Heritable disorders of the bile ducts

Gastroenterol Clin N Am 32 (2003) 857–875

Heritable disorders of the bile ducts Binita M. Kamath, MBBChir, David A. Piccoli, MD* Division of Gastroenterology and Nutrition, University of Pennsylvania School of Medicine 34th Street and Civic Center Boulevard, Philadelphia, PA 19104, USA

Diseases of the bile ducts encompass a wide range of disorders. These include those disorders primarily affecting extra and intrahepatic bile ducts and those that may be classified as panbiliary. A simple classification of bile duct disorders is represented in Box 1. This discussion focuses on heritable disorders of the bile ducts. For a discussion of so-called ‘‘isolated’’ or ‘‘sporadic’’ cystic and other bile duct disorders, the reader is referred elsewhere. As can be seen in Box 1, the major heritable bile duct disorders are those affecting the intrahepatic ducts, namely syndromic bile duct paucity, or Alagille syndrome (AGS), and the fibrocystic cholangiopathies autosomal recessive polycystic kidney disease (ARPKD)/congenital hepatic fibrosis (CHF), and autosomal dominant polycystic kidney disease (ADPKD). Therefore the remainder of the article will be limited to discussion on intrahepatic biliary disease. Embryology of the intrahepatic ducts It is not possible to fully appreciate disorders of intrahepatic bile ducts without a brief review of the embryology. The intrahepatic ducts develop primarily by a process of differentiation from the hepatocytes at the margins of the portal tracts. This differentiation results in the formation of the socalled ‘‘ductal plate,’’ which takes place in a centripetal fashion beginning from the hilus, through a process termed by Desmet [1] as ‘‘remodeling.’’ After completion of this process, the ductal plate disappears, leaving only the centrally located differentiated interlobular duct. The ductal plates make their first appearance in the 7th to 8th weeks of gestation, and a few persisting elements of the plates may be present at or beyond term [1]. Persistence of the ductal plate in the postnatal liver, accompanied by an increase in portal tract fibrous tissue, creates a lesion known as the ductal * Corresponding author. E-mail address: [email protected] (D.A. Piccoli). 0889-8553/03/$ - see front matter Ó 2003 Elsevier Inc. All rights reserved. doi:10.1016/S0889-8553(03)00054-2

858

B.M. Kamath, D.A. Piccoli / Gastroenterol Clin N Am 32 (2003) 857–875

Box 1. Bile duct disorders A. Intrahepatic bile duct disorders (1) Bile duct paucity (i) Syndromic bile duct paucity—Alagille syndrome (ii) Non-syndromic bile duct paucity (2) Cystic disorders (i) Solitary, sporadic cysts (ii) Polycystic, heritable conditions (iia) non-communicating cysts—ADPKD (iib) communicating cysts with DPM—ARPKD/CHF, malformation syndromes (iic) communicating cysts without DPM—Caroli disease B. Extrahepatic bile duct disorders (i) Anatomic anomalies—bile duct stenosis or perforation (ii) Choledochal cyst C. Panbiliary duct disorders (i) Biliary atresia ADPKD, autosomal dominant polycystic kidney disease; ARPKD/CHF, autosomal recessive polycystic kidney disease/congenital hepatic fibrosis; DPM, ductal plate malformation.

plate malformation [2], biliary dysgenesis, or congenital hepatic fibrosis. This lesion is found in combination with renal abnormalities (usually cysts) in several heritable conditions in which there is actual or potential cystic dilatation of the biliary ducts [3]. The interlobular ducts formed from the differentiation and remodeling of the ductal plate are joined by intrahepatic extensions of the extrahepatic ducts, (themselves derived from the cephalic portion of the hepatic diverticulum) to complete the bile duct system. The physiologic and biochemical factors governing the differentiation and remodeling of the ductal plate are essentially unknown at present, though the role of ductal-vascular interactions are increasingly being recognized. Understanding these factors may well be the key to understanding the genesis of duct paucity and cystic diseases of the liver. Intrahepatic cystic bile duct disorders Cystic diseases of the intrahepatic bile ducts present a wide range of disorders (Box 1). They include both sporadically occurring and heritable conditions, and extend from lesions typically discovered incidentally to frank malignancies. The distinction between communicating and

B.M. Kamath, D.A. Piccoli / Gastroenterol Clin N Am 32 (2003) 857–875

859

non-communicating cysts is clinically significant. Duct cysts that communicate with the biliary tree, they have a greater likelihood of causing clinical disease. Communicating duct cysts can be associated with cholangitis, stone formation, and (uncommonly) neoplasia. Non-communicating duct cysts are usually asymptomatic, but if sufficiently large, may present as an abdominal mass or biliary obstruction. Heritable diseases involving intrahepatic duct cysts are often associated with ductal plate malformation. As discussed later, this consists of plates or cisternae of duct elements characteristically found at the circumference of the portal tracts and is associated with increased portal tract fibrous tissue [2]. The relevance of this lesion to cystic bile ducts lies in the fact that the abnormal ducts have larger diameters than normal ducts and have a propensity to become dilated. Discussions of intrahepatic duct cystic dilatation must include mention of Caroli disease. Many radiologists and clinicians classify virtually all patients with intrahepatic duct dilatation as having ‘‘Caroli disease.’’ The authors believe this usage is imprecise and inadequately specific in view of the genetic and pathologic information that has accumulated since Caroli’s original papers. Most reported cases of ‘‘Caroli Disease’’ described since Caroli’s report seem to be associated with the ductal plate malformation, and thus seem to be examples of what we now know as congenital hepatic fibrosis (CHF) or autosomal recessive polycystic kidney disease (ARPKD), with prominent duct dilatation. This may rarely be found in autosomal dominant polycystic kidney disease (ADPKD). It has, however, been suggested that there is a group of cases distinct from ADPKD, CHF, and ARPKD by virtue of the level of ducts involved, which should be called Caroli’s Disease. Alternatively, the term Caroli Disease could be restricted to cases with no portal tract abnormality other than segmental duct dilatation, as in the ‘‘simple’’ form described by Caroli. The major heritable conditions characterized by intrahepatic bile duct cysts are ADPKD, also called ‘‘adult’’ polycystic disease, and ARPKD, which has formerly been termed infantile polycystic disease (IPCD). The latter is intimately related to, if not identical with CHF. There are also several heritable malformation syndromes characterized by potential bile duct cysts and renal disease. The ductal plate malformation is seen in all of these conditions (least commonly in ADPKD). Whenever the ductal plate malformation is the basis for the cysts, the cysts communicate proximally and distally with the biliary tree. Renal cysts of tubular origin or other renal developmental lesions are typically present in all these conditions. The renal lesions tend to be dissimilar in the different clinical conditions. ARPKD/CHF ARPKD/CHF is composed of a characteristic hepatopathology, cystic disease of the kidneys, portal hypertension, and an increased risk of

