austria) (11). The chi-square test and Wilcoxon rank-sum nonparametric test were used for comparisons of patient characteristics, dTBA, dTBA/sTBA ratio, and dTBA/sGGT ratio between infants with BA and those without BA. Correlations were assessed by Spearman rank correlation test. A P value of <0.05 was regarded as significant. The cut-off values for dTBA, dTBA/sTBA ratio, and dTBA/sGGT ratio were determined by receiver operating characteristic (ROC) analysis.
The protocol was approved by the Ethics Committee of Osaka University, and written informed consent was obtained from the parents of the enrolled children before study initiation.
In this study, 37 infants were evaluated for TBA concentration in duodenal fluid obtained by DTT to rule out BA. Patient characteristics were shown in Table 1. BA was diagnosed in 16 infants by intraoperative cholangiogram, and the etiology of other 21 infants included neonatal hepatitis (n = 12), parenteral nutrition–associated cholestasis (n = 3), neonatal intrahepatic cholestasis caused by citrin deficiency (n = 2), paucity of intrahepatic bile duct with unknown etiology (n = 2), cytomegalovirus hepatitis (n = 1), and progressive familial intrahepatic cholestasis (PFIC) type 1 (n = 1). Fifteen of 16 infants with BA and 11 of 21 infants without BA presented with acholic stools. The mean age at DTT was 53.5 ± 25.1 days (standard deviation) and 5 infants were born before 37 weeks of gestation. Serum GGT was significantly elevated in infants with BA than those without BA (747 ± 568 vs 123 ± 69 U/L, P < 0.01), and direct bilirubin was significantly elevated in infants with BA than in those without BA (5.7 ± 1.7 vs 4.1 ± 2.2 mg/dL, P < 0.05).
The results of DTT with colorimetric assessment, surgical cholangiogram, and final diagnosis are shown in Figure 1. Of the total, 21 infants were noted to have bile-negative DTT as assessed by the color of duodenal fluid, and 19 infants underwent surgical cholangiogram. Two infants did not undergo surgical cholangiogram despite the bile-negative DTT. One of them could not undergo surgical cholangiogram because he had low body weight of 1.2 kg. The other showed normal gallbladder when examined using abdominal ultrasound, and his GGT was within the normal range. Cholestasis in both infants was resolved spontaneously in 1 week. Among 19 infants who had bile-negative DTT and who underwent surgical cholangiogram, 16 infants were diagnosed with BA, and 3 infants were diagnosed with neonatal hepatitis, PFIC type 1, and paucity of bile ducts with unknown etiology.
Sixteen infants showed bile-positive DTT as assessed by the color of duodenal fluid, and 15 infants were not explored surgically. One infant underwent surgical cholangiogram as his hepatobiliary scintigraphy did not rule out the possibility of BA. Finally, all infants were diagnosed as other causes of NC. Thus, sensitivity, specificity, and positive and negative predictive values evaluated by the color of duodenal fluid were 100%, 76.2%, 76.2%, and 100%, respectively.
In each DTT, 3.8 ± 2.2 duodenal fluid samples were measured for TBA concentration. The number of duodenal samples was significantly higher in infants with BA than those without BA (4.7 ± 2.1 vs 3.2 ± 2.1, P < 0.05). The dTBA level was significantly lower in infants with acholic stools (558 ± 2203 vs 2416 ± 4588 μmol/L, P < 0.01), but there were no correlations between dTBA level and age at DTT in both all infants participated in this study including ones born in prematurely, and patients with BA (P = 0.658 and P = 0.822, respectively).
dTBA, dTBA/sTBA ratio, and dTBA/sGGT ratio were significantly lower in infants with BA than in those without BA (Table 2). Dot-plot and ROC curve for each parameter are shown in Figure 2. When dTBA <16.8 μmol/L was defined as bile-negative, sensitivity, specificity, and positive and negative predictive values were 100%, 90.5%, 88.9%, and 100%, respectively. Furthermore, with the optimal cut-off values of 0.088 and 0.076 μmol/U, specificity was improved to 95.2% and 100% for dTBA/sTBA ratio and dTBA/sGGT ratio, while keeping the sensitivity of 100%. AUCROC of dTBA, dTBA/sTBA ratio, and dTBA/sGGT ratio were 0.993 (95% confidence interval [95% CI]: 0.977–1.0), 0.997 (95% CI: 0.989–1.0), and 1.0 (95% CI: 1.0–1.0), respectively.
