Neonatal cholestasis (NC) affects approximately 1 in 2500 births and is defined as conjugated hyperbilirubinemia that occurs in the newborn period (1). The differential diagnosis for NC is varied. Timely diagnosis and treatment of structural and metabolic disorders are critical for optimal outcome. Diagnostic tests typically include laboratory tests, abdominal imaging, and percutaneous liver biopsy. Diagnosis may be delayed due to late referral or lengthy and cumbersome investigations. Unfortunately, clinical features and the routine liver function blood tests fail to distinguish extrahepatic biliary atresia (EHBA) from other causes of NC. Even after extensive workup, the diagnosis of EHBA is achieved in only 84% of affected infants before laparotomy (2). The criterion standard for diagnosis is intraoperative cholangiogram, but the infant should be thoroughly investigated before surgery.
In 1976, Waye (3) reported the first successful endoscopic retrograde cholangiopancreatography (ERCP) in a 3.5-month-old child using an adult-size duodenoscope. Since then, the development of smaller-diameter duodenoscopes has enabled the performance of ERCP in babies. A few studies have suggested that ERCP has a role in the diagnosis algorithm of NC (4–6), but the data supporting the use of ERCP in NC remain limited. The use of ERCP in infants is technically difficult and its use is confined to large centers (7). The aim of the present study was to determine the use, complication rate, and success rate of ERCP in a cohort of infants who received ERCP as part of the evaluation of NC in a single institution.
Between January 2000 and January 2010, 100 children younger than 16 years, including 27 infants younger than 3 months, underwent ERCP in the Department of Gastroenterology of Hadassah-Hebrew University Hospital, Jerusalem, Israel. During this period, only 2 gastroenterologsits performed ERCP (D.W. and E.G.). Our center is regarded as a national referral center for pediatric ERCP, and a regional referral center for the Kasai procedure. Data on demographics, diagnosis, treatment, and complications were collected.
Informed consent was obtained from the parents before ERCP. All of the infants underwent the procedure under general anesthesia by mask induction with oxygen and halothane until 2003, and thereafter sevoflurane or intravenous propofol 0.5 to 1 mg/kg followed by a continuous drip of intravenous propofol 3 mg · kg−1 · hour−1 were used. Endotracheal intubation was performed in all 27 children (all weighing <10 kg). ERCP was performed using the Olympus PJF 7.5 pediatric duodenoscope (Olympus, Tokyo, Japan). Cannulation was performed with an Olympus tapered Teflon catheter (PR–10Q). In all of the cases, 50% Urografin contrast medium was injected slowly under fluoroscopy. No more than 3 attempts were made to visualize the biliary tract. Diagnosis of abnormal ERCP findings was made according to previous publications (8,9). Failed ERCP was defined as the inability to cannulate the ampulla.
During the 10-year period, ERCP was performed on 27 infants younger than 3 months as part of the evaluation for cholestasis. Of the 27, 14 were boys (52%). The median age at time of procedure was 55 days (range 33–89). Before ERCP, each patient underwent extensive assessment for the causes of cholestasis that included infectious, metabolic, and genetic workup. Abdominal ultrasound (US) was normal in 16 infants, and the others showed small gallbladder (4), biliary stones (3), dilated bile ducts (3), and no gallbladder (1) (Fig. 1).
From the US examination, it was evident that 6 children did not have EHBA (biliary stones  and dilated ducts ) (Fig. 1). In the 21 infants who had normal US or small or no gallbladder, suspicion of EHBA was high. Thirteen infants had liver biopsies; in 6, the results were equivocal, 2 had neonatal hepatitis, and 5 were suggestive of EHBA. Only 13 infants underwent scintigraphy scan, all of which were abnormal. Of the 21 infants with suspicion of EHBA (Fig. 1), 15 had biliary atresia (Fig. 2), 4 had normal ERCP (Fig. 3), in 1 infant the ducts were not filled completely, and in 3 infants cannulation of ampulla failed. Of the infants who were diagnosed as having EHBA and had normal US, 8 (88%) had normal gallbladders with obliteration of the common bile duct and patency of the proximal bile ducts (type 1 EHBA).
ERCP indicated the diagnosis of biliary atresia in 13 infants. Two infants were diagnosed by operative cholangiography after unsuccessful ERCP attempts. In 4 other infants, the ERCP was normal. Biliary stones were confirmed in 2 infants, in which only flushing the bile duct was done. In 1 infant, dilatation of bile duct was found and was presumed to have had biliary stones. No further treatment was administered to these infants, with decreased bilirubin 48 hours after procedure. In 1 infant, a choledochal cyst was found (Fig. 4). Of the group of infants who had a high suspicion of EHBA, ERCP excluded EHBA and prevented surgery and intraoperative cholangiogram in 4 (19%).
The final diagnoses in our cohort (Fig. 1) were EHBA (15), biliary stones (5), neonatal hepatitis (4), choledochal cyst (1), paucity of intrahepatic bile ducts (1), and congenital hepatic fibrosis (1). Of the infants with examination failure (5), 2 had biliary atresia, 2 had biliary stones, and 1 had paucity of bile ducts. There were no complications during or after the ERCP procedure.