860

B.M. Kamath, D.A. Piccoli / Gastroenterol Clin N Am 32 (2003) 857–875

ascending cholangitis. The disorder has a proposed incidence of 1/20,000 live births and a variable clinical spectrum [4,5]. Pathology of ARPKD/CHF The hepatic lesion of ductal plate malformation is found in all cases of ARPKD/CHF. The renal lesion when identified in infancy is characterized by radially arranged tubular cysts occupying most of the large externally smooth renal mass with widely spaced glomeruli. The longer patients survive, the less characteristic the renal lesions become, because the cysts become more rounded [6]. The pathogenesis of the renal lesion is at least partially understood. A variety of animal models have been developed to investigate the mechanisms of cystogenesis in the kidneys. Evidence from several studies has suggested a role for the epidermal growth factor-a (EGFR) and transforming growth factor-a (TGF) receptor axis in promoting epithelial hyperplasia and subsequent renal cyst formation in murine and human ADPKD and ARPKD [7]. Furthermore, similar abnormalities of EGFR expression have been suggested to mediate biliary epithelial hyperplasia and ductal ectasia in a mouse model [8,9]. An inhibitor of tyrosine EGFR tyrosine kinase activity has recently been shown to markedly reduce collecting tubule cystic lesions, improve renal function, decrease biliary epithelial abnormalities and improve life span in a mouse model of ARPKD [10]. Genetics of ARPKD/CHF Mutations in the polycystic kidney and hepatic disease 1 gene (PKHD1) on chromosome 6p21.1-p12 have recently been identified as the molecular cause of ARPKD/CHF [11,12]. Despite a variety of clinical manifestations of ARPKD, linkage data suggests a single locus for the disorder with no clear evidence of genetic heterogeneity. The gene was identified by two independent groups, one studying the polycystic kidney rat model and the other using positional cloning techniques [11,12]. The gene encompasses 470 kilobases and at least 86 exons [12]. The predicted protein product has been termed polyductin and is believed to represent a novel integral membrane protein with a large highly glycosylated extracellular portion, a single transmembrane-spanning domain, and a short intracytoplasmic tail. The function of polyductin has yet to be determined, though it shares extracellular domains with hepatocyte growth factor receptor and the plexin superfamily [5]. The identification of PKHD1 has little enhanced our understanding of the pathogenesis of ARPKD/CHF. Mutations have been detected in approximately 60% of affected individuals and further mutation detection is believed to be limited by technical difficulties in screening this large gene. To date some 63 different PKHD1 mutations have been identified and these are scattered throughout the gene with no clustering at specific sites. Most of the mutations found are unique to a particular pedigree. Only one locus on exon

B.M. Kamath, D.A. Piccoli / Gastroenterol Clin N Am 32 (2003) 857–875

861

3, termed the T36M mutation, may be a mutational hotspot. Approximately 45% mutations are predicted to be protein truncating [5]. It had been difficult to establish clear genotype-phenotype correlations in ARPKD/CHF. It does seem, however, that individuals with two proteintruncating mutations have a more severe phenotype with early death in the perinatal or neonatal period [5]. Relationship between CHF and ARPKD The relationship of ARPKD to CHF is still somewhat controversial. The hepatic conditions in both lesions are essentially similar, as they all consist of the ductal plate malformation. The renal lesions, which consist of tubular cysts in both, classically differ markedly in both pathology and clinical severity. In newborn patients with ARPKD, the renal lesions are diffuse and prominent clinically, whereas in patients who exhibit the clinical picture of CHF the renal lesions are often not as evident in early life and are minor. However, with long survival of patients with ARPKD the lesions become increasingly similar. This suggests that the two conditions are actually one disorder with the apparent differences being related in part to the length of survival of the patients or the variable expression of the abnormalities in the affected organs. Evidence contrary to this unitary point of view lies in the observation that the clinicopathologic presentations tend to ‘‘breed true’’ within a given family [13], although variability has been described in some families. Thus, there is evidence both for and against the pathogenic identity of CHF and ARPKD that can only be resolved by identification of the molecular defects responsible for these disorders. For the purpose of this presentation, they will be treated as a single entity, with different clinicopathologic presentations. Clinical presentation of ARPKD/CHF The clinical manifestations of ARPKD/CHF vary according to the age at first presentation. The renal disease predominates in neonates and infants (ARPKD), whereas the hepatic-related disease predominates in older children and adults (CHF). The clinical profile has been divided into four groups at presentation: perinatal, neonatal, infantile, and juvenile groups [13]; but these subdivisions seem to have no genetic or pathogenetic implications, and in our view serve no useful purpose. Renal disease in ARPKD The renal disease may vary from an incidental finding in older children to a major cause of early mortality. It has been estimated that 30% to 50% of infants affected with ARPKD will die in the perinatal period [14], although recent studies suggest a better long-term prognosis [15]. In infants who present with the renal manifestations of infantile polycystic disease the

