Glomerular alterations in children with biliary atresia YoshiYuki O h t o m o , MD, Yutaka Fukuda, MD~ PhD, Keijiro Y a b u t a , MD, Takeshi Miyano; MD, Yasuhiko Tomino, MD, a n d Toshikazu Shirai, MD, PhD From the Departments of Pediatrics, Pediat0c Surgery, Nephrology, and Pathology, Juntendo University School of Medicine, Tokyo, Japan
To investigate the association of glomerular involvement with bi!iary atresia,we carried out the following studies: (I) review of the clinical records of 120patients, (2) histo!ogic study of the kidneys obtained at autopsy from 28 patients, and (3) measurements of circulating immune complexes. Of 90 patients with a d e q u a t e follow-up information, 40 (44.4%) had hematuria, proteinuria, or both. All the kidney specimens showed a wide variety of mesangial proliferation, and immunoglobulins were present in 23 of 26 cases. There was a good correlation between the glomerular alterations and the period of reduced hepatic function. When igA was present in the mesangium, IgA2 and secretory components were detected. Elevated serum IgA and circulating IgA-containing immune complex levels were found in patients with prolonged obstructive jaundice. These findings indicate that glomerular alterations may occur subsequently in patients with biliary atresia, and that IgA of intestinal mucosal origin plays some role in the development of these lesions. (J PEDiATR1992;120:404-8)
Cirrhosis of the liver has been known to be associated with glomerular changes since 1942) and these histologic changes have been termed hepatic glomerulonephritis.2 There are, however, few reports about HGN in childhood3 because few children with advanced liver disease are seen. In biliary atresia, inflammatory destruction of the extrahepatic bile ducts is rapidly followed by a biliary type of cirrhosis. We found that some patients with BA have urinary abnormalities during the course of the illness.4 The purpose of our study was to investigate the association of HGN with BA by reviewing the medical charts and by Supported in part by a scientific research grant (No. 02670461) from the Japanese Ministry of Education. Presented at the 1lth International Congress of Nephrology, Tokyo, Japan; July 1990. at the 23rd Annual Meeting of the American Society of Nephrology, Washington, D.C.. December 1990. and at the 5th International Sendal Symposiumon Biliary Atresia, Sendai. Japan, May 1991. Submitted for publication July 19. 1991: accepted Oct. 24, 1991. Reprint requests: YoshiyukiOhtomo. MD, Department of Pediatrics, Juntendo University School of Medicine, 2-1-1. Hongo, Bunkyo-ku. Tokyo 113. Japan.
9/20/34568
404
studying the serum, urine, and kidney specimens of our patients. METHODS The clinical records of 120 patients with BA seen at Juntendo University Hospital (Tokyo, Japan) between January 1979 and October 1989 were reviewed; 90 of the patients who had had adequate follow-up information were enrolled in the study. All the patients had extrahepatic BA, diagnosed by operative exploration and intraoperative cholangiography.
I
BA HGN
Biliary atresia Hepatic glomerulonephritis
I
Of these 90 patients, 50 were dead (group 1; the age at death ranged from 2 months to 8 years 9 months, with a mean of 2 years 5 months); 40 patients had been followed for more than 2 years after portoenterostomy (group 2). The patients in group 2 were further divided into two groups: group 2A (average follow-up period 5 years 11 months) consisted of 15 patients with prolonged obstructive jaundice and a serum level of direct bilirubin > 18 #mol/L (1.0 mg/
Volume 120 Number 3
Table
Glomerular alterations in children with biliary atresia
4 05
I. Incidence of urinary abnormalities in patients with BA Groups
No. (%) with hematuria No. (%) with proteinuria 70-300 mg/dl >300 mg/dl No. (%) with no urinary abnormalities
I (n = 50)
2A (n = 15)
2B (n = 25)
Total (n - 90)
21 (42.0)
8 (53.3)
4 (16.0)
33 (36.7)
9 (18.0) 6 (12.0) 23 (46.0)
1 (6.7) 2 (13.3) 7 (46.7)
2 (8.0) 1 (4.0) 20 (80.0)
12 (13.3) 9 (10.0) 50 (55.6)
II. Incidence of urinary changes and the results of immunofluorescence studies in relation to the grade of histologic
Table
change Glomerular changes
No. of patients
Hematuria and/ or proteinuria
IgA +
IgG +
IgM +
Grade 2
13
Grade 3
13
Grade 4
2
TOTAL
28
6/13 (46.2) 8/13 (61.5) 2/2 (lOO.O) 16/28 (57.1)
6/12 (50.0) 8/12 (66.7) 2/2 (lOO.O) 16/26 (61.5)
7/12 (58.3) 7/12 (58.3) 2/2 (lOO.O) 16/26 (61.5)
8/12 (66.7) 11/12 (91.7) 2/2 (lOO.O) 21/26 (80.8)
Results are shownas numberof positivecases per numberof cases studied.Numbersin parenthesesare percentages.