ERCP is the procedure of choice for investigation of biliary and pancreatic diseases in adults. In children the indications and use of ERCP are still evolving, but increasing data support its use. More than a decade ago, a North American Society of Pediatric Gastroenterology, Hepatology, and Nutrition committee (10) summarized the use of ERCP in children, but did not reach a consensus based on the available data. Since then, several reviews and prospective studies assessing the use of ERCP in children (5,11,12) have been published; however, there have been extremely few studies on ERCP in neonates (4,13). Guelrud et al (8) reported on 32 infants with NC who were highly suspected of having EHBA and underwent ERCP. This was a prospective study in which infants underwent ERCP who would otherwise have undergone intraoperative cholangiogram. Shanmugam et al (6) reported 48 who underwent ERCP because of inconsistency of the liver biopsy with clinical features such as acholic stools or accompanied symptoms. ERCP demonstrated patency of the biliary tract in 42% of infants. Another study by Keil et al (7) reviewed 104 infants with cholestasis, in which ERCP was performed in infants with acholic stools and normal US. In our study, infants were referred when noninvasive tests were inconclusive or discordant with the clinical features: normal US with cholestasis or small gallbladder with questionable acholic stools. In cases in which there was an increased possibility of EHBA, patients would not undergo ERCP, but rather were referred directly for intraoperative cholangiogram. This may have led to a bias in our cohort, but we stress the role of ERCP in the diagnostic algorithm for NC only in unclear cases when other tests are equivocal.
Timely differentiation of biliary atresia from other causes of NC is of particular importance because delayed surgical correction is associated with poor long-term prognosis (14,15). Nevertheless, even after an extensive workup, diagnosis of EHBA is achieved in only 84% of infants prelaparotomy (2). Methods such as radionuclide biliary tract scan (16) and magnetic resonance cholangiopancreatography (MRCP) (17) have unacceptably poor sensitivity and specificity for EHBA. No studies comparing MRCP to ERCP have been reported, but MRCP, although noninvasive, may be difficult to interpret because of the small size of bile ducts and artifacts resulting from rapid breathing in infants (18).
To date, the nonoperative method of choice for differentiating EHBA from other forms of cholestasis not requiring early surgery is liver biopsy. Findings of varying degrees of portal tract fibrosis, edema, ductular proliferation, and bile plugs support EHBA, whereas evidence of giant cell transformation suggests other causes of neonatal hepatitis. Early liver biopsy, especially before 6 weeks of age, may not show typical features, and serial biopsy samples may be necessary for definite diagnosis (19). Equivocal histopathologic features suggest alternative conditions such as α1-antitrypsin deficiency, Alagille syndrome, neonatal sclerosing cholangitis, cystic fibrosis, or exposure to total parenteral nutrition, all of which can mimic EHBA (20–22). In our study, not all of the infants underwent liver biopsy because the referring centers did not have the ability to perform liver biopsy in such small infants. The age of referral did not allow further investigations that would postpone surgical intervention. Thus, these infants underwent ERCP and most had operations scheduled if needed within 24 hours.
Of all of the diagnostic procedures, intraoperative cholangiogram provides the final diagnosis for EHBA. If successful, ERCP provides similar data as cholangiography, and, as shown by our study and others’ (7,11,13), it causes minimal complications; however, it should be confined to specialized centers. Our study suggests that ERCP should not be performed in all infants with cholestasis, but rather in those infants whose diagnosis is questionable or in whom ERCP may have therapeutic use. Most of the infants with EHBA in our cohort (88%) were type 1 (main site of obliteration is the common bile duct and gallbladder is intact), whereas type 3 (complete obliteration of the entire biliary system) is regarded as the most common. Type 3 is easier to diagnose because US is usually abnormal. In 4 of the infants, ERCP precluded the need for laparotomy. In these infants, US was normal, but because of acholic stools with equivocal liver biopsy, there was a high suspicion of EHBA. These infants would have otherwise undergone explorative laparotomy. The high incidence of EHBA type 1 may be the result of referral bias because these infants were referred because of normal US, but a high clinical suspicion for EHBA persisted. Nevertheless, this finding raises the question of whether different genetic background leads to the different form of disease seen in other countries. This issue needs further investigation.
Cholelithiasis with cholestasis is another indication for ERCP in children and adults (23). Papillotomy is performed routinely in adults, but in infants it is uncommon (9,24). There is no consensus for managing those infants. Spontaneous resolution of symptoms may occur in 20% to 80% of infants (23,25). In our cohort, cholestasis caused by cholelithiasis improved 48 hours after ERCP, and none of the infants underwent further investigations or interventions. It is difficult to assess whether resolution of symptoms was spontaneous or attributed to the catheterization and contrast medium injection during the ERCP.