862

B.M. Kamath, D.A. Piccoli / Gastroenterol Clin N Am 32 (2003) 857–875

kidneys are enlarged and severely dysfunctional. They may be palpable on examination, and an abdominal radiograph will demonstrate bilaterally enlarged kidneys. Excretory urography may only poorly visualize the collecting system. The ‘‘nephrogram’’ (characteristic of the neonatal presentation) demonstrates a radiolucent mottled parenchyma because of the cystic changes of the nephrons. Many infants with ARPKD develop uremia and chronic renal failure. Respiratory distress occurs from compression exerted by the enlarged kidneys, fluid retention, congestive heart failure, concomitant pulmonary hypoplasia, or pneumonia. Progressive renal failure and hypertension may occur over the first few weeks or months of life. Mortality is high in these patients. In contrast, those who survive the first month of life generally do quite well [4,15]. In these children the hepatic fibrosis may be progressive but is rarely a clinically important factor. Renal disease in CHF In patients who present with the hepatic manifestations of CHF, palpable kidneys are often noted at initial evaluation, occasionally in association with arterial hypertension [16,17]. The intravenous pyelogram demonstrates enlarged kidneys and tubular ectasia with alternating dense and radiolucent streaks radiating from the medulla to the cortex. Renal dysfunction is present in approximately twenty percent of patients as evidenced by decreased maximal concentrating capacity, an elevated serum BUN, and a chronic mild metabolic acidosis [18]. Even in some patients with a normal IVP initially, there may be an evolution to the typical radiographic findings in later life and in most cases cysts are present on pathological evaluation. Hepatic disease in CHF Portal hypertension In the older patient with ARPKD/CHF, the most significant abnormality is portal hypertension. The precise pathogenesis is unknown, but is thought to be associated with the hepatic fibrosis or portal vein abnormalities. Clinically, hematemesis or melena is the presenting sign in 30% to 70% of patients from pediatric and mixed population studies [17,19]. In children the age for presentation of hematemesis may be as early as the first year of life [20], but it usually ranges from 5 to 13 years. Firm or hard hepatomegaly is present in nearly all patients, often with a prominent left lobe, and this is usually one of the presenting findings. Splenomegaly occurs in most, accompanied by hypersplenism with thrombocytopenia. Portal vein abnormalities, characteristically duplication of the intrahepatic branches are common [21]. Biliary lesions and ascending cholangitis Dilatation of the intrahepatic ducts is common in this condition [22,23], as is an increased risk for cholangitis [16,24–26]. Although this dilatation is

B.M. Kamath, D.A. Piccoli / Gastroenterol Clin N Am 32 (2003) 857–875

863

occasionally noted macroscopically on CT or ultrasound evaluation, magnetic resonance cholangiopancreatography seems to be an excellent imaging modality for diffuse or focal ductal dilatation [22], although it has not been proven to predict risk for subsequent cholangitis. Cholangitis may be occult, acute or chronic in nature, and contributes significantly both the morbidity and mortality of congenital hepatic fibrosis. Vascular abnormalities In addition to the duplication of the intrahepatic portal venous system, other vascular abnormalities and congenital heart disease [27], are recognized associations. These include cerebral [28,29], hepatic, splenic and renal aneurysms [23], and cerebellar hemangioma. Diagnosis of ARPKD/CHF The diagnosis in a patient with hepatomegaly or portal hypertension is suggested on clinical and radiographic observations. The liver is usually enlarged and quite firm, with a prominent left lobe. The spleen, and occasionally the kidneys, are palpable. In most patients the biochemical parameters of hepatic synthetic function are normal. There may be a mild elevation of the transaminases in some cases, but the bilirubin is usually normal. The white blood count, sedimentation rate, and globulin level should be determined as evidence of chronic cholangitis. An elevated BUN, creatinine, or decreased creatinine clearance provides evidence for renal involvement. The initial radiographic evaluation should be an ultrasound with Doppler evaluation of the portal vasculature. Evidence of portal hypertension, splenomegaly, and intense hepatic echogenicity support the diagnosis. Evidence for duplication of the intrahepatic vasculature is also confirmatory. The renal ultrasound may show increased size and echogenicity of the kidneys. An intravenous pyelogram will confirm the diagnosis in most cases, but may not be necessary. Percutaneous liver biopsy will show ductal plate malformation in most patients, although a few older patients will have hepatic fibrosis without obvious biliary dysgenesis. It is important to culture all liver specimens for bacterial pathogens, in addition to evaluating the tissue for evidence of cholangitis. Particularly in the older patients, the demonstration by biopsy (or otherwise) of cystic renal disease is helpful in establishing the diagnosis. Therapy for CHF Portosystemic shunting has been the treatment of choice, as there is a low incidence of postoperative encephalopathy or hyperammonemia. Prospective trials of other alternative approaches, such as sclerotherapy or pharmacological management of varices are not yet available. Nevertheless, the presence of spontaneous portosystemic shunts in some children suggest that sclerotherapy may be beneficial if it can be shown to hasten the

864

B.M. Kamath, D.A. Piccoli / Gastroenterol Clin N Am 32 (2003) 857–875

development of hemodynamically significant shunts without surgery. If surgery is selected as the treatment for portal hypertension, the type of shunt should be carefully chosen to prevent the limitation of options for either hepatic or renal transplantation in later life. Prolonged cholangitis is a major complication and has been responsible for hepatic failure and death. Therefore, unexplained fever or serologic evidence of inflammation, even in the absence of fever, warrant a diagnostic liver biopsy and aspirate for culture. Manipulation of the extrahepatic biliary tree carries an increased risk of infection in patients with abnormal ducts or bile stasis [30]. In cases of refractory cholangitis, surgical management and external or internal drainage may be necessary to resolve the hepatobiliary infection. In patients with stasis and refractory cholangitis a choleretic agent may significantly augment therapy. Ursodeoxycholic acid therapy and prophylactic antibiotic administration have not been adequately studied in CHF-related cholangitis, but may have a role in selected patients. Prognosis for CHF In general, the prognosis for those older children who present with CHF is good. The limitations are those imposed by complications of the disease, namely portal hypertension, cholangitis, and occasionally, renal or hepatic failure. Chronic renal failure is most common in patients with a presentation in infancy. As noted, portal hypertension is usually successfully managed and rarely complicated by hepatic encephalopathy. Ascending cholangitis with sepsis and hepatic failure is a major cause of death in most series. In those patients with chronic cholangitis or progressive hepatic dysfunction, liver transplantation may prove to be the optimal therapy. Occasionally, patients have received combined renal and hepatic transplantations for multi-organ failure. ADPKD ‘‘Adult’’ polycystic disease can be anatomically identified even in fetal life. It is important to recognize for its genetic implications, even though the functional significance of the finding is not apparent until beyond childhood. The hepatic lesions are primarily duct cysts that are readily demonstrated ultrasonographically. Cysts increase in size from childhood until 40 to 50 years of age. They are recognized and are perhaps present at an earlier age in women than in men. Commonly, the cysts in this condition are dilated ductal elements that are not demonstrated to communicate with the distal biliary tree. However, there may also be portal tract lesions consistent with the (communicating) ductal plate malformation in a smaller percentage of patients [31]. The renal lesion consists of cysts that seem to arise from multiple areas along the nephron and increase in size with age, eventuating in the kidneys and becoming large cystic reniform masses with inadequate numbers of functioning nephrons [32].

