In children, intrahepatic cholestasis may have a viral, metabolic, or toxic origin or due to structural abnormalities of the intrahepatic bile ducts, which has been described under various names (intrahepatic biliary hypoplasia, hypoplasia of interlobular bile ducts, ductular hypoplasia, ductopenia, paucity of interlobular bile ducts). Paucity of interlobular bile ducts (PILBD) is the generally accepted term. Clinically two categories are recognized: syndromic (Alagille's syndrome, also called arteriohepatic dysplasia) and nonsyndromic (1,2). Nonsyndromic PILBD has been described in association with various metabolic (alpha-1 antitrypsin deficiency) (3), viral (rubella, cytomegalovirus) (4), and other conditions such as chromosomal abnormalities, cystic fibrosis, altered bile acid metabolism (5-7). Most cases are primary or idiopathic.
Bile duct paucity can be defined only histologically, and requires a sufficiently large liver biopsy that contains at least five portal tracts (2). The ratio of interlobular ducts to the number of portal tracts lies between 0.9 and 1.8 in term children and ratios under 0.5 confirm the diagnosis of PILBD. Since 1975, evidence has accumulated that PILBD, like extrahepatic biliary atresia, may not be only a static condition due to agenesis or nonformation of the bile ducts, but may also correspond to a progressive, necroinflammatory, destructive, and sclerosing cholangiopathy of unknown origin, resulting in a reduced number of interlobular ducts (8,9).
While Alagille's syndrome is recognized as an important and relatively common cause of neonatal cholestasis, reports about nonsyndromic PILBD are quite rare (10). We analyzed the clinical, biochemical, and histopathological features of 10 patients with nonsyndromic PILBD.
MATERIALS AND METHODS
In the period from 1984 to 1995, 10 liver biopsies in infants and children fulfilled the histopathological criteria of PILBD: the absence or marked reduction in the number of bile ducts in portal tracts (ratio of bile ducts to portal tracts < 0.5, >5 portal areas). None of the patients had peculiar facies, cardiovascular abnormalities, vertebral arch defects, or posterior embryotoxon. Liver biopsies were obtained surgically in three and with Tru-Cut needle in seven patients. Biopsy samples were stained with hematoxylin and eosin, trichrome, silver and periodic acid-Schiff stains after being fixed in 10% formolsaline, and investigated by light microscopy.
Clinical examination; liver biochemical tests; markers for hepatitis B and TORCH infections; determination of α1-antitrypsin level, serum and urine amino acids; sweat electrolyte determination for cystic fibrosis; abdominal ultrasonography; x-rays of the thoracic spine; and [99mTc]HIDA excretion were performed on all the patients. Three children had diagnostic laparotomies for exploration of the extrahepatic biliary system, as the [99mTc]HIDA scans showed no passage to the duodenum despite good uptake of hepatocytes. In two children, gallbladders seemed to be hypoplasic by ultrasonography.
Seven boys and three girls fulfilled the diagnostic criteria (Table 1). All but one were term babies. Birth weights were appropriate for gestational age in eight patients, while one was large and the other small for gestational age. Age at referral ranged from 17 days to 3 years (median 2.5 months).
All of the patients had jaundice at the first admission. Persistent or intermittent icterus was identified at the first month in seven infants. The other children's jaundice was observed at 4, 17, and 30 months, respectively (cases 4, 5, and 9). Stools were pale intermittently in seven and persistently in three patients. Among the six infants presenting with icterus and pale stools in the first 3 months of life, two had septicemia (cases 1 and 6), with corneal ulcer perforation in case 6. Septicemia developed after pneumonia in both with high neutrophil counts, but unfortunately, immunologic tests were not performed. Case 8 had chronic fatty diarrhea diagnosed by high fat excretion in stool collected for 72 h. Jejunal biopsy was not available. Pruritus was a prominent symptom in five children (cases 4, 5, 7, 9, and 10), having its onset between 2 months and 2.5 years of age (median 1.1 years).
Hepatomegaly, defined as a liver edge palpable more than 2 cm below the costal margin was present in eight patients. The spleen was palpable 0.5-3 cm below the costal margin in seven patients at the first examination.
