intrahepatic bile duct changes in human hepatosplenic schistosomiasis mansoni

Liver, 1989: 9, 100-109 Key words: human hepatosplenic form; infrahepafic bile ducts; liver biopsies; schisfosomal granuloma; schisfoso- miasis rnansoni.

Intrahepatic bile duct changes in human hepatosplenic schistosomiasis mansoni

M. REGINA VIANNA, L. C. C. GAYOTTO, R. TELMA, M. SANTOS, V. A. FERREIRA ALVES, JULIA FUKUSHIMA

AND T. DE BRIT0

Department of Pathology and Institute of Tropical Medicine, University of S. Paulo Medical School and Institute “Adolfo Lutz”, S. Paulo Health Ser- vice, S. Paulo, Brazil

ABSTRACT - Wedge liver biopsies of 132 patients with hepatosplenic mansonian schistosomiasis were studied and divided in two groups according to the presence (Group I - 69 cases) or absence (Group I1 - 63 cases) of markers of the actual presence of the parasite in the liver tissue. Histological variables indicating bile duct injury were analysed in each case: periductal fibrosis, hyperplasia of the bile duct epithelium, bile duct degeneration, and marginal ductular proliferation. The presence of one or more of these variables defined two sub-groups: A - bile duct lesions present (73 cases), and B - bile duct lesions absent (59 cases). The variables “bile duct degeneration” and “ductular proliferation” were related to the actual presence of the parasite in the host. In 55.3% of all cases of human mansonian schistosomiasis a spectrum of injuries to the bile ducts was present. Epithelial hyperplasia alone or associated with patterns of mucopolysaccharide production was observed in 87.6% cases of the Sub-group A. The bile ducts changes in mansonian schistosomiasis are close to those described in liver fluke infestations such as clonorchiasis, fascioliasis and opistorchiasis. Statistical analysis revealed that high mucopolysaccharide production was associated with epithelial hyperplasia. The pathogenesis of the bile duct changes in human mansonian schistosomiasis and its relation to the parasitic infestations and their antigens is discussed.

Accepted for publication 14 September I988

Previous experimental studies in mice infested with Schistosoma mansoni showed intrahepatic bile duct changes (1, 2), such as duct cell hyperplasia, cytoplasmic vacuolation and oncocytic change of the cytoplasm of the altered cells. S . mansoni antigens were detected by immunohistochemical methods in isolated cells of the damaged bile duct lining (2).

Bedi & Isseroff (3) reported common bile duct enlargement, thickening of the bile duct wall and hyperplasia of the inner epithelial lining, also in

mice infested with S. mansoni. They suggested that proline released by schistosomes into the portal circulation might induce the bile duct changes.

To our knowledge, there are no previous re- ports on intrahepatic bile duct injury in human schistosomiasis mansoni and its possible link with the parasitic infestation. Hence, this is the main purpose of the present study. Furthermore, an attempt was made to demonstrate S. mansoni antigens in the damaged epithelial lining of the bile ducts.

INTRAHEPATIC BILE DUCTS IN HUMAN SCHISTOSOMIASIS MANSONI 101

Fig. 1. Onion skin pattern of periductal fibrosis of an interlobular bile duct and slight chronic peripheral inflammation. HE x 150.

Patients and methods a) Wedge liver biopsies of 167 patients with clinical and haemodynamic evidence of hepatosplenic schistosomiasis mansoni received between 1979 and 1986 were reviewed. Patients were submitted to different surgical procedures to alleviate portal hypertension, and the wedge biopsies were obtained during surgery. All cases with periportal Symmers fibrosis (4) and injury to the venous branches, characterized by narrowed lumen and muscle layer frag- mentation were initially studied. After excluding the cases with clinical and pathological evidence of extrahepatic biliary tract lesion, hepatitis B and/or piecemeal necrosis, 132 biopsies remained to be studied.

The biopsies were divided into two groups:

Group I: 69 cases with granulomata and/or eggs with miracidia and/or schistosomal pigment. Such findings were considered markers of the actual presence of the parasite in the host.

Group 11: 63 cases without the above markers of actual infestation.

All the liver specimens were fixed in 10% formalin and routinely embedded in paraffin. About 5-pm sections were cut and processed for conventional stainings which included haemdtoxylin-eosin, Masson’s trichrome, Perl’s

stain for iron and Gomori’s reticulin stain. The com- bined alcian blue (AB) (PH 2.5) and periodic acid-Schiff (PAS) with diastase digestion method to distinguish blue-stained acidic from red-stained neutral mucopoly- saccharides (MPS) could be done in 71 of the 73 cases in Group A (see below). Five normal livers were used as controls for the MPS stain.

