- Email: [email protected]
Role of reovirus type 3 in persistent infantile cholestasis
Role of reovirus type 3 in persistent infantile cholestasis The relationship between reovirus type 3 and persistent infantile cholestasis was studied by measuring antibody to the virus in the sera o f affected and control babies younger than 1 year o f age. One hundred sixty-seven infants were divided into four groups: those with extrahepatie biliary atresia, idiopathic neonatal hepatitis, or other cholestatic disorders, and controls. When available, maternal sera obtained simultaneously with infant sera were also studied. The results indicate that 62% o f babies with extrahepatie biliary atresia and 52% o f infants with idiopathic neonatal hepatitis have reovirus 3 antibodies. In contrast, less than 12% o f either normal infants or babies with other eholestatie disorders have antibodies. These observations suggest that perinatal infection with reovirus type 3 may serve as an initiating event in the genesis o f two closely related forms o f infantile obstructive cholangiopathy: extrahepatie biliary atresia and idiopathic neonatal hepatitis. (J PEOtArR 10S:912, 1984)
Joy H. Glaser, M.D., William F. Balistreri, M.D., and Rachel Morecki, M.D., Bronx, New
York, and Cincinnati, Ohio
SEROLOGIC EVIDENCE for an association between reovirus type 3 and extrahepatic biliary atresia has been reported), 2 Sixty_eight percent of children with this disease were shown to have indirect immunofluorescent antibodies to Reo 3, whereas only 8% of healthy controls had similar antibodies) The high incidence of Reo 3 antibodies in children with EHBA has raised questions regarding its incidence in other disease entities characterized by persisten~ cholestasis in infancy, in an attempt to clarify the latter, the serologic studies have been extended to include two additional groups of patients with cholestasis: infants with idiopathic neonatal hepatitis and babies with persistent infantile cholestasis resulting from other causes. From the Departments o f Pediatrics and Pathology and the Liver Research Center, Albert Einstein College o f Medicine; and the Children's Hospital Research Foundation, Cincinnati Children's Hospital. Supported in part by The March o f Dimes and by Grant AM-I Z702 from the National Institutes o f Health. Presented in part at the International Symposium on Biliary Atresia and Related Disorders, Sendal Japan, 1983. Submitted for publication Feb. 6, 1984; accepted June 8, 1984. Reprint requests: Joy H. Glaser, M.D., Department o f Pediatrics, Albert Einstein College o f Medicine, 1300 Morris Park Ave., Bronx, N Y 10461.
912
The Journal o f P E D I A T R I C S
Because the acquisition of Reo 3 antibodies is known to rise in normal infants after 1 year of age, the current study was restricted to those younger than 1 year.
METHODS Sera were obtained from 167 infants _<1 year of age who were cared for in New York, Ohio, Virginia, and California. Sequential sera were available from 18 patients with EHBA, 15 with N H , seven with Other cholestatic
EHBA IFA NH Reo 3 TORCH
Extrahepatic biliary atresia Immunofluorescent antibodies Neonatal hepatitis Reovirus type 3 Toxop!asmosis, rubella, cytomegalovirus, herpesvirus
conditions, and 22 control subjects. The results of the latest serum sample were used for statistical analysis. Most serum samples were obtained from infants younger than 6 months. For purposes of comparison, the study population was divided into four major groups (Table I). Group 1 included 34infants with surgically and morphologically confirmed
Volume 105 Number 6
Reovirus type 3 and neonatal cholestasis
EHBA (this group was further subdivided into A and B, depending on whether only one or more samples of infant sera were available for study). Group 2 consisted of 23 patients with prolonged cholestasis diagnosed as "idiopathic" N H after exclusion of known metabolic disorders, infection with T O R C H agents, and hepatitis B by appropriate tests and liver biopsy. Group 3 consisted of 33 patients with Cholestasis related to other identifiable disorders. Group 4 consisted of 77 infants without cholestasis, including four infants with liver abnormalities unaccompanied by cholestasis (one each with intrahepatic abscess, GM-1 gangliosidosis, glycogen storage disease, and disseminated neur0blastoma ). All of the patients with persistent cholestasis and 90% of the controls were hospitalized for varying lengths of time in the first few months of life. The most common reasons for hospitalization among the controls included prematurity and various congenital anomalies. Because this is a cumulative study, some patients have been described previously.!