B.M. Kamath, D.A. Piccoli / Gastroenterol Clin N Am 32 (2003) 857–875

865

Cysts may also be found in other organs, including spleen, pancreas, thyroid, ovary, endometrium, seminal vesicles, and epididymis. Arterial aneurysms are present in 15% of cases. Genetics of ADPKD ADPKD results from mutations in one of at least three distinct genetic loci, termed PKD1, PKD2, and PKD3. PKD1 has been mapped to chromosome 16p13.3, and is responsible for most ADPKD in Caucasians. The gene responsible for this form of ADPKD has been identified [33,34], and prenatal diagnosis is thus possible. The gene product of PKD1, polycystin-1, is thought to mediate cellular interactions [35]. A second ADPKD gene, PKD2, has been localized to chromosome 4q13-4q23, and the mutant gene has been identified [36]. The PKD2 gene product has been termed polycystin-2 and most likely represents an endoplasmic reticulumbound cation channel [35]. The onset of renal disease in PKD2 occurs later in life than PKD1. Cyst formation in ADPKD is believed to result from a two-hit mechanism in which there is a germline mutation in PKD1 or PKD2 and an additional somatic mutation. This view is supported by mouse knockout models [35]. There is also evidence for a third locus for ADPKD, termed PKD3 [37].

Other malformation syndromes The potential for bile duct cyst formation is also present in several malformation syndromes of which the ductal plate malformation is a part. Landing has demonstrated morphometric differences between some of these and ARPKD and CHF [38]. These syndromes are said to include Meckel Syndrome, Ivemark Syndrome, Zellweger Syndrome, Jeune Syndrome, Elejalde Syndrome, Glutaric aciduria Syndrome Type II, Majewski syndrome, Robert syndrome, Saldino–Noonan syndrome, Smith–Lemli– Opitz [39] syndrome, and Trisomy 9 and l3 syndromes. Most of these are heritable conditions. Most of these malformation syndromes also have cystic renal disease, usually renal dysplasia, as a component. The pathogenetic implications of this coexistence of renal and hepatic cysts in these malformation syndromes and in ADPKD and ARPKD/CHF is not clear, particularly since the renal disease varies considerably in character among the various conditions.

Bile duct paucity Decrease in ductal number (paucity) is one of the most significant abnormalities of the intralobular bile ducts in children. Bile duct paucity can only be defined histologically. In patients at or beyond 37 weeks gestational

866

B.M. Kamath, D.A. Piccoli / Gastroenterol Clin N Am 32 (2003) 857–875

age, paucity is present when histologic examination demonstrates that ratio of ducts to portal tracts is less than 0.9.

Alagille syndrome—syndromic bile duct paucity Alagille syndrome (AGS) is a multi-system developmental disorder characterized by a marked reduction in the number of the interlobular bile ducts and cholestasis, occurring in association with cardiac, musculoskeletal, ocular, facial, renal, pancreatic, and neurodevelopmental abnormalities [40]. It is a familial disease with a wide variability in its clinical spectrum, even within individual pedigrees. The disease gene in AGS has been identified as Jagged1 (JAG1) on chromosome 20p12 [41,42]. Although bile duct paucity has been considered to be the most important and constant feature of AGS, paucity is not uniformly present. Furthermore, paucity is not present on the initial biopsy in many infants with AGS. In addition, ductular proliferation is present in a small number of infants with AGS, leading to potential diagnostic confusion with biliary atresia. Newer studies only identify paucity in 80% to 85% of AGS patients [43,44]. Several studies of serial liver biopsies have demonstrated that paucity is more common in older infants and children [45]. The progression to paucity typically accompanies a worsening of clinical hepatic disease in infancy over a period of months or years which is a characteristic of AGS. Hypotheses for the cause of progression to paucity in AGS include either a destruction of ducts post-natally, or a differential maturation of portal tracts and their associated ducts. The factors that lead to the decrease in the number of ducts are not yet understood. With the realization that AGS is caused by a defect in JAG1, which functions as a ligand in the Notch signaling pathway, the authors can infer that this pathway plays a key role in duct development. Genetics of Alagille syndrome AGS is a dominant genetic disorder with variable expressivity ranging from sub-clinical to severely affected individuals [46,47]. In 1997, mutations in the JAG1 gene were shown to be the cause of AGS [41,42]. JAG1 is a cell surface protein that functions as a ligand for the Notch transmembrane receptors, and these receptor–ligand pairs are part of the evolutionarily conserved Notch signaling pathway. Notch receptors appear on the cell surface as two associated peptides, one extracellular and the other consisting of the transmembrane segment and intracellular domain. On stimulation of the receptor by ligand (such as JAG1), the intracellular domain translocates into the nucleus where it mediates downstream effects in conjunction with intracellular regulatory proteins. The molecular outcome of the initiation of Notch signaling is the transcription of Notch sensitive genes, which then act to regulate cellular differentiation.

B.M. Kamath, D.A. Piccoli / Gastroenterol Clin N Am 32 (2003) 857–875

867

To date, over 230 AGS probands have been studied molecularly, and JAG1 mutations have been demonstrated in approximately 70% [48]. Among these reported patients, 4% had a deletion of the entire JAG1 gene, 49% had protein truncating mutations (frameshift and nonsense), 9% had splicing mutations, 9% had missense mutations and no mutation could be identified in 31% [48]. The mutations are distributed across the entire coding region of the JAG1 gene. Fifty to 70% of mutations are de novo. Mutations in JAG1 could cause AGS either by inducing haploinsufficiency of the JAG1 protein, or by causing a dominant negative effect. Under a model of haploinsufficiency an alteration in one of the JAG1 genes leads to a complete lack of product or a severely defective product, resulting in insufficient protein. Genes showing haploinsufficiency code for products that are needed in specific tissues in large quantity, such that having only one functional copy does not allow enough of the protein product to be made. The fact that large deletions of 20p12, including the entire JAG1 gene cause AGS is good evidence that some cases of AGS are caused by haploinsufficiency. A second potential mechanism for a dominant disease is that of a dominant negative effect. In this case, the mutant protein antagonizes the activity of the remaining wild type protein, so that normal function of the gene is obliterated. While haploinsufficiency for JAG1 is certainly the mechanism of disease causation in patients with total gene deletions and some missense mutations, a dominant negative mechanism cannot be rule out in some cases of missense mutations or prematurely truncated proteins. As stated earlier, the AGS phenotype demonstrates variable expressivity with clinical effects ranging from sub-clinical to life threatening. The only reliable indicator of mortality has been the presence of complex heart disease. It is not possible to predict which patients will progress to end stage liver disease. It had been hoped that genetic studies would reveal a correlation between the genetic mutation and phenotype, however this has not been found to be the case in patients with AGS. Even within families segregating a single mutation, the expressivity of the disorder has been found to range from mild to severe [46,47]. A possible exception to the lack of genotype–phenotype correlation was recently reported. A large family in which apparently isolated heart disease was segregating (most commonly tetralogy of Fallot) was analyzed by linkage analysis, and the disease locus was found to map to 20p12 in the vicinity of JAG1 [49]. Analysis of JAG1 in this family revealed that eleven individuals with heart disease had a missense mutation in exon 6. This mutation lies in a region of the EGF repeats that has been shown to be necessary for proper receptor-ligand interaction. This missense mutation has not yet been identified in a patient who meets the clinical criteria for AGS, nor has it been demonstrated in controls. This family is remarkable in that none of the mutation positive individuals were reported to have extracardiac manifestations. As this is the first time that a genotype–phenotype