dl) at 1 year of age; and group 2B (average follow-up period 7 years 10 months) consisted of 25 patients who were free of jaundice after operation (Table I). The laboratory data of all 90 patients were reviewed with respect to urinalyses, biochemical tests of liver function, serum creatinine concentration, serum immunoglobulin levels, and serum complement activities. Among several biochemical tests, we used serum cholinesterase levels as an index of hepatic damage. These tests were done by standard laboratory techniques. Nephritic urinary changes were defined as proteinuria of more than 70 mg/dl, hematuria of more than 10 to 15 erythrocytes per high-power field, or both. We studied kidneys available from autopsy of 28 patients of group 1. All patients had biliary fibrosis or cirrhosis when they died. An autopsy was performed less than 24 hours after death. All kidney specimens were examined by light microscopy, and 26 by immunofluorescence. For light microscopy, kidney tissue was fixed in neutral-buffered formalin. After dehydration and wax embedment, sections that were 2 to 4 #m thick were stained with hematoxylin and eosin, the periodic acid-Schiff technique, and Masson trichrome. For immunofluorescence, frozen sections were cut with a cryostat at 2 or 3 #m, washed in buffer, and stained with commercial direct fluorescein isothiocynate-tagged antisera to human IgA, IgG, IgM, C3, Clq, fibrinogen (Cappel Laboratories, Organon Teknika BV, Turnhout, Belgium),
IgA1, and IgA2 (Beckton Dickinson Immunocytometry Systems, Mountain View, Calif.). By using the indirect immunoftuorescence technique, we stained tissue with antisera to human secretory component (Australian Monoclonal Development Pty Ltd., Hawthorn, Australia). These sections were examined and photographed with a Zeiss fluorescence microscope. All the kidney specimens were assigned to one of the following four grades on the basis of light microscopic findings according to Nakamoto et al.5: grade 1, no appreciable change; grade 2, mild to moderate mesangial proliferation; grade 3, moderate to marked mesangial proliferation with local circumferential mesangial interposition; and grade 4, moderate to marked mesangial proliferation with diffuse circumferential mesangial interposition. In 20 patients, IgA-class circulating immune complex measurements were made by using the solid phase anti-C3 enzyme immunoassay.6 These serum samples were kept frozen at - 4 0 ~ C until they were assayed. Data were analyzed with the Student t test. Differences were considered significant at p
To investigate the association of glomerular involvement with bi!iary atresia,we carried out the following studies: (I) review of the clinical records of 120patients, (2) histo!ogic study of the kidneys obtained at autopsy from 28 patients, and (3) measurements of circulating immune complexes. Of 90 patients with a d e q u a t e follow-up information, 40 (44.4%) had hematuria, proteinuria, or both. All the kidney specimens showed a wide variety of mesangial proliferation, and immunoglobulins were present in 23 of 26 cases. There was a good correlation between the glomerular alterations and the period of reduced hepatic function. When igA was present in the mesangium, IgA2 and secretory components were detected. Elevated serum IgA and circulating IgA-containing immune complex levels were found in patients with prolonged obstructive jaundice. These findings indicate that glomerular alterations may occur subsequently in patients with biliary atresia, and that IgA of intestinal mucosal origin plays some role in the development of these lesions. (J PEDiATR1992;120:404-8)
Cirrhosis of the liver has been known to be associated with glomerular changes since 1942) and these histologic changes have been termed hepatic glomerulonephritis.2 There are, however, few reports about HGN in childhood3 because few children with advanced liver disease are seen. In biliary atresia, inflammatory destruction of the extrahepatic bile ducts is rapidly followed by a biliary type of cirrhosis. We found that some patients with BA have urinary abnormalities during the course of the illness.4 The purpose of our study was to investigate the association of HGN with BA by reviewing the medical charts and by Supported in part by a scientific research grant (No. 02670461) from the Japanese Ministry of Education. Presented at the 1lth International Congress of Nephrology, Tokyo, Japan; July 1990. at the 23rd Annual Meeting of the American Society of Nephrology, Washington, D.C.. December 1990. and at the 5th International Sendal Symposiumon Biliary Atresia, Sendai. Japan, May 1991. Submitted for publication July 19. 1991: accepted Oct. 24, 1991. Reprint requests: YoshiyukiOhtomo. MD, Department of Pediatrics, Juntendo University School of Medicine, 2-1-1. Hongo, Bunkyo-ku. Tokyo 113. Japan.