Even when TBA concentration in duodenal fluid was analyzed in infants who presented with acholic stools, the values of dTBA, dTBA/sTBA ratio, and dTBA/sGGT ratio were significantly lower in infants with BA compared with infants without BA (Table, Supplementary Digital Content 1, http://links.lww.com/MPG/B398). The sensitivity and specificity of dTBA were 100% and 81.8% with cut-off value of 9.3 μmol/L, and the specificities were improved to 90.9% and 100% for dTBA/sTBA and dTBA/sGGT ratio with cut-off values of 0.061 and 0.044 μmol/U, respectively, while keeping the sensitivities of 100% (Figure, Supplementary Digital Content 2, http://links.lww.com/MPG/B399).
The duration from admission to the time of obtaining the result of DTT was 0.8 ± 1.4 days. The time required for the fluoroscopy to place the nasoduodenal tube was 8.6 ± 5.9 minutes. Mild hypoglycemia was observed in 1 case with citrin deficiency and treated promptly with glucose infusion. No other complications associated with DTT were observed.
The prognosis of BA is extremely poor if portoenterostomy is not performed. Therefore, the sensitivity of the screening test to rule out BA was kept 100% in this study. In addition, screening tests should be performed in a short time as late diagnosis and surgical treatment can lead to poor results.
Bile acids are synthesized and conjugated in hepatocytes and are excreted into the intestinal lumen. In cholestatic conditions, impaired bile formation in hepatocytes and/or obstruction of bile flow through the intrahepatic and extrahepatic biliary trees results in decreased secretion of bile acids into the duodenum and increased efflux in the systemic circulation. DTT was a preoperative screening method used to examine the patency of intrahepatic and extrahepatic bile ducts by the color of duodenal fluid and avoid unnecessary surgical exploration. If the color of duodenal fluid was bile-pigmented yellow, the patency of the extrahepatic bile duct was proved, and BA was denied. DTT was first reported in 1979 (5) and several studies explore the efficacy of DTT (6–10). In previous studies, the outcome of DTT was accessed by the color of duodenal fluid, and they showed high sensitivity (90.6%–100%) and relatively low specificity (60%–93.7%). We propose that the relatively low specificity was due to the colorimetric evaluation of duodenal fluid which was not objective and quantitative. In this study, we hypothesized that infants with BA, who had intrahepatic or extrahepatic bile duct obstruction, had lower TBA concentration in duodenal fluid than other causes of NC, and duodenal TBA concentration may serve as a quantitative and objective marker with better accuracy for the diagnosis of BA than the colorimetric assessment.
In this retrospective study that included 37 infants, DTT examined by maximum duodenal TBA concentration showed higher accuracy as a preoperative evaluation to rule out BA than the colorimetric assessment. Of the total, 5 infants were false-positive by colorimetric assessment. Among them, cholestasis was resolved spontaneously in 2 before surgical cholangiogram, and they were diagnosed with NC. Their dTBA were 6.5 and 116.5 μmol/L, respectively. The rest of the 5 infants underwent surgical cholangiogram. They were finally diagnosed with PFIC type 1, neonatal hepatitis, and nonsyndromic paucity of bile duct. Their dTBA were 9.2, 31.2, and 87.9 μmol/L, respectively. If DTT was assessed by TBA in duodenal fluid, at least 2 of the infants could have avoided surgical exploration.