868

B.M. Kamath, D.A. Piccoli / Gastroenterol Clin N Am 32 (2003) 857–875

correlation has been observed with JAG1 mutations, this mutation may act by a unique mechanism. Gene testing for AGS is currently available. Evaluation by fluorescent insitu hybridization for deletions at 20p12 identifies only 5% to 7% of patients [50]. Molecular testing is available in several centers, but is difficult because of the large size of the gene, and the lack of significant mutational hot spots. Most mutations identified to date are unique, and scattered widely across the gene. In addition, current screening techniques cannot identify a mutation in approximately 30% of clearly affected probands [48]. When a defect can be identified in a proband, however, it is then easy and useful to identify relatives who are minimally affected yet carry a significant risk to their offspring. Prenatal testing is available if the mutation is identified in the proband. Molecular testing has also aided in the diagnosis of AGS for patients with very minor or atypical manifestations. Given the large number of patients in whom no mutation has yet been identified, it remains possible that other Notch/ligand gene defects, or other unrelated genes may account for some patients diagnosed clinically as AGS. Clinical manifestations of AGS Hepatic manifestations Jaundice is present in most symptomatic patients, and presents as a conjugated hyperbilirubinemia in the neonatal period. It must be distinguished from biliary atresia. In half of these infants jaundice is persistent, resolving only in later childhood. The magnitude of the hyperbilirubinemia is minor compared with the degree of cholestasis. Cholestasis is manifest by pruritus, which is among the most severe in any chronic liver disease. It is rarely present before 3 to 5 months of age, but is seen in nearly all children by the third year of life even in those who are anicteric. The presence of severe cholestasis results in the formation of xanthomas, characteristically on the extensor surfaces of the fingers, the palmar creases, nape of the neck, popliteal fossa, buttocks, and around inguinal trauma sites. The lesions persist throughout childhood but may gradually disappear after 10 years of age. The timing for the formation of xanthomas correlates with a serum cholesterol greater than 500 mg/dL. Hypercholesterolemia and hypertriglyceridemia may be profound reaching levels exceeding 1000 mg/mL and 2000 mg/mL respectively. The most common laboratory abnormalities are elevations of serum bile aids, conjugated bilirubin, alkaline phosphatase, and c-glutamyl transpeptidase, which suggest a defect in biliary excretion in excess of the abnormalities in hepatic metabolism or synthesis. There are elevations of the serum aminotransferases, up to 10-fold, though hepatic synthetic function is usually well preserved. Nevertheless, progression to cirrhosis and hepatic failure, is recognized in approximately 20% of AGS patients.

B.M. Kamath, D.A. Piccoli / Gastroenterol Clin N Am 32 (2003) 857–875

869

Diminished bile salt excretion and low intraluminal bile salt concentrations result in ineffective solubilization and absorption of dietary lipid, essential fatty acids and fat soluble vitamins. Cardiovascular manifestations. A wide range of cardiovascular abnormalities have been reported in patients with AGS [43,51,52]. An audible murmur is present at some time in 97% of patients [43]. The most common lesions are pulmonary artery stenoses at various sites in the proximal and distal tree. Of the intracardiac lesions, tetralogy of Fallot is the most common (7%–9%). Other lesions include truncus arteriosus, secundum atrial septal defect, patent ductus arteriosus, ventriculoseptal defects, and pulmonary atresia. In a large series of patients, Emerick et al [43] demonstrated that only cardiac disease predicted increased mortality in AGS patients. Characteristic facies. The facies in AGS consist of a prominent forehead, moderate hypertelorism with deep set eyes, a small pointed chin, and a saddle or straight nose. The usefulness of the facies as a major criterion for diagnosis of AGS has been challenged because of interobserver differences. It has been suggested that these facies are a common result of cholestasis [53], but recent evidence supports the idea of AGS-specific facies [54]. Vertebral and musculoskeletal abnormalities. The most characteristic finding is the sagittal cleft or butterfly vertebrae. This is an uncommon anomaly that may occur in normal individuals. The affected vertebral bodies are split sagittally into paired hemivertebrae, because of failure of the fusion of the anterior arches of the vertebrae. Generally these are asymptomatic and of no structural significance. Ocular abnormalities. Several abnormalities have been described in AGS. A few of the findings are secondary to chronic vitamin deficiencies. Of the primary ocular abnormalities posterior embryotoxon is the most important diagnostically. Posterior embryotoxon is a prominent, centrally positioned Schwalbe’s ring (or line), at the point where the corneal endothelium and the uveal trabecular meshwork join. Posterior embryotoxon occurs in up to 89% of patients with AGS, but it also occurs in 8% to 15% of normal eyes. Vascular anomalies. Vascular anomalies have been noted in AGS from some of the earliest descriptions of this syndrome [55,56]. The literature documents multiple case reports of intracranial vessel abnormalities and other vascular anomalies in AGS. The latter include involvement of the renal, abdominal aorta, celiac, superior mesenteric, and subclavian arteries [57–60]. Unexplained intracranial bleeding is a recognized complication and cause of mortality in AGS. Intracranial bleeds accounted for 25% of the mortality

870

B.M. Kamath, D.A. Piccoli / Gastroenterol Clin N Am 32 (2003) 857–875

seen in AGS in a large series of patients [43]. Intracranial bleeds are seen in up to 16% of patients [61]. There does not seem to be any pattern to the location or severity of the intracranial bleeding which ranges from massive fatal events to asymptomatic cerebral infarcts. Underlying central nervous system vascular abnormalities have been described in some AGS individuals, which could explain the intracranial events [62,63]. Moyamoya syndrome, another form of vasculopathy, has also been reported in children with AGS [64].