9/20/34568
404
studying the serum, urine, and kidney specimens of our patients. METHODS The clinical records of 120 patients with BA seen at Juntendo University Hospital (Tokyo, Japan) between January 1979 and October 1989 were reviewed; 90 of the patients who had had adequate follow-up information were enrolled in the study. All the patients had extrahepatic BA, diagnosed by operative exploration and intraoperative cholangiography.
I
BA HGN
Biliary atresia Hepatic glomerulonephritis
I
Of these 90 patients, 50 were dead (group 1; the age at death ranged from 2 months to 8 years 9 months, with a mean of 2 years 5 months); 40 patients had been followed for more than 2 years after portoenterostomy (group 2). The patients in group 2 were further divided into two groups: group 2A (average follow-up period 5 years 11 months) consisted of 15 patients with prolonged obstructive jaundice and a serum level of direct bilirubin > 18 #mol/L (1.0 mg/
Volume 120 Number 3
Table
Glomerular alterations in children with biliary atresia
4 05
I. Incidence of urinary abnormalities in patients with BA Groups
No. (%) with hematuria No. (%) with proteinuria 70-300 mg/dl >300 mg/dl No. (%) with no urinary abnormalities
I (n = 50)
2A (n = 15)
2B (n = 25)
Total (n - 90)
21 (42.0)
8 (53.3)
4 (16.0)
33 (36.7)
9 (18.0) 6 (12.0) 23 (46.0)
1 (6.7) 2 (13.3) 7 (46.7)
2 (8.0) 1 (4.0) 20 (80.0)
12 (13.3) 9 (10.0) 50 (55.6)
II. Incidence of urinary changes and the results of immunofluorescence studies in relation to the grade of histologic
Table
change Glomerular changes
No. of patients
Hematuria and/ or proteinuria
IgA +
IgG +
IgM +
Grade 2
13
Grade 3
13
Grade 4
2
TOTAL
28
6/13 (46.2) 8/13 (61.5) 2/2 (lOO.O) 16/28 (57.1)
6/12 (50.0) 8/12 (66.7) 2/2 (lOO.O) 16/26 (61.5)
7/12 (58.3) 7/12 (58.3) 2/2 (lOO.O) 16/26 (61.5)
8/12 (66.7) 11/12 (91.7) 2/2 (lOO.O) 21/26 (80.8)
Results are shownas numberof positivecases per numberof cases studied.Numbersin parenthesesare percentages.
dl) at 1 year of age; and group 2B (average follow-up period 7 years 10 months) consisted of 25 patients who were free of jaundice after operation (Table I). The laboratory data of all 90 patients were reviewed with respect to urinalyses, biochemical tests of liver function, serum creatinine concentration, serum immunoglobulin levels, and serum complement activities. Among several biochemical tests, we used serum cholinesterase levels as an index of hepatic damage. These tests were done by standard laboratory techniques. Nephritic urinary changes were defined as proteinuria of more than 70 mg/dl, hematuria of more than 10 to 15 erythrocytes per high-power field, or both. We studied kidneys available from autopsy of 28 patients of group 1. All patients had biliary fibrosis or cirrhosis when they died. An autopsy was performed less than 24 hours after death. All kidney specimens were examined by light microscopy, and 26 by immunofluorescence. For light microscopy, kidney tissue was fixed in neutral-buffered formalin. After dehydration and wax embedment, sections that were 2 to 4 #m thick were stained with hematoxylin and eosin, the periodic acid-Schiff technique, and Masson trichrome. For immunofluorescence, frozen sections were cut with a cryostat at 2 or 3 #m, washed in buffer, and stained with commercial direct fluorescein isothiocynate-tagged antisera to human IgA, IgG, IgM, C3, Clq, fibrinogen (Cappel Laboratories, Organon Teknika BV, Turnhout, Belgium),
IgA1, and IgA2 (Beckton Dickinson Immunocytometry Systems, Mountain View, Calif.). By using the indirect immunoftuorescence technique, we stained tissue with antisera to human secretory component (Australian Monoclonal Development Pty Ltd., Hawthorn, Australia). These sections were examined and photographed with a Zeiss fluorescence microscope. All the kidney specimens were assigned to one of the following four grades on the basis of light microscopic findings according to Nakamoto et al.5: grade 1, no appreciable change; grade 2, mild to moderate mesangial proliferation; grade 3, moderate to marked mesangial proliferation with local circumferential mesangial interposition; and grade 4, moderate to marked mesangial proliferation with diffuse circumferential mesangial interposition. In 20 patients, IgA-class circulating immune complex measurements were made by using the solid phase anti-C3 enzyme immunoassay.6 These serum samples were kept frozen at - 4 0 ~ C until they were assayed. Data were analyzed with the Student t test. Differences were considered significant at p