-3 Serum from only one patient, included i n group 3, with sYndromatic paucity of intrahepatic bile ducts was available for study. Maternal sera. Maternal serum obtained simultaneously with the infant's initial or only serum sample was available in 40 cases. The maternal sera were taken when the infant was younger than 6 month s of age. This allowed comparisons to be made between a mother's Reo 3 immunofluorescent antibodies and her infant's serology. Antibody determination. The indirect immunofluorescent testing was performed as previously described. ~ All sera were numerically coded, and tested and scored as positive or negative at a dilution of 1:10. Scoring was based on at least three separate determinations. The diagnosis in each case was not known until all tests were completed and the results reported to the referring physician. RESULTS Infants. Antibodies to Reo 3 were idehtified in 21 of 34 (62%) patients with EHBA (group 1), in 12 of 23 (52%) with N H (group 2), in four of 33 (12%) with other cholestatic diseases (group 3), and in five of 77 (6%) controls (group 4) (Table II). The high incidence of antibodyin patients with EHBA and N H is in contrast to the serologic results in the infants with a variety of other diseases associated with prolonged neonatal cholestasis (group 3). In this population the prevalence of antibodies was 12%, similar to the 6% incidence found among control infants (group 4). In addition, antibody was not detected in four babies with noncholestatic liver involvement included as part of the control group. Because of the etiologic diversity of disease in infants in group 3, no conclusions can be reached regarding Reo i
913
Table I. Clinical characteristics of 167 study patients Group
Diagnosis
Number of patients
1 2 3
Extrahepatic biliary atresia Idiopathic neonatal hepatitis Cholestasis of other causes Total parenteral alimentation Cytomegalovirus-associated hepatitis Cystic fibrosis a~-Antitrypsin deficiency Niemann-Pick disease Other* Control
34 23 33 14 7 4 2 2 4 77
4
*One each Schwachman syndrome, malignant histiocytosis,tyrosinemia, and syndromaticintrahepaticpaucityof bile ducts.
3 antibody for each individual condition. However, only one of 14 Children in whom cholestasis was secondary to total parenteral alimentation had Reo 3 IFA. Positive serology among the other infants with cholestasis was seen in one child with cystic fibrosis, one with syndromatic paucity of bile ducts, and one of seven infants with hepatitis associated With cytomegaiovirus. Mothers. Reo 3 antibodies were demonstrated in 16 of 21 (76%) mothers in group 1, in 6 of 11 (54%) in group 2, and in five of eight (62%) in group 4. Maternal serum was not available for group 3 infants. The low incidence of Reo 3 I F A in 77 control subjects, including 38 infants younger than 2 months of age, can be contrasted with the high incidence of such antibodies in infants with EHBA and N H (Table II). In the latter groups, 13 of 22 infants were seropositive simultaneously with their mothers, and all of the children remained seropositive from 2 to 15 weeks later. None of five control babies of seropositive mothers had antibody or developed it subsequently (Table III). In addition, eight initially seronegative infants with EHBA or N H became seropositive, whereas 22 seronegative control infants remained seronegative on subsequent testing (data not shown). DISCUSSION The results of these serologic studies support clinical and pathologic observations made over the past 20 years, which suggest a common cause for EHBA and NH. After the initial description of "neonatal giant cell hepatitis mimicking extrahepatic biliary atresia," awareness of the many similarities between these two disorders progressively increased. 4 Histologic analysis of percutaneous liver biopsies demonstrated aspects of N H commonly observed in patients with EHBA, especially in the early phase of disease. 5 In addition, histologic studies of bile duct remnants resected during portoenterostomy revealed an
9 14
Glaser, Balistreri, and M o r e c k i
The Journal o f Pediatrics December 1984
Table II. Comparison o f reovirus type 3 immunofluorescent antibody by age in patients and controls
Age
Group 1
Group 2
Group 3
Group 4
Extrahepatie biliaryatresia
Neonatal hepatitis
Other cholestasis
Control
(too)
Total
Positive
%
Total
Positive
%
Total
Positive
%
Total
Positive
%
0to 2 3to5 6 to 12 Total
16 12 6 34
8 10 3 21
50 83 50 62
12 8 3 23
5 5
42 63 67 52
16 8 9 33
1 0 ~ 4
6 0 33 12
38* 13 2{5 77
1 1
3 8 12 6
2 12
5
*Sixteen were --<4weeksof age; all were seronegative.