Furthermore, aiming to improve the specificity of the screening method while keeping the sensitivity 100%, we combined some serum markers with duodenal TBA concentration and examined the efficacy. When we involved serum TBA, which supposed to be elevated in cholestatic conditions, the specificity improved to 95.2% with the ratio of dTBA/sTBA, which means that 3 infants could have avoided surgery. GGT has been reported the most reliable parameter in blood tests to differentiate BA from other causes of NC (12,13). In this study, sGGT was significantly elevated in infants with BA than those without BA, and dTBA/sGGT ratio had the highest accuracy (sensitivity and specificity were both 100%) for the diagnosis of BA. Therefore, we believe that TBA concentration in duodenal fluid combined with serum GGT levels can be a potential differentiating parameter.
Acholic stool is one of the primary symptoms of BA. It is, however, well-known that it is insufficient to diagnose or rule out BA solely by the color of stools. Infants without BA might present with complete acholic stools if the degree of intrahepatic cholestasis was severe. On the contrary, infants with BA whose extrahepatic bile duct was obstructed may present with pigmented stools as seen in one of the patients with BA in this study. Therefore, further examinations including DTT are required to rule out BA. dTBA, dTBA/sTBA ratio, and dTBA/sGGT ratio kept high accuracy on infants with acholic stools and this suggested that DTT had sufficient power to rule out BA in infants with acholic stools.
Several other procedures other than DTT have been performed to rule out BA. Abdominal ultrasound can be performed safely and rapidly, but this procedure is operator dependent, and reported sensitivities (55%–87%) and specificities (71%–83%) were insufficient (12,14–17). Hepatobiliary scintigraphy can also be performed safely, and reported sensitivities were high (83%–100%), but specificities were low (45%–79%) (12,13). Kwatra et al (18) reported that hepatobiliary scintigraphy after pretreatment with phenobarbital had high sensitivity (100%) and specificity (93%), but this examination required 5 days for phenobarbital pretreatment and could not be performed rapidly. Magnetic resonance cholangiopancreatography can be performed safely, but sensitivity and specificity for the diagnosis of BA are poor (14,19). Endoscopic retrograde cholangiopancreatography can visualize the biliary tree and may provide similar data as surgical cholangiogram (20); however, it is invasive and technically difficult in infants. Liver biopsy is the usual diagnostic procedure in some countries (3). Results of varying degrees of portal tract fibrosis, edematous portal tracts with inflammatory changes, bile duct proliferation, and intraportal bile plugs support the diagnosis of BA; however, evidence of giant cell transformation suggests the diagnosis of idiopathic hepatitis (21). It is reported to have high sensitivities (89%–99%) and specificities (82.5%–98%), but as a screening method to rule out BA, it is invasive and needs a time delay of 1 to 3 days (12,22). Interobserver variation may be another issue (23).
In this study, DTT evaluated using maximum TBA concentration in duodenal fluid, maximum TBA in duodenal fluid/serum TBA ratio, and maximum TBA in duodenal fluid/serum GGT ratio had high accuracy and could be performed safely and rapidly. This study suggested that DTT is a useful preoperative screening method to avoid unnecessary intraoperative cholangiogram, when biliary atresia cannot be denied by their clinical history, physical presentation, and laboratory tests, and can replace some other diagnostic procedures.
This study had several limitations. This was a retrospective analysis, and the number of subjects was relatively small. A prospective study is needed to confirm the results of our study.
In conclusion, this study demonstrates that DTT evaluated using duodenal TBA concentration is a safe and prompt preoperative screening method with high accuracy to rule out BA.