Additional manifestations. Structural and functional renal abnormalities occur in 40% to 50% of patients with AGS. These include solitary kidney, ectopic kidney, bifid pelvis, reduplicated ureters, and multicystic and dysplastic kidneys. Renal artery stenosis is a cause of systemic hypertension in AGS. Tubulointerstitial nephropathy, renal tubular acidosis, a characteristic ‘‘lipodosis’’ of the glomeruli, and adult-onset renal insufficiency may also occur. Severe growth retardation is common in patients with AGS, particularly in the first 4 years of life. Malnutrition caused by malabsorption is a major factor in this failure to thrive. There seem to be limitations in linear growth even when protein-calorie malnutrition is not evident. Intrinsic pancreatic disease may also occur in AGS. Clinically, the identification of pancreatic insufficiency is important, as therapy with enzyme supplementation is available.

Diagnosis of AGS Traditionally the diagnosis of AGS requires histologic bile duct paucity associated with at least three of five major criteria: cholestasis, characteristic facies, vertebral anomalies, ocular anomalies and cardiac disease. Since then a wide range of clinical findings have been associated with AGS. Renal disease, pancreatic disease and intracranial vascular events are now recognized to be significant manifestations of AGS. Large series of patients [43,44,55,61] have demonstrated differing frequencies of the manifestations of AGS. As a result of molecular testing, it is now common to identify family members who would not meet clinical criteria for AGS, but who have the same JAG1 mutation as a severely affected relative. Because of the successful identification of JAG1 mutations in over 200 families with AGS, the diagnostic criteria for AGS can be modified significantly. For the index case (proband) in a family, it seems reasonable to apply the original Alagille criteria, requiring bile duct paucity with at least three features from the list of cholestasis, characteristic facies, posterior embryotoxon, butterfly vertebrae, and typical AGS renal disease or cardiac disease. In infants less than 6 months of age, when paucity is commonly absent, three or four clinical features should be adequate to make the diagnosis. In families

B.M. Kamath, D.A. Piccoli / Gastroenterol Clin N Am 32 (2003) 857–875

871

with one definite proband, other members with two, and even one feature are likely to have the gene defect, (and thus should be considered to have AGS). Treatment Infants with intrahepatic cholestasis may have significant fat malabsorption. Optimal diets include increased amounts of medium chain triglycerides added to the diet. Fat soluble vitamin deficiency is present to a variable degree in most patients with bile duct paucity necessitating oral or parenteral supplementation. Pruritus is the most significant symptom for many patients with chronic cholestasis. Antihistamines may give some relief, and care should be taken to keep the skin hydrated with emollients. Cholestyramine may improve pruritus, but some children will develop a severe acidosis on this therapy. Rifampin, which inhibits uptake of bile acids into the hepatocyte, also seems to provide significant relief of pruritus. Ursodeoxycholic acid, a potent choleretic, may have a dramatic effect of reducing symptomatic cholestasis, although in some patients it seems to exacerbate pruritus. Recently biliary diversion has been successfully undertaken to relieve intractable pruritis [65]. Hepatic transplantation may be required for chronic liver failure, portal hypertension, or severe intractable pruritus. Survival following transplantation has varied significantly in different studies, from 45% to 100% [66,67]. Transplantation does seem to have a higher risk for patients with AGS, due in part to the severity of cardiopulmonary disease. Caution should be taken when considering relatives as potential donors for living related transplantation, as unsuspected disease in the parent has thwarted donation [68]. Prognosis The outcome of syndromic bile duct paucity is variable and is most directly related to the severity of the hepatic and the cardiac lesions, with mortality predominantly attributable to these two organs. Complex congenital cardiac disease is a major cause of early mortality, while hepatic complications account for most of the later morbidity and mortality. The predicted probability of survival to 20 years of age for all patients is 75% [43]. The probability of survival to age 20 is 80% for patients who do not require liver transplantation and 60% in those who underwent transplantation.

References [1] Desmet VJ. Congenital diseases of intrahepatic bile ducts: variations on the theme ‘‘ductal plate malformation.’’ Hepatology 1992;16(4):1069–83. [2] Jorgensen MJ. The ductal plate malformation. Acta Pathol Microbiol Scand 1977; (Suppl 257):1–87.