Table III. Reovirus type 3 immunofluorescent antibody in 40 mothers and their infants
Group 1 (n
21)
Extrahepatic biliary atresia A
B
S1
Group 4 (n = 8)
Neonatal hepatitis
Control
Infant
Infant Mother
Group 2 (n = 11)
Mother
S1
Infant $2
Mother
Infant
SI
$2
Mother
S1
$2
-
(3)
+
-
(2)
-
(8)
+
+ (1)
+ (4)
+
-
(4)
-
(8)
-
(6)
-
-
(4)
-
(6)
+
-
-
-
-
(4)
-
(8)
+ (8)
+
- (4)
- (8)
+
+ (2)
+ (17)
-
- (4)
- (8)
-
(8)
§
+ (4)
+
-
-
+ (24)
-
-
-
+ (12) + (12) + (24)
+ + +
- (4) - (4) + (6)
+ (20) + (20) + (8)
-
- (3) - (5) - (8)
- (15) + (10) + (17)
+ + +
- (4) - (4) - (4)
- (8) - (8) - (8)
+ (16)
+
-
-
-
(24)
(10)
(2)
+
+ (6)
+
+
-
+
+ (6)
+ (16)
+
- (12)
-
-
-
+
+ (15)
+
+ (8) + (8) + (12)
-
- (16)
+
(8)
(14)
(2)
(12)
+ (16) + (24) + (20)
(12)
(14)
(4)
(4)
(8)
(8)
Testing performed as described in the text at a serum dilution of 1:10. Maternal serum was obtained simultaneouslywith infant's first serum sample. Numbers in parenthesescorrespondto age in weeks at time sample was taken.
inflammatory process in segments of bile ducts adjacent to the site of atresia. 6,7 Clinical observations, on the other
were then initiated to test the hypothesis o f a common
hand, included occasional infants in whom transition from N H to E H B A took place. 8
a large proportion (>50%) of infants with idiopathic N H
Landing, 9 in 1974, proposed the term "progressive obliterative cholangiopathy," expressing the viewpoint that
in this respect to infants with E H B A . Although the serologic findings are best interpreted as supporting Reo 3
a common etiologic factor, most probably viral, initiated
as the common cause for these diseases, the results might
factor responsible for both disorders. Results indicate that have evidence of infection with Reo 3, making t h e m similar
hepatobiliary injury in both E H B A and N H . However,
also indicate an increased susceptibility of infants with
only occasional cases of E H B A and N H were associated
cholestasis to infection with this virus. To clarify the
with cytomegalovirus or rubella, and no serologic evidence has been found for infection with hepatitis A or B. ~~ The
significance of Reo 3 I F A in infants with E H B A and
high incidence of Reo 3 antibodies in patients with E H B A ,
patients with persistent infantile cholestasis related to a
coupled with a recent report documenting the identifica-
variety of other causes. Our study clearly distinguishes babies with E H B A and N H from infants with other types
tion of Reo 3 in the p o r t a hepatis of an infant undergoing portoenterostomy, provides the evidence for the role of Reo 3 in this disease23 Serologic studies in infants with N H
idiopathic N H , similar serologic studies were done in
o f cholestasis, in whom the low incidence of antibodies to Reo 3 is similar to that observed in normal infants.