The following histological variables were analysed in each case and semi-quantitatively graded (0) when absent, (1) when slight, (2) when moderate and (3) when marked:

Periductal fibrosis (Fig. 1). Bile-duct hyperplasia: including interlobular and septal ducts, characterized by increased number and size of the cells which appeared like a pseudo-stratified epithelium. Bile-duct degeneration: characterized by basal MPS negative cytoplasmatic vacuolation and nuclear changes, mainly pyknosis, of the epithelial cells of the interlobular and septa1 ducts (Fig. 2). Marginal ductular proliferation.

The finding of one or more of these variables was sup- posed to define two groups:

Sub-group A: Presence of bile duct damage (73 cases). Sub-group B: Absence of bile duct damage (59 cases).

b) The immunoperoxidase procedure was done as fol-

102 VIANNA ET AL.

Fig. 2. Epithelial lining of a septa1 bile duct exhibiting slight epithelial hyperplasia characterized by increased size of cells and rows of nuclei at different heights. Evidence of cell degeneration such as nuclear pyknosis (asterisk) and basal cytoplasmic vacuolation (arrow) are present. HE x 500.

lows: rabbit anti-serum was raised against S. mansoni worms. Goat anti-rabbit IgG from Diagnostic Reagent, USA (linking antibody) and peroxidase-antiperoxidase complex from Dako Corporation, USA, were used in l / 50 and 1/2000 dilutions respectively in the PAP method. Staining was performed according to the method of Sternberger et al. (5). The rabbit IgG against S. mansoni was incubated overnight and the linking antibody for 30 min.

The substrates used were 6% hydrogen peroxide (0.6 ml), 3,3’-diaminobenzidine (Sigma Corp. USA) (30 mg) diluted in 100 ml of PBS, pH = 7.4. Sections were slightly counterstained with Harris haematoxylin. Positive stain was not observed in normal human liver and was abol- ished when PBS or non-immune serum was used as the first layer.

c) Gammaglutamyltransferase was studied in 46, alka- line phosphatase in 54, and GPT in 56 patients.

d) Statisfical analysis: in order to test the homogeneity among groups or variables the chi-square test was used. For tables, the 2 x 2 Fisher exact test was performed.

For the biochemical evaluation logarithmic transformation was applied in an attempt to normalize data

so that the analysis of variance could be used through the minimal square method to adjust a general linear model.

Data were run in a IBM 4341 computer with a statistical analysis system.

Results In 73 cases (55.3%) altered intrahepatic bile ducts were observed (Sub-group A), and its distribution in Groups I and I1 is displayed in Table 1.

Table 1 Distribution of cases with (sub-group A) and without bile duct damage (sub-group B) in relation to the presence (Group I) or absence (Group 11) of markers of actual parasitic infestation

Subgroup

Group A B Total

I 43 (62.3%) 26 (37.7%) 69 (100%) I1 30 (47.7%) 33 (52.3%) 63 (100%)

Total 73 59 132


Table 2 Association between bile-duct degeneration and hyperplasia

Bile-duct degeneration

Hyperplasia o + 1 2+3 Total

o + 1 2+3

22 (84.62%) 4 (15.38%) 26 (100.00%) 27 (57.45%) 20 (42.55%) 47 (lO0.OOX)

P=0.0155.

Periductal fibrosis was seen in 71 cases (97.2%), and no difference was observed in its distribution in Groups I and 11.

Epithelial hyperplasia was observed in 64 (87.6%) and epithelial degeneration in 54 cases (73.9%). The absence of cases with a marked degree of epithelial degeneration in Group I1 made the statistical analysis non-viable. However, when scores 0 together with 1 were compared to 2 together with 3, epithelial degeneration was significantly higher in Group I (p < 0.0001).

Marginal ductular proliferation was observed in 39 cases (53.4%) and, similarly to epithelial degeneration, when the scores 0 with 1 and 2 with 3 were grouped, a significant predominance in Group I (p < 0.0001) resulted.

Statistical studies among the variables disclosed a positive association between hyperplasia and epithelial degeneration (Table 2) (p = 0.01 55).

When the 71 cases with bile duct damage (Sub- group A) were stained by ABIPAS, the following three patterns were detected:

Fig. 3. Pattern a: septa1 bile duct with a rim deposit of both neutral and acidic MPS on the luminal border. Ductal epithelial lining shows a focal area of slight epithelial hyperplasia (arrow). ABjPAS stain counterstained by methanil yellow and haematoxylin x 300.