1. Poddar U, Thapa BR, Das A, et al. Neonatal cholestasis
: differentiation of biliary atresia from neonatal hepatitis in a developing country. Acta Paediatr
2. Fawaz R, Baumann U, Ekong U, et al. Guideline for the evaluation of cholestatic jaundice in infants: joint recommendation of the North American Society for Pediatric Gastroenterology, Hepatology and Nutrition and the European Society for Pediatric Gastroenterology, Hepatology, and Nutrition. J Pediatr Gastroenterol Nutr
3. Hartley JL, Davenport M, Kelly DA, et al. Biliary atresia. Lancet
4. Serinet MO, Wildhaber BE, Broué P, et al. Impact of age at Kasai operation on its results in late childhood and adolescence: a rational basis for biliary atresia screening. Pediatrics
5. Greene HL, Helinek GL, Moran R, et al. A diagnostic approach to prolonged obstructive jaundice by 24-hour collection of duodenal fluid. J Pediatr
6. Larrosa-Haro A, Caro-Lopez AM, Coello-Ramirez P, et al. Duodenal tube test in the diagnosis of biliary atresia. J Pediatr Gastroenterol Nutr
7. Meisheri IV, Kasat LS, Kumar A, et al. Duodenal intubation and test for bile—a reliable method to rule out biliary atresia. Pediatr Surg Int
8. Boskovic A, Kitic I, Prokic D, 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 Gastroentrol
9. Lai MW, Chang MH, Hsu SC, et al. Differential diagnosis of extrahepatic biliary atresia from neonatal hepatitis: a prospective study. J Pediatr Gastroenterol Nutr
10. Faweya AG, Akinyinka OO, Sodeinde O, et al. Duodenal intubation and aspiration test: utility in the differential diagnosis of infantile cholestasis. J Pediatr Gastroenterol Nutr
11. Kanda Y. Investigation of the freely available easy-to-use software ‘EZR‘ for medical statistics. Bone Marrow Transplant
12. Agin M, Tumgor G, Alkan M, et al. Clues to the diagnosis of biliary atresia in neonatal cholestasis
. Turk J Gastroenterol
13. Arora NK, Kohli R, Gupta DK, et al. Hepatic technetium-99m-mebrofenin imiodiacetate scans and serum (-glutamyl transpeptidase levels interpreted in series to differentiate between extrahepatic biliary atresia and neonatal hepatitis. Acta Paediatr
14. Yang JG, Ma DQ, Peng Y, et al. Comparison of different diagnostic methods for differentiating biliary atresia from idiopathic neonatal hepatitis. Clin Imaging
15. Lin WY, Lin CC, Changlai SP, et al. Comparison technetium of Tc-99m disofenin cholescintigraphy with ultrasonography in the differentiation of biliary atresia from other forms of neonatal jaundice. Pediatr Surg Int
16. Takamizawa S, Zaima A, Muraji T, et al. Can biliary atresia be diagnosed by ultrasonography alone? J Pediatr Surg
17. Azuma T, Nakamura T, Nakahira M, et al. Pre-operative ultrasonographic diagnosis of biliary atresia—with reference to the presence of absence of the extrahepatic bile duct. Pediatr Surg Int
18. Kwatra N, Shalaby-Rana E, Narayanan S, et al. Phenobarbital-enhanced hepatobiliary scintigraphy in the diagnosis of biliary atresia: two decades of experience at a tertiary center. Pediatr Radiol
19. Lee MJ, Kim MJ, Yoon CS, et al. Gadopentetate dimeglumine-enhanced MR cholangiopancreatography in infants with cholestasis. Pediatr Radiol
20. Shteyer E, Wengrower D, Benuri-Silbiger I, et al. Endoscopic retrograde cholangiopancreatography in neonatal cholestasis
. J Pediatr Gastroenterol Nutr
21. Azar G, Beneck D, Lane B, et al. Atypical morphologic presentation of biliary atresia and value of serial liver biopsies. J Pediatr Gastroenterol Nutr
22. Moyer V, Freese DK, Whitington PF, et al. Guideline for the evaluation of cholestatic jaundice in infants: recommendations of the North American Society for Pediatric Gastroenterology, Hepatology and Nutrition. J Pediatr Gastroenterol Nutr
23. Russo P, Magee JC, Boitnott J, et al. Design and validation of the biliary atresia research consortium histologic assessment system for cholestasis in infancy. Clin Gastroenterol Hepatol