Volume 105 Number 6
Seroepidemiologic surveys have shown that most adults have Reo 3 antibodies. N o r m a l infants tend to acquire these slowly after the first year of life? 4 W e believe that these results demonstrating a high incidence of Reo 3 antibodies only in infants with E H B A and N H are consistent with a perinatal Reo 3 infection in affected infants. In addition, these studies would indicate that transplacental passage of Reo 3 I F A probably results in titers below 1:10 by the time infants are a few weeks of age. The identification of Reo 3 as a frequent etiologic factor in N H and E H B A is in keeping with experimental studies describing the affinity of Reo 3 for both murine hepatocytes and bile duct epithelium? 5 Although our study describes an agent with the potential to cause E H B A and N H , many questions remain regarding the pathogenesis of these disorders. The factor(s) that determine the major site of injury may reside in the varied bile acid composition, IgA secretion patterns, H L A type, or age at the time of infection. It is unclear why antibody is not detected in all children with E H B A and N H . It may be that in some patients Reo 3 antigen may coexist and bind to antibody, blocking further immunoreactivity detectable by the I F A method. It is also possible that other unidentified infectious agents may be the cause of these disorders in infants with negative Reo 3 serologic findings. The lower incidence in patients with N H is consistent with the heterogeneity of this group. Despite exclusion of known metabolic disorders, this group probably still includes other as yet unidentifiable inborn errors of metabolism. The results of our study suggest a specific role for Reo 3 virus in the genesis of E H B A and N H . Further studies are necessary to clarify the role of its virus in perpetuating a progressive injury observed in many of these patients. Because Reo 3 in a defective form is known to persist in the brain tissue of experimental animals, causing prolonged central nervous system disease, viral persistence in hepatobiliary tissue may also be involved in the chronic tissue injury seen in babies with E H B A and N H . 16 We thank Dr. Marshall S. Horwitz for his ~ontinuous guidance and interest in this project. For their help in obtaining the sera we wish especially to thank the following: In New, York, Dr. Janna Collins and Dr. Robert Truding, Albert Einstein College of Medicine, Dr. Phillip Lipsitz, Long Island Jewish Medical Center, Dr. Joseph Levy, Babies Hospital, and Dr. Cladd Stevens and Dr. Ann Taylor, New York Blood Center; in Virginia, Dr. J. Rainer Poley, Eastern Virginia Medical School; and in California, Dr. M. Michael Thaler, San Francisco Medical Center.
Reovirus type 3 and neonatal cholestasis
9 15
REFERENCES
1. Morecki R, Glaser JH, Cho S, Balistreri WF, Horwitz MS: Biliary atresia and Reovirus type 3 infection. N Engl J Med 307:481, 1982. 2. Bangaru B, Morecki R, Glaser JH, Gartner LM, Horwitz MS: Comparative studies of biliary atresia in the human newborn and Reovirus-induced cholangitis in weanling mice. Lab Invest 43:456, 1980. 3. Morecki R, Glaser JH, Horwitz MS: Pathogenesis of biliary atresia and reovirus type 3 infection. In Kasai M, editor: Biliary atresia and related disorders. Amsterdam, 1984, Excerpta Medica, pp 20-27. 4. Craig JM, Landing BH: A form of hepatitis in the neonatal period simulating biliary atresia. Arch Pathol 541:321, 1952. 5. Alagille D: Clinical aspects of neonatal hepatitis. Am J Dis Child 123:287, 1972. 6. Bill AH, Haas JE, Foster GL: Biliary atresia: Histopathologic observations and reflections upon its natural history. J Pediatr Surg 12:977, 1977. 7. Gautier M, Jehan P, Odievre M: Histologic study of biliary fibrous remnants in 48 cases of extrahepatic biliary atresia: Correlation with postoperative bile flow restoration. J PEDIATR 89:704, 1976. 8. McDonald JP, Stehman FB, Stewart DR: Infantile obstructive cholangiopathy. Am J Dis Child 133"518, 1979. 9. Landing BH: Considerations of the pathogenesis of neonatal hepatitis, biliary atresia and choledochal cyst: The concept of infantile obstructive cholangiopathy. Prog Pediatr Surg 6:113, 1974. 10. Lang D J, Marshall WC, Pincott JR, Stern H: Cytomegalovirus: Association with neonatal hepatitis and biliary atresia. In Javit NB, editor: Neonatal hepatitis and biliary atresia. Rockvil[e, Md., 1977, National Institutes of Health, pp3341. 11. Straus L, Bernstein J: Neonatal hepatitis in congenital rubella: A histopathological study. Arch Pathol 86:317, 1968. 12. Balistreri WF, Tabor E, Gerety R J: Negative serology for hepatitis A and B viruses in 18 cases of neonatal cholestasis. Pediatrics 66:269, 1980. 13. Moreeki R, Glaser J, Johnson AB, Kress Y: Evidence for reovirus type 3 in the porta hepatis of a patient with ~xtrahepatic biliary atresia: An immunocytochemical and ultrastructural study [abstract]. Hepatology 3:316, 1983. 14. Leers WD, Rozee KR: A survey of Reovirus antibodies in sera of urban children. Can Med Assoc J 96:1040, 1966. 15. Stanley NF: The reovirus murine models. In Melnick JE, editor: Progress in medical virology, vol 18. Basel, 1974, S. Karger, pp 257-272. 16. Spandidos DA, Graham AF: Generation of defective virus after infection of newborn rats with reovirus. J Virol 20:234, 1976.