Growth retardation was evident in five patients starting after 6 months of age. None of the patients had mental or motor developmental delay.
Laboratory Investigations (Table 1)
Investigations for α1-antitrypsin deficiency, cystic fibrosis, abnormal titers of serum and urine amino acids, and TORCH and hepatitis B infections were normal. Biochemical tests of liver function were abnormal in all but case 3, in the beginning. The maximum ALT levels ranged from 3 to 10 times the upper limit of normal, with the maximum serum bilirubin being 4-22 times normal. Although the serum bilirubin and ALT level normalized in one child (case 2), another whose ALT level was normal in the beginning (case 3), rose later.
Alkaline phosphatase levels were higher than normal in six patients, and in two they rose through the course. In two patients, either the serum cholesterol or triglyceride levels were elevated (cases 7 and 10). Low serum albumin levels were recorded in four children (cases 1, 3, 5, and 7). Abdominal ultrasonography revealed a normal biliary tract in all but two patients, whose gallbladder seemed to be hypoplasic. [99mTc]HIDA excretion test was performed to six patients. Bile passage to intestines was not evident in three children, while in three others poor hepatocyte uptake, indicating liver cell damage, was recognized.
Liver biopsies, performed at ages of 20 days to 3 years (median 5 months), revealed a paucity of interlobular bile ducts and intracellular cholestasis (Fig. 1) in all, and cirrhosis with regenerative nodules in two (ages 20 months and 3 years, cases 5 and 9). Mild to moderate portal tract fibrosis was present in cases 2, 3, 6, 7, and 10, while giant cells were found in another (case 4, 7.5 months old). Five patients had disturbed hepatocyte architecture (cases 5, 6, 7, 9, and 10), and three had minimal to moderate inflammatory cells in the portal tracts (cases 4, 5, and 10).
Parents of eight children were relatives (first degree in five). Three of our patients were the first product of their families. Five of our patients had had eight older sibs dying with similar symptoms. The five couples with multiple child deaths were all consanguineous. Among these 10 families, four were unable to give birth to healthy babies, in spite of two to three deliveries. None of the families were related to each other.
Mean follow up was 2.1 years with a range of 2 months to 4 years. One patient was lost to follow-up at 3 months of age. Five patients had prolonged prothrombin time at presentation that was corrected with parenteral vitamin K. Two of these (cases 3 and 7) also had rickets and hemolytic anemia cured by vitamin D and E supplementation, respectively. Hydrogen peroxidase tests and reticulocyte counts were high, and peripheral smears revealed hemolytic findings such as acanthocytosis and basophilic stippling in those patients (lowest hemoglobin levels 7.18 and 8.39 g/dl) before vitamin E was started. Another infant had corneal ulcer perforation due to vitamin A deficiency (vitamin A 3.2 μg/100 ml, normal range 30-65; carotene 0.34 mg/ml, normal range 0.93-2.3) and vitamin E-responsive hemolytic anemia. Two infants with normal prothrombin activity developed rickets and hemolytic anemia.
While all children received oral polyvitamin (vitamin A 2500 IU, vitamin D3 400 IU, vitamin B1 1.5 mg/cl, vitamin B2 1.7 mg/cl, vitamin B6 0.5 mg/cl, vitamin C 30 mg/dl, niacinamide 15 mg/dl, dl-pantenol 0.5 mg/cl), cases 3, 5, 6, and 7 also received parenteral fat-soluble vitamins, because of clinical deficiencies. Although pruritus was well controlled by cholestyramine in three and by phenobarbital in two patients, one was refractory to both these and prednisolone also. As the patient receiving ursodeoxycholic acid did not show any improvement of liver function tests or pruritus at the end of 2 months, cholestyramine was added, resulting in better clinical efficacy. Orthotopic liver transplantation was performed successfully in one child at two years of age because of decompensated cirrhosis.