872

B.M. Kamath, D.A. Piccoli / Gastroenterol Clin N Am 32 (2003) 857–875

[3] Bernstein J. Hepatic involvement in hereditary renal syndromes. Birth Defects Orig Artic Ser 1987;23(1):115–30. [4] Zerres K, Rudnik–Schoneborn S, Deget F, Holtkamp U, Brodehl J, Geisert J, et al. Autosomal recessive polycystic kidney disease in 115 children: clinical presentation, course and influence of gender. Arbeitsgemeinschaft fur Padiatrische, Nephrologie. Acta Paediatr 1996;85(4):437–45. [5] Bergmann C, Senderek J, Sedlacek B, Pegiazoglou I, Puglia P, Eggermann T, et al. Spectrum of Mutations in the Gene for Autosomal Recessive Polycystic Kidney Disease (ARPKD/PKHD1). J Am Soc Nephrol 2003;14(1):76–89. [6] Lieberman E, Salinas–Madrigal L, Gwinn JL, Brennan LP, Fine RN, Landing BH. Infantile polycystic disease of the kidneys and liver: clinical, pathological and radiological correlations and comparison with congenital hepatic fibrosis. Medicine [Baltimore] 1971;50(4):277–318. [7] Richards WG, Sweeney WE, Yoder BK, Wilkinson JE, Woychik RP, Avner ED. Epidermal growth factor receptor activity mediates renal cyst formation in polycystic kidney disease. J Clin Invest 1998;101(5):935–9. [8] Nauta J, Ozawa Y, Sweeney WE Jr, Rutledge JC, Avner ED. Renal and biliary abnormalities in a new murine model of autosomal recessive polycystic kidney disease. Pediatr Nephrol 1993;7(2):163–72. [9] Nauta J, Sweeney WE, Rutledge JC, Avner ED. Biliary epithelial cells from mice with congenital polycystic kidney disease are hyperresponsive to epidermal growth factor. Pediatr Res 1995;37(6):755–63. [10] Sweeney WE, Chen Y, Nakanishi K, Frost P, Avner ED. Treatment of polycystic kidney disease with a novel tyrosine kinase inhibitor. Kidney Int 2000;57(1):33–40. [11] Ward CJ, Hogan MC, Rossetti S, Walker D, Sneddon T, Wang X, et al. The gene mutated in autosomal recessive polycystic kidney disease encodes a large, receptor-like protein. Nat Genet 2002;30(3):259–69. [12] Onuchic LF, Furu L, Nagasawa Y, Hou X, Eggermann T, Ren Z, et al. PKHD1, the polycystic kidney and hepatic disease 1 gene, encodes a novel large protein containing multiple immunoglobulin-like plexin-transcription-factor domains and parallel beta-helix 1 repeats. Am J Hum Genet 2002;70(5):1305–17. [13] Blyth H, Ockenden BG. Polycystic disease of kidney and liver presenting in childhood. J Med Genet 1971;8(3):257–84. [14] Guay–Woodford LM, Galliani CA, Musulman–Mroczek E, Spear GS, Guillot AP, Bernstein J. Diffuse renal cystic disease in children: morphologic and genetic correlations. Pediatr Nephrol 1998;12(3):173–82. [15] Roy S, Dillon MJ, Trompeter RS, Barratt TM. Autosomal recessive polycystic kidney disease: long-term outcome of neonatal survivors. Pediatr Nephrol 1997; 11(3):302–6. [16] Alvarez F, Hadchouel M, Bernard O. Latent chronic cholangitis in congenital hepatic fibrosis. Eur J Pediatr 1982;139(3):203–5. [17] Kerr DN, Okonkwo S, Choa RG. Congenital hepatic fibrosis: the long-term prognosis. Gut 1978;19(6):514–20. [18] Anand SK, Chan JC, Lieberman E. Polycystic disease and hepatic fibrosis in children. Renal function studies. Am J Dis Child 1975;129(7):810–3. [19] Alvarez F, Bernard O, Brunelle F, Hadchouel M, Leblanc A, Odievre M, et al. Congenital hepatic fibrosis in children. J Pediatr 1981;99(3):370–5. [20] Fiorillo A, Migliorati R, Vajro P, Caldore M, Vecchione R. Congenital hepatic fibrosis with gastrointestinal bleeding in early infancy. Clin Pediatr [Phila] 1982;21(3):183–5. [21] Odievre M, Chaumont P, Montagne JP, Alagille D. Anomalies of the intrahepatic portal venous system in congenital hepatic fibrosis. Radiology 1977;122(2):427–30. [22] Jung G, Benz–Bohm G, Kugel H, Keller KM, Querfeld U. MR cholangiography in children with autosomal recessive polycystic kidney disease. Pediatr Radiol 1999;29(6):463–6.

B.M. Kamath, D.A. Piccoli / Gastroenterol Clin N Am 32 (2003) 857–875

873

[23] Murray–Lyon IM, Shilkin KB, Laws JW, Illing RC, Williams R. Non-obstructive dilatation of the intrahepatic biliary tree with cholangitis. Q J Med 1972;41(164):477–89. [24] Howlett SA, Shulman ST, Ayoub EM, Alexander RA, Donnelly WH, Cerda JJ. Cholangitis complicating congenital hepatic fibrosis. Am J Dig Dis 1975;20(8):790–5. [25] Murray–Lyon IA, Shilkin KB, Laws JW, Illing RC, Williams R. Cholangitis complicating congenital hepatic fibrosis. Gut 1972;13(4):319. [26] Sanchez C, Gonzalez E, Garau J. Trimethoprim-sulfamethoxazole treatment of cholangitis complicating congenital hepatic fibrosis. Pediatr Infect Dis 1986;5(3):360–3. [27] Naveh Y, Roguin N, Ludatscher R, Auslaender L, Schramek A, Aharon M. Congenital hepatic fibrosis with congenital heart disease. A family study with ultrastructural features of the liver. Gut 1980;21(9):799–807. [28] Neumann HP, Krumme B, van Velthoven V, Orszagh M, Zerres K. Multiple intracranial aneurysms in a patient with autosomal recessive polycystic kidney disease. Nephrol Dial Transplant 1999;14(4):936–9. [29] King K, Genta RM, Giannella RA, Weesner RE. Congenital hepatic fibrosis and cerebral aneurysm in a 32-year-old woman. J Pediatr Gastroenterol Nutr 1986;5(3):481–4. [30] Dusol Jr M, Levi JU, Glasser K, Schiff ER. Congenital hepatic fibrosis with dilation of intrahepatic bile ducts. A therapeutic approach. Gastroenterology 1976;71(5):839–43. [31] Grunfeld JP, Albouze G, Jungers P, Landais P, Dana A, Droz D, et al. Liver changes and complications in adult polycystic kidney disease. Adv Nephrol Necker Hosp 1985;14:1–20. [32] Kissane JM. Congenital Malformations. In: Heptinstall RH, editor. Pathology of the Kidney. 2nd edition. Boston: Little Brown and Co; 1974. p. 69. [33] Hughes J, Ward CJ, Peral B, Aspinwall R, Clark K, San Millan JL, et al. The polycystic kidney disease 1 (PKD1) gene encodes a novel protein with multiple cell recognition domains. Nat Genet 1995;10(2):151–60. [34] Polycystic kidney disease: the complete structure of the PKD1 gene and its protein. The International Polycystic Kidney Disease Consortium. Cell 1995;81(2):289–98. [35] Gallagher AR, Hidaka S, Gretz N, Witzgall R. Molecular basis of autosomal-dominant polycystic kidney disease. Cell Mol Life Sci 2002;59(4):682–93. [36] Mochizuki T, Wu G, Hayashi T, Xenophontos SL, Veldhuisen B, Saris JJ, et al. PKD2, a gene for polycystic kidney disease that encodes an integral membrane protein. Science 1996;272(5266):1339–42. [37] Daoust MC, Reynolds DM, Bichet DG, Somlo S. Evidence for a third genetic locus for autosomal dominant polycystic kidney disease. Genomics 1995;25(3):733–6. [38] Landing BH, Wells TR, Claireaux AE. Morphometric analysis of liver lesions in cystic diseases of childhood. Hum Pathol 1980;11(Suppl 5):549–60. [39] Bernstein J, Stickler GB, Neel IV. Congenital hepatic fibrosis: evolving morphology. APMIS Suppl 1988;4:17–26. [40] Alagille D, Odievre M, Gautier M, Dommergues JP. Hepatic ductular hypoplasia associated with characteristic facies, vertebral malformations, retarded physical, mental, and sexual development, and cardiac murmur. J Pediatr 1975;86(1):63–71. [41] Li L, Krantz ID, Deng Y, Genin A, Banta AB, Collins CC, et al. Alagille syndrome is caused by mutations in human Jagged1, which encodes a ligand for Notch1. Nat Genet 1997;16(3):243–51. [42] Oda T, Elkahloun AG, Pike BL, Okajima K, Krantz ID, Genin A, et al. Mutations in the human Jagged1 gene are responsible for Alagille syndrome. Nat Genet 1997;16(3):235–42. [43] Emerick KM, Rand EB, Goldmuntz E, Krantz ID, Spinner NB, Piccoli DA. Features of Alagille syndrome in 92 patients: frequency and relation to prognosis. Hepatology 1999;29(3):822–9. [44] Lykavieris P, Hadchouel M, Chardot C, Bernard O. Outcome of liver disease in children with Alagille syndrome: a study of 163 patients. Gut 2001;49(3):431–5. [45] Emerick KM, Whitington PF. Clinical aspects of familial cholestasis (with molecular explanations). Curr Gastroenterol Rep 1999;1(3):223–30.