Joy H. Glaser, M.D., William F. Balistreri, M.D., and Rachel Morecki, M.D., Bronx, New
York, and Cincinnati, Ohio
SEROLOGIC EVIDENCE for an association between reovirus type 3 and extrahepatic biliary atresia has been reported), 2 Sixty_eight percent of children with this disease were shown to have indirect immunofluorescent antibodies to Reo 3, whereas only 8% of healthy controls had similar antibodies) The high incidence of Reo 3 antibodies in children with EHBA has raised questions regarding its incidence in other disease entities characterized by persisten~ cholestasis in infancy, in an attempt to clarify the latter, the serologic studies have been extended to include two additional groups of patients with cholestasis: infants with idiopathic neonatal hepatitis and babies with persistent infantile cholestasis resulting from other causes. From the Departments o f Pediatrics and Pathology and the Liver Research Center, Albert Einstein College o f Medicine; and the Children's Hospital Research Foundation, Cincinnati Children's Hospital. Supported in part by The March o f Dimes and by Grant AM-I Z702 from the National Institutes o f Health. Presented in part at the International Symposium on Biliary Atresia and Related Disorders, Sendal Japan, 1983. Submitted for publication Feb. 6, 1984; accepted June 8, 1984. Reprint requests: Joy H. Glaser, M.D., Department o f Pediatrics, Albert Einstein College o f Medicine, 1300 Morris Park Ave., Bronx, N Y 10461.
912
The Journal o f P E D I A T R I C S
Because the acquisition of Reo 3 antibodies is known to rise in normal infants after 1 year of age, the current study was restricted to those younger than 1 year.
METHODS Sera were obtained from 167 infants _<1 year of age who were cared for in New York, Ohio, Virginia, and California. Sequential sera were available from 18 patients with EHBA, 15 with N H , seven with Other cholestatic
EHBA IFA NH Reo 3 TORCH
Extrahepatic biliary atresia Immunofluorescent antibodies Neonatal hepatitis Reovirus type 3 Toxop!asmosis, rubella, cytomegalovirus, herpesvirus
conditions, and 22 control subjects. The results of the latest serum sample were used for statistical analysis. Most serum samples were obtained from infants younger than 6 months. For purposes of comparison, the study population was divided into four major groups (Table I). Group 1 included 34infants with surgically and morphologically confirmed
Volume 105 Number 6
Reovirus type 3 and neonatal cholestasis
EHBA (this group was further subdivided into A and B, depending on whether only one or more samples of infant sera were available for study). Group 2 consisted of 23 patients with prolonged cholestasis diagnosed as "idiopathic" N H after exclusion of known metabolic disorders, infection with T O R C H agents, and hepatitis B by appropriate tests and liver biopsy. Group 3 consisted of 33 patients with Cholestasis related to other identifiable disorders. Group 4 consisted of 77 infants without cholestasis, including four infants with liver abnormalities unaccompanied by cholestasis (one each with intrahepatic abscess, GM-1 gangliosidosis, glycogen storage disease, and disseminated neur0blastoma ). All of the patients with persistent cholestasis and 90% of the controls were hospitalized for varying lengths of time in the first few months of life. The most common reasons for hospitalization among the controls included prematurity and various congenital anomalies. Because this is a cumulative study, some patients have been described previously.!