A 5-month-old baby died because of surgical complications after laparotomy. Two died, one at 8 months and one at 1.5 years, due to liver failure and/or complications of portal hypertension. One patient whose liver biopsy was performed at 4 months of age cleared her jaundice right after the biopsy. She is 4 years old now, without any complaint or physical or biochemical pathological finding. This family gave birth to another girl who had prolonged jaundice and intracranial hemorrhage due to vitamin K deficiency at 40 days of age. Jaundice disappeared after phenobarbital and cholestyramine therapy at 3 months of age. She is now 21 months old, healthy, without any cholestatic findings or sequelae. Unfortunately, the parents refused liver biopsies. Five patients (ages 12 and 16 months and 2, 3, and 4 years) are alive, three with pruritus. There was no predictive histologic marker for prognosis. Fibrosis, inflammation, or degree of cholestasis did not differ between the survivors and non-survivors.
Only a few series of nonsyndromic PILBD cases have been published. In the series of Kahn et al. (10), among 17 patients, nine patients had associated well-defined primary diseases, including Down's syndrome, hypopituitarism, cystic fibrosis, α1-antitrypsin deficiency, CMV infection, and Ivemark's syndrome. In our report, all 10 cases were idiopathic. Alagille and Odievre (11) described 24 patients with nonsyndromic paucity.
A definite male preponderance was found in our study, while both sexes were equally affected in Kahn's series. Although most children in that study were born prematurely, all but one of our patients were term babies. The high incidence of jaundice and acholic stools among our patients at presentation is similar to other reports.
Three of our patients' liver biopsies were performed before 90 days of age. Kahn et al. (10) have shown that biopsies obtained before that age are sufficient for diagnosis, in contrast with Alagille syndrome, as duct paucity and portal fibrosis were evident as early as 1 week of age. In Kahn's series, giant cells were found 76.5% in early biopsies, and 33.3% in biopsies of the children older than 90 days. While only one of our patients (aged 7 months) showed giant cells, none of the three young ones had them.
Growth retardation and poor weight gain were present in 40% of patients in this study and were found frequently in others (10,11). Anorexia, a major problem causing reduced food intake, may be a factor. It may be secondary to malabsorption caused by low bile salt concentrations in the intestinal lumen. The morbidity in childhood is considerable. In our series, 70% had complications attributed to malabsorption of fat-soluble vitamins.
The prognosis of patients with nonsyndromic paucity remains quite variable. Ishak and Sharp (12) feel that patients with nonsyndromic PILBD usually have a benign course, except for those with increased trihydroxycoprostonic acid production. The patients of Eyssen et al. (7), with high trihydroxycoprostonic acid in duodenal fluid, had some congenital stigmata such as simian creases, epicanthal folds, frontal bossing, clump feet, broad metopic suture, and third fontanel; these stigmata were not present in our patients. A peroxisomal disorder is suggested for this entity. However, their familial nature and early age of death (6 and 8.5 months) related to liver disease are features similar to those in our patients. Witzleben (13) separates patients with nonsyndromic PILBD into two groups: those with progressive disease and those with a benign course. Two of our patients died at 8 months and 1.5 years of age of causes related to liver disease. Another patients survived only after an orthotopic liver transplantation. This was performed in another country, and no data about his liver histology at the time of operation are available. On the other hand, two children are being followed, without jaundice or pruritus and are growing well, at the ages of 2 and 4 years. There are recent reports about PILBD patients in whom bile ducts reappeared spontaneously, at the ages of 5 and 9 years (14,15). As second liver biopsies are not available for our patients, it is not possible to comment about their liver histologies. It seems, therefore, that nonsyndromic paucity does not constitute a homogeneous entity, but rather is the common hepatic response to different noxious factors. Outcome might then depend on the degree of severity of the damage.
As a group, the diseases associated with PILBD can progress to cirrhosis in 30% of patients. This progression is seen in only 14% of the syndromic cases, in contrast to 45% in the nonsyndromic patients (16). The average age at the time of description of cirrhosis in the former group is 12 years, whereas it is only 4.5 years in the latter. The mechanism for the marked difference in the incidence of cirrhosis between the two groups is unclear. Cirrhosis was demonstrated in biopsies of two of our patients at relatively younger ages (20 months and 3 years). Although Kahn et al. (10) suggested that increased tortuosity of interlobular bile ducts, ductular proliferation associated with leukocytic portal infiltration, and disruption of the limiting plate represent unfavorable histologic signs, our patients who died did not show these changes.
