Delayed Postnatal Presentation of Biliary Atresia in 2 Premature Neonates

Biliary atresia (BA) is the leading cause of neonatal cholestasis (1). It is the end result of a destructive process affecting both intrahepatic and extrahepatic bile ducts, leading to the obliteration of the biliary tract and eventually to secondary biliary cirrhosis. The outcome for infants with BA depends on early referral and timely Kasai operation. In experienced centers, up to 60% of children achieve biliary drainage after the Kasai operation and have normal serum total bilirubin levels within 6 months (1,2). Around 20% of children who underwent the Kasai operation will reach adulthood without undergoing liver transplantation (3). Although BA is the most frequent indication for liver transplantation in children, its pathogenesis remains poorly understood (1,2). Physiopathological hypotheses include genetic predisposition, dysimmunity, viral infection, and ischemia, but the cause is probably multifactorial, with obliterative cholangiopathy being the common endpoint (1). Two forms of BA have been described: a more common “perinatal” isolated form (90% of cases) and a less common “embryonic” form (10% of cases) with a high frequency of associated malformations, such as BA splenic malformation syndrome (BASM), characterized by splenic anomalies, situs inversus, cardiovascular defects, intestinal malrotation, preduodenal porta vein, and/or absence of inferior vena cava (1). These terms that imply an explicit assumption regarding the timing of the etiological cause are unadapted because evidence for biliary injury beginning early in the second trimester of gestation exists in the “perinatal” form (4,5). In the literature, few cases of BA have been reported in preterm infants and the clinical description was not fully detailed (6–8). We report on 2 cases of BA in premature babies born at 31 and 35 weeks of gestation who had initially normally colored stools. These cases illustrate that diagnosis of BA in preterm cholestatic neonates is difficult and requires a high index of suspicion and careful workup. PATIENT 1 A girl neonate weighing 2720 g was born at 35 weeks of gestation. There was no peculiar familial medical background and antenatal ultrasonographies were normal. She developed neonatal respiratory distress and hypoxia due to a hyaline membrane disease (appearance, pulse, grimace, activity, respiration [APGAR] score was not evaluated). Feeding with premature formula was tolerated and she did not receive parenteral nutrition. She developed jaundice soon after birth, but her stools were normally colored. She was referred to our unit for a diagnostic workup of neonatal jaundice and suspicion of cholestasis. At admission, 17 days after birth, she still was jaundiced, but stools remained normally colored. Liver was slightly enlarged and serum liver tests showed cholestasis (Table 1). Main causes of neonatal cholestasis other than BA were excluded (9,10). Abdominal ultrasonography (high-frequency transducer, 12 MHz) showed a small anechoic cyst (4 × 6 mm) located at the porta hepatis. The so-called ultrasonographic triangular cord was not observed (11). Bile ducts were not dilated and the main bile duct was not visible. The cyst was in communication with a linear anechoic structure corresponding to an extremely small and abnormal gallbladder with an irregular wall (Fig. 1A). This aspect was suggestive of BA, but stools remained normally colored. There was no sign of BASM. A magnetic resonance cholangiography was performed but did not provide any more valuable information (12). Progressively, stools discolored and became acholic 5 weeks after birth (corrected age: 1 day). A laparotomy was performed 6 weeks after birth (corrected age: 1 week). Intraoperative cholangiography was in favor of complete BA (type III) and a portoenterostomy was performed. The BA diagnosis was confirmed by the histological examination of the biliary remnant. Liver histology showed septal fibrosis and severe cholestasis with bile duct proliferation and bile ducts containing bile plugs. Stool color was normal 6 days after surgery. Serum bilirubin level decreased and normalized 76 days after surgery. The child received a symptomatic treatment for cholestasis including ursodeoxycholic acid. After 6 months, stools remained normal, as did the serum total bilirubin level. Three years after Kasai operation, the serum total bilirubin concentration was slightly increased (Table 1).TABLE 1: Evolution of serum liver tests in 2 preterm neonates with delayed postnatal development of biliary atresiaFIGURE 1: Liver color Doppler ultrasonography in preterm neonates who developed biliary atresia. A, In patient 1, born at 35 weeks of gestation, ultrasonography performed 17 days after birth showed a small anechoic cyst (4 × 6 mm) located at the porta hepatis. The cyst was in communication with a linear anechoic structure corresponding to an extremely small and abnormal gallbladder with an irregular wall. B, In patient 2, born at 31 weeks of gestation, ultrasonography performed 45 days after birth showed a small anechoic cyst (4 × 6 mm) located at the porta hepatis.PATIENT 2 A girl neonate weighing 1590 g was born at 31 weeks of gestation. There was no peculiar familial medical background, and antenatal ultrasonographies were normal. She developed neonatal respiratory distress and hypoxia due to a hyaline membrane disease. APGAR score was 6 at 1 minute and 8 at 5 minutes. Feeding with premature formula was tolerated and she did not receive parenteral nutrition. She was jaundiced 2 days after birth, but stools were normally colored. She was referred to our unit for diagnostic workup of neonatal jaundice and suspicion of cholestasis. At the admission, 45 days after birth, she still was jaundiced and stools