-3 Serum from only one patient, included i n group 3, with sYndromatic paucity of intrahepatic bile ducts was available for study. Maternal sera. Maternal serum obtained simultaneously with the infant's initial or only serum sample was available in 40 cases. The maternal sera were taken when the infant was younger than 6 month s of age. This allowed comparisons to be made between a mother's Reo 3 immunofluorescent antibodies and her infant's serology. Antibody determination. The indirect immunofluorescent testing was performed as previously described. ~ All sera were numerically coded, and tested and scored as positive or negative at a dilution of 1:10. Scoring was based on at least three separate determinations. The diagnosis in each case was not known until all tests were completed and the results reported to the referring physician. RESULTS Infants. Antibodies to Reo 3 were idehtified in 21 of 34 (62%) patients with EHBA (group 1), in 12 of 23 (52%) with N H (group 2), in four of 33 (12%) with other cholestatic diseases (group 3), and in five of 77 (6%) controls (group 4) (Table II). The high incidence of antibodyin patients with EHBA and N H is in contrast to the serologic results in the infants with a variety of other diseases associated with prolonged neonatal cholestasis (group 3). In this population the prevalence of antibodies was 12%, similar to the 6% incidence found among control infants (group 4). In addition, antibody was not detected in four babies with noncholestatic liver involvement included as part of the control group. Because of the etiologic diversity of disease in infants in group 3, no conclusions can be reached regarding Reo i
913
Table I. Clinical characteristics of 167 study patients Group
Diagnosis
Number of patients
1 2 3
Extrahepatic biliary atresia Idiopathic neonatal hepatitis Cholestasis of other causes Total parenteral alimentation Cytomegalovirus-associated hepatitis Cystic fibrosis a~-Antitrypsin deficiency Niemann-Pick disease Other* Control
34 23 33 14 7 4 2 2 4 77
4
*One each Schwachman syndrome, malignant histiocytosis,tyrosinemia, and syndromaticintrahepaticpaucityof bile ducts.
3 antibody for each individual condition. However, only one of 14 Children in whom cholestasis was secondary to total parenteral alimentation had Reo 3 IFA. Positive serology among the other infants with cholestasis was seen in one child with cystic fibrosis, one with syndromatic paucity of bile ducts, and one of seven infants with hepatitis associated With cytomegaiovirus. Mothers. Reo 3 antibodies were demonstrated in 16 of 21 (76%) mothers in group 1, in 6 of 11 (54%) in group 2, and in five of eight (62%) in group 4. Maternal serum was not available for group 3 infants. The low incidence of Reo 3 I F A in 77 control subjects, including 38 infants younger than 2 months of age, can be contrasted with the high incidence of such antibodies in infants with EHBA and N H (Table II). In the latter groups, 13 of 22 infants were seropositive simultaneously with their mothers, and all of the children remained seropositive from 2 to 15 weeks later. None of five control babies of seropositive mothers had antibody or developed it subsequently (Table III). In addition, eight initially seronegative infants with EHBA or N H became seropositive, whereas 22 seronegative control infants remained seronegative on subsequent testing (data not shown). DISCUSSION The results of these serologic studies support clinical and pathologic observations made over the past 20 years, which suggest a common cause for EHBA and NH. After the initial description of "neonatal giant cell hepatitis mimicking extrahepatic biliary atresia," awareness of the many similarities between these two disorders progressively increased. 4 Histologic analysis of percutaneous liver biopsies demonstrated aspects of N H commonly observed in patients with EHBA, especially in the early phase of disease. 5 In addition, histologic studies of bile duct remnants resected during portoenterostomy revealed an
9 14
Glaser, Balistreri, and M o r e c k i
The Journal o f Pediatrics December 1984
Table II. Comparison o f reovirus type 3 immunofluorescent antibody by age in patients and controls
Age
Group 1
Group 2
Group 3
Group 4
Extrahepatie biliaryatresia
Neonatal hepatitis
Other cholestasis
Control
(too)
Total
Positive
%
Total
Positive
%
Total
Positive
%
Total
Positive
%
0to 2 3to5 6 to 12 Total
16 12 6 34
8 10 3 21
50 83 50 62
12 8 3 23
5 5
42 63 67 52
16 8 9 33
1 0 ~ 4
6 0 33 12
38* 13 2{5 77
1 1
3 8 12 6
2 12
5
*Sixteen were --<4weeksof age; all were seronegative.