874

B.M. Kamath, D.A. Piccoli / Gastroenterol Clin N Am 32 (2003) 857–875

[46] Kamath BM, Krantz ID, Spinner NB, Heubi JE, Piccoli DA. Monozygotic twins with a severe form of Alagille syndrome and phenotypic discordance. Am J Med Genet 2002;112(2):194–7. [47] Krantz ID, Piccoli DA, Spinner NB. Clinical and molecular genetics of Alagille syndrome. Curr Opin Pediatr 1999;11(6):558–64. [48] Krantz ID, Colliton RP, Genin A, Rand EB, Li L, Piccoli DA, et al. Spectrum and frequency of jagged1 (JAG1) mutations in Alagille syndrome patients and their families. Am J Hum Genet 1998;62(6):1361–9. [49] Eldadah ZA, Hamosh A, Biery NJ, Montgomery RA, Duke M, Elkins R, et al. Familial Tetralogy of Fallot caused by mutation in the jagged1 gene. Hum Mol Genet 2001; 10(2):163–9. [50] Krantz ID, Rand EB, Genin A, Hunt P, Jones M, Louis AA, et al. Deletions of 20p12 in Alagille syndrome: frequency and molecular characterization. Am J Med Genet 1997;70(1):80–6. [51] Silberbach M, Lashley D, Reller MD, Kinn WF Jr, Terry A, Sunderland CO. Arteriohepatic dysplasia and cardiovascular malformations. Am Heart J 1994; 127(3):695–9. [52] Crosnier C, Lykavieris P, Meunier–Rotival M, Hadchouel M. Alagille syndrome. The widening spectrum of arteriohepatic dysplasia. Clin Liver Dis 2000;4(4):765–78. [53] Sokol RJ, Heubi JE, Balistreri WF. Intrahepatic ‘‘cholestasis facies’’: is it specific for Alagille syndrome? J Pediatr 1983;103(2):205–8. [54] Kamath BM, Loomes KM, Oakey RJ, Emerick KE, Conversano T, Spinner NB, et al. Facial features in Alagille syndrome: Specific or cholestasis facies? Am J Med Genet 2002;112(2):163–70. [55] Alagille D, Estrada A, Hadchouel M, Gautier M, Odievre M, Dommergues JP. Syndromic paucity of interlobular bile ducts (Alagille syndrome or arteriohepatic dysplasia): review of 80 cases. J Pediatr 1987;110(2):195–200. [56] Watson GH, Miller V. Arteriohepatic dysplasia: familial pulmonary arterial stenosis with neonatal liver disease. Arch Dis Child 1973;48(6):459–66. [57] Berard E, Sarles J, Triolo V, Gagnadoux MF, Wernert F, Hadchouel M, et al. Renovascular hypertension and vascular anomalies in Alagille syndrome. Pediatr Nephrol 1998;12(2):121–4. [58] Connor SE, Hewes D, Ball C, Jarosz JM. Alagille syndrome associated with angiographic moyamoya. Childs Nerv Syst 2002;18(3–4):186–90. [59] Quek SC, Tan L, Quek ST, Yip W, Aw M, Quak SH. Abdominal coarctation and Alagille syndrome. Pediatrics 2000;106(1):E9. [60] Shefler AG, Chan MK, Ostman–Smith I. Middle aortic syndrome in a boy with arteriohepatic dysplasia (Alagille syndrome). Pediatr Cardiol 1997;18(3):232–4. [61] Hoffenberg EJ, Narkewicz MR, Sondheimer JM, Smith DJ, Silverman A, Sokol RJ. Outcome of syndromic paucity of interlobular bile ducts (Alagille syndrome) with onset of cholestasis in infancy. J Pediatr 1995;127(2):220–4. [62] Berard E, Triolo V. Intracranial hemorrhages in Alagille syndrome. J Pediatr 2000; 136(5):708–10. [63] Moreau S, Bourdon N, Jokic M, de Rugy MG, Babin E, Valdazo A, et al. Alagille syndrome with cavernous carotid artery aneurysm. Int J Pediatr Otorhinolaryngol 1999;50(2):139–43. [64] Woolfenden AR, Albers GW, Steinberg GK, Hahn JS, Johnston DC, Farrell K. Moyamoya syndrome in children with Alagille syndrome: additional evidence of a vasculopathy. Pediatrics 1999;103(2):505–8. [65] Emerick KM, Whitington PF. Partial external biliary diversion for intractable pruritus and xanthomas in Alagille syndrome. Hepatology 2002;35(6):1501–6. [66] Tzakis AG, Reyes J, Tepetes K, Tzoracoleftherakis V, Todo S, Starzl TE. Liver transplantation for Alagille’s syndrome. Arch Surg 1993;128(3):337–9.

B.M. Kamath, D.A. Piccoli / Gastroenterol Clin N Am 32 (2003) 857–875

875

[67] Cardona J, Houssin D, Gauthier F, Devictor D, Losay J, Hadchouel M, et al. Liver transplantation in children with Alagille syndrome–a study of twelve cases. Transplantation 1995;60(4):339–42. [68] Gurkan A, Emre S, Fishbein TM, Brady L, Millis M, Birnbaum A, et al. Unsuspected bile duct paucity in donors for living-related liver transplantation: two case reports. Transplantation 1999;67(3):416–8.