The frequency of similar symptoms in siblings of both sexes of the five affected children and the very high consanguinity rate among the parents in our series is an interesting finding that has not been reported before. This suggests an inherited autosomal recessive genetic disorder. However, accurate diagnosis of primary idiopathic nonsyndromic PILBD needs chromosome and bile acid analysis that are lacking in our patients.
PILBD should be considered in children with cholestasis, especially when recurrent icterus and acholic stools are present. Differentiating PILBD from other forms of cholestasis, essentially extrahepatic biliary atresia, is especially important in early infancy to avoid needless surgery. Accurate diagnosis may improve cholestasis and prevent permanent liver injury.
1. Alagille D, Odievre M, Gautier M, et al. Hepatic ductular hypoplasia associated with characteristic facies, vertebral malformations, retarded physical, mental and sexual developement and cardiac murmur. J Pediatr
2. Piccoli DA, Witzleben CL. Disorders of the intrahepatic bile ducts. In: Walker AW, Durie PR. Hamilton JR. Walker Smith JA, Watkins JB, eds. Pediatric gastrointestinal disease
. Philadelphia: BC Decker, 1991:1124-40.
3. Hadchouel M, Gautier M. Histopathologic study of the liver in the early cholestatic phase of α1
antitrypsin deficiency. J Pediatr
4. Kage M, Kosai KI, Kojiro M, Nokamura Y, Fukuda S. Infantile cholestasis due to cytomegalovirus infection of the liver. A possible cause of paucity of interlobular bile ducts. Arch Pathol Lab Med
5. Puri P, Guinay EJ. Intrahepatic biliary atresia in Down's syndrome. J Pediatr Surg
6. Furuya KN, Roberts EA, Canny GJ, Phillips MJ. Neonatal hepatitis syndrome with paucity of interlobular bile ducts in cystic fibrosis. J Pediatr Gastroenterol Nutr
7. Eyssen H, Parmentier G, Compernolle F, Boon J, Eggermant E. Trihydroxycoprostanic acid in the duodenal fluid of two children with intrahepatic bile duct anomalies. Biochim Biophys Acta
8. Desmet VJ. The Cholangiopathies. In: Suchy FJ, ed. Liver disease in children
. St. Louis, MO: Mosby, 1994:145-65.
9. Desmet VJ. Congenital diseases of intrahepatic bile ducts: variations on the theme “ductal plate malformation.” Hepatology
10. Kahn E, Daum F, Markowitz J, Teichberg S, et al. Nonsyndromic Paucity of interlobular bile ducts: light and electron microscopic evaluation of sequential liver biopsies in early childhood. Hepatology
11. Alagille D, Odievre M. Cholestasis in children. In: Alagille D, Odievre M, eds. Liver and biliary tract disease in children
. New York: John Wiley & Sons, 1979:163-95.
12. Ishak K, Sharp H. Developmental abnormalities in liver disease in childhood. In: Macsween R, Anhony P, Scheuer P, eds. Pathology of the liver
. New York: Churchill Living-stone, 1979:72-3.
13. Witzleben CL. Bile duct paucity (intrahepatic atresia). In: Rosenberg HS, Bernstein J, eds. Perspective in pediatric pathology
. New York: Maison Publishing, 1982:185-201.
14. Fujisawa T, Kage M, Ushijima K, Kimura A, Ono E, Kato H. Alagille syndrome with a spontaneous appearance of the interlobular bile ducts. Acts Pediatr Jpn
15. Berezin S, Beneck D, Altman P, Schwarz SM. Resolution of nonsyndromic paucity of intrahepatic bile ducts in infancy. Dig Dis Sci
16. Perrault J. Paucity of interlobular bile ducts: getting to know it better. Dig Dis Sci
Controversies in Pediatric Inflammatory Bowel Disease Conference
Sponsored by the Children's Hospital Medical Center, University of Cincinnati and the Crohn's & Colitis Foundation, this conference on issues iun pediatric inflammatory bowel disease will be held June 13-15, 1997 at the Westin Hotel in Cincinnati, Ohio. For information contact Dr. Mitchell Coihen, Children's Hospital Medical Center, 3333 Burnet Avenue, Cincinnati, Ohio 45229: Tel: (513) 559-4415.