Table III. Reovirus type 3 immunofluorescent antibody in 40 mothers and their infants
Group 1 (n
21)
Extrahepatic biliary atresia A
B
S1
Group 4 (n = 8)
Neonatal hepatitis
Control
Infant
Infant Mother
Group 2 (n = 11)
Mother
S1
Infant $2
Mother
Infant
SI
$2
Mother
S1
$2
-
(3)
+
-
(2)
-
(8)
+
+ (1)
+ (4)
+
-
(4)
-
(8)
-
(6)
-
-
(4)
-
(6)
+
-
-
-
-
(4)
-
(8)
+ (8)
+
- (4)
- (8)
+
+ (2)
+ (17)
-
- (4)
- (8)
-
(8)
§
+ (4)
+
-
-
+ (24)
-
-
-
+ (12) + (12) + (24)
+ + +
- (4) - (4) + (6)
+ (20) + (20) + (8)
-
- (3) - (5) - (8)
- (15) + (10) + (17)
+ + +
- (4) - (4) - (4)
- (8) - (8) - (8)
+ (16)
+
-
-
-
(24)
(10)
(2)
+
+ (6)
+
+
-
+
+ (6)
+ (16)
+
- (12)
-
-
-
+
+ (15)
+
+ (8) + (8) + (12)
-
- (16)
+
(8)
(14)
(2)
(12)
+ (16) + (24) + (20)
(12)
(14)
(4)
(4)
(8)
(8)
Testing performed as described in the text at a serum dilution of 1:10. Maternal serum was obtained simultaneouslywith infant's first serum sample. Numbers in parenthesescorrespondto age in weeks at time sample was taken.
inflammatory process in segments of bile ducts adjacent to the site of atresia. 6,7 Clinical observations, on the other
were then initiated to test the hypothesis o f a common
hand, included occasional infants in whom transition from N H to E H B A took place. 8
a large proportion (>50%) of infants with idiopathic N H
Landing, 9 in 1974, proposed the term "progressive obliterative cholangiopathy," expressing the viewpoint that
in this respect to infants with E H B A . Although the serologic findings are best interpreted as supporting Reo 3
a common etiologic factor, most probably viral, initiated
as the common cause for these diseases, the results might
factor responsible for both disorders. Results indicate that have evidence of infection with Reo 3, making t h e m similar
hepatobiliary injury in both E H B A and N H . However,
also indicate an increased susceptibility of infants with
only occasional cases of E H B A and N H were associated
cholestasis to infection with this virus. To clarify the
with cytomegalovirus or rubella, and no serologic evidence has been found for infection with hepatitis A or B. ~~ The
significance of Reo 3 I F A in infants with E H B A and
high incidence of Reo 3 antibodies in patients with E H B A ,
patients with persistent infantile cholestasis related to a
coupled with a recent report documenting the identifica-
variety of other causes. Our study clearly distinguishes babies with E H B A and N H from infants with other types
tion of Reo 3 in the p o r t a hepatis of an infant undergoing portoenterostomy, provides the evidence for the role of Reo 3 in this disease23 Serologic studies in infants with N H
idiopathic N H , similar serologic studies were done in
o f cholestasis, in whom the low incidence of antibodies to Reo 3 is similar to that observed in normal infants.
Volume 105 Number 6
Seroepidemiologic surveys have shown that most adults have Reo 3 antibodies. N o r m a l infants tend to acquire these slowly after the first year of life? 4 W e believe that these results demonstrating a high incidence of Reo 3 antibodies only in infants with E H B A and N H are consistent with a perinatal Reo 3 infection in affected infants. In addition, these studies would indicate that transplacental passage of Reo 3 I F A probably results in titers below 1:10 by the time infants are a few weeks of age. The identification of Reo 3 as a frequent etiologic factor in N H and E H B A is in keeping with experimental studies describing the affinity of Reo 3 for both murine hepatocytes and bile duct epithelium? 5 Although our study describes an agent with the potential to cause E H B A and N H , many questions remain regarding the pathogenesis of these disorders. The factor(s) that determine the major site of injury may reside in the varied bile acid composition, IgA secretion patterns, H L A type, or age at the time of infection. It is unclear why antibody is not detected in all children with E H B A and N H . It may be that in some patients Reo 3 antigen may coexist and bind to antibody, blocking further immunoreactivity detectable by the I F A method. It is also possible that other unidentified infectious agents may be the cause of these disorders in infants with negative Reo 3 serologic findings. The lower incidence in patients with N H is consistent with the heterogeneity of this group. Despite exclusion of known metabolic disorders, this group probably still includes other as yet unidentifiable inborn errors of metabolism. The results of our study suggest a specific role for Reo 3 virus in the genesis of E H B A and N H . Further studies are necessary to clarify the role of its virus in perpetuating a progressive injury observed in many of these patients. Because Reo 3 in a defective form is known to persist in the brain tissue of experimental animals, causing prolonged central nervous system disease, viral persistence in hepatobiliary tissue may also be involved in the chronic tissue injury seen in babies with E H B A and N H . 16 We thank Dr. Marshall S. Horwitz for his ~ontinuous guidance and interest in this project. For their help in obtaining the sera we wish especially to thank the following: In New, York, Dr. Janna Collins and Dr. Robert Truding, Albert Einstein College of Medicine, Dr. Phillip Lipsitz, Long Island Jewish Medical Center, Dr. Joseph Levy, Babies Hospital, and Dr. Cladd Stevens and Dr. Ann Taylor, New York Blood Center; in Virginia, Dr. J. Rainer Poley, Eastern Virginia Medical School; and in California, Dr. M. Michael Thaler, San Francisco Medical Center.