Our study has several limitations. First, it is a retrospective analysis with inherent limitations such as data acquisition and clinical follow-up. Second, because some patients were referred from other centers, long-term follow-up, especially in the infants with NC, was not always available. In these infants, diagnosis may evolve. Although this is a cohort of a single national center, the success rate is lower than that of other centers. This result emphasizes that ERCP in infants should be conducted in only highly trained and experienced centers with a high number of patients.
In conclusion, our study further supports the data that ERCP is feasible in infants with no major complications. ERCP should be used in infantile cholestasis when there is discordance between US and clinical features. When these indications were applied, ERCP precluded explorative laparotomy in 19% of cases and type 1 EHBA was diagnosed in the majority of cases.
We thank Dr Steven Werlin for critical review of this manuscript.
1. Balistreri WF. Neonatal cholestasis. J Pediatr
2. Manolaki AG, Larcher VF, Mowat AP, et al. The prelaparotomy diagnosis of extrahepatic biliary atresia
. Arch Dis Child
3. Waye JD. Endoscopic retrograde cholangiopancreatography in the infant. Am J Gastroenterol
4. Petersen C, Meier PN, Schneider A, et al. Endoscopic retrograde cholangiopancreaticography prior to explorative laparotomy avoids unnecessary surgery in patients suspected for biliary atresia
. J Hepatol
5. Jang JY, Yoon CH, Kim KM. Endoscopic retrograde cholangiopancreatography in pancreatic and biliary tract disease in Korean children. World J Gastroenterol
6. Shanmugam NP, Harrison PM, Devlin J, et al. Selective use of endoscopic retrograde cholangiopancreatography in the diagnosis of biliary atresia
younger than 100 days. J Pediatr Gastroenterol Nutr
7. Keil R, Snajdauf J, Rygl M, et al. Diagnostic efficacy of ERCP
in cholestatic infants
—a retrospective study on a large series. Endoscopy
8. Guelrud M, Jaen D, Mendoza S, et al. ERCP
in the diagnosis of extrahepatic biliary atresia
. Gastrointest Endosc
9. Guelrud M, Mendoza S, Jaen D, et al. ERCP
and endoscopic sphincterotomy in infants
and children with jaundice due to common bile duct stones. Gastrointest Endosc
10. Fox VL, Werlin SL, Heyman MB. Endoscopic retrograde cholangiopancreatography in children. Subcommittee on Endoscopy and Procedures of the Patient Care Committee of the North American Society for Pediatric Gastroenterology and Nutrition. J Pediatr Gastroenterol Nutr
11. Paris C, Bejjani J, Beaunoyer M, et al. Endoscopic retrograde cholangiopancreatography is useful and safe in children. J Pediatr Surg
12. Otto AK, Neal MD, Slivka AN, et al. An appraisal of endoscopic retrograde cholangiopancreatography (ERCP
) for pancreaticobiliary disease in children: our institutional experience in 231 cases. Surg Endosc
13. Vegting IL, Tabbers MM, Taminiau JA, et al. Is endoscopic retrograde cholangiopancreatography valuable and safe in children of all ages? J Pediatr Gastroenterol Nutr
14. Mieli-Vergani G, Howard ER, Portman B, et al. Late referral for biliary atresia
—missed opportunities for effective surgery. Lancet
15. Davenport M, Kerkar N, Mieli-Vergani G, et al. Biliary atresia
: the King's College Hospital experience (1974–1995). J Pediatr Surg
16. Gilmour SM, Hershkop M, Reifen R, et al. Outcome of hepatobiliary scanning in neonatal hepatitis syndrome. J Nucl Med
17. Kim CJ, Nam HS, Lee CY, et al. Catamenial hemoptysis: a nationwide analysis in Korea. Respiration
18. Lee MJ, Kim MJ, Yoon CS, et al. Gadopentetate dimeglumine-enhanced MR cholangiopancreatography in infants
with cholestasis. Pediatr Radiol
19. Azar G, Beneck D, Lane B, et al. Atypical morphologic presentation of biliary atresia
and value of serial liver biopsies. J Pediatr Gastroenterol Nutr
20. Nord KS, Saad S, Joshi VV, et al. Concurrence of alpha 1-antitrypsin deficiency and biliary atresia
. J Pediatr
21. Amedee-Manesme O, Bernard O, Brunelle F, et al. Sclerosing cholangitis with neonatal onset. J Pediatr
22. Shapira R, Hadzic N, Francavilla R, et al. Retrospective review of cystic fibrosis presenting as infantile liver disease. Arch Dis Child
23. Vrochides DV, Sorrells DL Jr, Kurkchubasche AG, et al. Is there a role for routine preoperative endoscopic retrograde cholangiopancreatography for suspected choledocholithiasis in children? Arch Surg
24. Thomas M, Kadiwar K, Domajnko B, et al. Choledocholithiasis in a 4-month-old infant. J Pediatr Surg
25. Keller MS, Markle BM, Laffey PA, et al. Spontaneous resolution of cholelithiasis in infants