were alternatively normally colored and slightly discolored. Liver was slightly enlarged and serum liver tests showed cholestasis (Table 1). Total serum bile acid concentration was increased (76 μmol/L). The main causes of neonatal cholestasis other than BA were excluded (9,10). Abdominal ultrasonography (high-frequency transducer, 12 MHz) showed a small anechoic cyst (4 × 6 mm) located at the porta hepatis (Fig. 1B). The ultrasonographic “triangular cord” was not observed (11). Bile ducts were not dilated and the main bile duct was not visible. The gallbladder was not visible. There was no sign of BASM. This aspect was suggestive of BA, but stools remained normally colored. A magnetic resonance cholangiography was performed, but did not provide any more valuable information (12). Progressively, stools discolored and became acholic 9 weeks after birth (corrected age: 1 day). A laparotomy was performed 10 weeks after birth (corrected age: 1 week). Intraoperative cholangiography was in favor of complete BA (type III) and a portoenterostomy was performed. BA diagnosis was confirmed by the histological examination of the biliary remnant. Liver histology showed septal fibrosis, cholestasis, and bile duct proliferation. Stool color was normal 8 days after surgery. Serum bilirubin level decreased and normalized 1 month after surgery. The child received a symptomatic treatment for cholestasis including ursodeoxycholic acid. After 6 months, stools remained normal, as did the serum total bilirubin level. Three years after the Kasai operation, the serum total bilirubin concentration was normal (Table 1). DISCUSSION In premature neonates, cholestasis is often multifactorial in origin and related to risk factors such as anoxia, ischemia, immaturity of bile acid metabolism/enterohepatic circulation, genetic predisposition, infection, lack of enteral feeding, necrotizing enterocolitis, and/or prolonged total parenteral nutrition, in the context of perinatal distress (9,10,13). Such cholestasis is usually spontaneously resolving within a few months and is classified as “transient neonatal cholestasis,” which implies that all of the other causes of neonatal cholestasis have been excluded (14). Its prognosis is good, and in most cases, stools are not acholic. Nevertheless, BA may be associated by chance with perinatal distress, and the latter should not be used as an argument to overdiagnose transient neonatal cholestasis. Our report illustrates this point, showing that in preterm neonates with cholestasis, in fact related to BA, discoloration of stools can occur several weeks after birth, whereas initial hepatic ultrasonographic findings may already indicate BA (15). In preterm children reported here, the first trap that has been avoided was to consider that the babies did not have cholestasis because initial conjugated bilirubin represented only around 15% (>17 μmol/L) of serum total bilirubin concentration and stools were yellow. In a jaundiced neonate with normally colored stools, it is a rule to consider cholestasis when the serum conjugated bilirubin concentration is either >17 μmol/L or >20% of serum total bilirubin concentration; all the more that other serum liver test parameters are abnormal, including γ-glutamyl transferase activity and total bile acid concentration (1,9,16). The second trap that has been avoided was to consider that these babies did not have BA because stools were initially normally colored, at a time corresponding to the last weeks of a full-term pregnancy. In both preterm babies, stools became acholic at a time corresponding to a normal term birth (corrected age: 1 day). This is in accordance with previous data, which showed that in normal-term babies with BA, stools became acholic within 2 weeks after birth (17). The third trap that has been avoided was to interpret favorably the decrease in total serum bilirubin concentration, whereas the proportion of conjugated bilirubin increased signing cholestasis. The fact that the main medical causes of neonatal cholestasis were excluded suggested BA, all the more that liver ultrasonography identified early signs highly suggestive of BA (9,10,15). Surgical and histological findings obtained at a corrected age of 1 week were typical of BA. Although the presence of early cholestasis in these preterm neonates with BA suggests that bile duct injury has started before or during the third trimester of pregnancy, the precise timing of the onset of the injurious process cannot be determined (4–6). It has been emphasized recently that increased age at Kasai operation has a negative effect on the long-term outcome of children with BA (2). Indeed, survival rates with native liver increased when age at surgery decreased, the best rate being obtained when the Kasai operation is performed before age 31 days. In preterm neonates reported here, surgery was performed early, at a corrected age of 1 week, and both children were alive with their native liver with almost normal or normal serum total bilirubin levels at age 3 years. Although neonates with BA may benefit from early Kasai operation, surgery should be performed after 2 weeks of life and complete umbilical cord healing because in our experience severe sepsis due to omphalitis may complicate surgery performed too early (unpublished data). These cases illustrate that diagnosis of cholestasis in preterm jaundiced neonates is difficult and that careful workup and follow-up is mandatory to detect BA. Acknowledgments We thank Drs Belzic (Lorient, France) and Suc (Tours, France) for referring the patients to our unit.

Récupéré en direct depuis OpenAlex et désinversé. Les résumés ne sont pas conservés dans cette base de données : les index inversés représentent 8,6 Go des 9,3 Go de texte de la base, et le serveur dispose de 13 Go libres.