Reovirus type 3 and neonatal cholestasis
9 15
REFERENCES
1. Morecki R, Glaser JH, Cho S, Balistreri WF, Horwitz MS: Biliary atresia and Reovirus type 3 infection. N Engl J Med 307:481, 1982. 2. Bangaru B, Morecki R, Glaser JH, Gartner LM, Horwitz MS: Comparative studies of biliary atresia in the human newborn and Reovirus-induced cholangitis in weanling mice. Lab Invest 43:456, 1980. 3. Morecki R, Glaser JH, Horwitz MS: Pathogenesis of biliary atresia and reovirus type 3 infection. In Kasai M, editor: Biliary atresia and related disorders. Amsterdam, 1984, Excerpta Medica, pp 20-27. 4. Craig JM, Landing BH: A form of hepatitis in the neonatal period simulating biliary atresia. Arch Pathol 541:321, 1952. 5. Alagille D: Clinical aspects of neonatal hepatitis. Am J Dis Child 123:287, 1972. 6. Bill AH, Haas JE, Foster GL: Biliary atresia: Histopathologic observations and reflections upon its natural history. J Pediatr Surg 12:977, 1977. 7. Gautier M, Jehan P, Odievre M: Histologic study of biliary fibrous remnants in 48 cases of extrahepatic biliary atresia: Correlation with postoperative bile flow restoration. J PEDIATR 89:704, 1976. 8. McDonald JP, Stehman FB, Stewart DR: Infantile obstructive cholangiopathy. Am J Dis Child 133"518, 1979. 9. Landing BH: Considerations of the pathogenesis of neonatal hepatitis, biliary atresia and choledochal cyst: The concept of infantile obstructive cholangiopathy. Prog Pediatr Surg 6:113, 1974. 10. Lang D J, Marshall WC, Pincott JR, Stern H: Cytomegalovirus: Association with neonatal hepatitis and biliary atresia. In Javit NB, editor: Neonatal hepatitis and biliary atresia. Rockvil[e, Md., 1977, National Institutes of Health, pp3341. 11. Straus L, Bernstein J: Neonatal hepatitis in congenital rubella: A histopathological study. Arch Pathol 86:317, 1968. 12. Balistreri WF, Tabor E, Gerety R J: Negative serology for hepatitis A and B viruses in 18 cases of neonatal cholestasis. Pediatrics 66:269, 1980. 13. Moreeki R, Glaser J, Johnson AB, Kress Y: Evidence for reovirus type 3 in the porta hepatis of a patient with ~xtrahepatic biliary atresia: An immunocytochemical and ultrastructural study [abstract]. Hepatology 3:316, 1983. 14. Leers WD, Rozee KR: A survey of Reovirus antibodies in sera of urban children. Can Med Assoc J 96:1040, 1966. 15. Stanley NF: The reovirus murine models. In Melnick JE, editor: Progress in medical virology, vol 18. Basel, 1974, S. Karger, pp 257-272. 16. Spandidos DA, Graham AF: Generation of defective virus after infection of newborn rats with reovirus. J Virol 20:234, 1976.