104 VIANNA ET AL.

a. Observed in 21 cases, the pattern was characterized by a purple rim made up of a mixture of acidic and neutral MPS over the luminal border of the ductal epithelium. Rare small PAS positive granules were seldom observed in the supra-nuclear cytoplasm (supra-nuclear neutral mucin). This pattern of MPS deposit was also observed in the five normal livers used as controls (Fig. 3).

b. The pattern was found in 25 cases and was characterized by a marked thickening of the MPS purple rim made up of an admixture of acidic and neutral MPS over the luminal border of the epithelial lining and by the now conspicuous presence of supra-nuclear granules of neutral mucin. Occasionally, heavy deposits of either acidic or neutral MPS formed a large cytoplasmic vacuole that gave the cell a goblet cell appearance (intestinal metaplasia) (Fig. 4 and 5).

c. The pattern was observed in 25 cases, and char-

acterized by the findings described in b and, in addition, new-gland formation due to invagination of the epithelium of the interlobular ducts (Fig. 6 and 7)

Statistical analysis failed to show significant predominance of any of these patterns in either group. However, high MPS production was associated with marked epithelial hyperplasia (p<0.0019) (Fig. 8).

The odds-ratio revealed that the probability of finding ductal damage in the group with actual infestation was 1.8 higher than in the group without it. However, this ratio was not significant at a 0.05 level (6).

As previously reported in experimental animals (2, 7) and in humans (8), S. mansoni antigen deposits were detected in miracidia, dead worms and occasionally in rare monocytes which were part of poorly organized granulomata. S. mansoni anti-

Fig. 4. Pattern b septa1 bile duct lining showing moderate epithelial hyperplasia and MPS negative basal vacuoles (arrow). The MPS cytoplasm deposits of the partial goblet cell metaplasia are linked to the thickened MPS stained rim seen on the luminal border (asterisk). AB/PAS stain counterstained by methanil yellow and haematoxylin x 500.


Fig. 5. Pattern b interlobular bile duct showing marked epithelial hyperplasia and a thickened rim of acidic and neutral MPS on the luminal border. Partial goblet cell metaplasia of the epithelium (arrow) is also seen. AB/PAS stain counterstained by methanil yellow and haematoxylin x 300.

gen was not detected in the cytoplasm of the cells lining the bile ducts.

Analysis of variance did not show any biochemical variable able to differentiate Group I from Group TI.

Discussion A spectrum of injuries to the intrahepatic bile ducts, observed in patients with hepatosplenic schistosomiasis mansoni, comprise bile duct hyperplasia either alone or associated with changes in the pattern of MPS production. Epithelial hyperplasia was present in 87.6% of the cases, and the statistical analysis showed a parallel between this variable and the patterns of MPS production. In other words, marked epithelial hyperplasia of the bile duct lining was accompanied by enhance- ment of MPS production with newly formed glands and goblet cell formation (intestinal metaplasia). The stimulation of epithelial hyperplasia

and MPS production in mansonian schistosomiasis is apparently not a specific one and is close to the findings already reported in the bile ducts of humans and animals infected by Clonorchis sinensis (9, lo), Fasciola hepatica (1 1) and Opis- thorchis viverrini (12). It is noteworthy that these parasites live in the biliary tree and the constant movement of the worms and/or metabolic products of the living or degradation products of the dead parasites probably result in mechanical and chemical irritation to the epithelia (10). In schistosomiasis mansoni portal inflammation around eggs and/or dead worms probably leads to the dissemination of products of the parasite through the intrahepatic peribiliary vascular plexus with injury to the interlobular and septal bile ducts.

In experimental murine schistosomiasis mansoni (1, 2), bile duct changes are more marked and show a somewhat different morphological aspect to those described in man. However, in

106 VIANNA ET AL.

Fig. 6 . Pattern c: septa1 bile-duct with changes similar to those described for the pattern b associated with periductal new glands formation with acidic and neutral MPS production. This pattern is quite close to the bile duct change associated with the Clonorchis sinensis infestation. ABjPAS stain counterstained with methanil yellow and haematoxylin x 100.

experimental infestation there are more eggs and dead worms, followed by a more marked portal inflammation, when compared to the disease in humans. This probably explains the more severe bile-duct lesions described in murine schistosomiasis mansoni.

Kozuka et al. (13) believe that “intestinal metaplasia in the human biliary tract mostly represents neoplastic transformation or conditions liable to neoplastic transformation”. Mucous gland hyperplasia, regarded as a proliferative type of intestinal metaplasia, is considered by some as a possible predisposing factor of cancer (1 3). Moreover, Ku- rashina et al. (14) showed that intrahepatic bile duct hyperplasia, which is significantly frequent in bile ducts undergoing inflammatory changes, is a precursor of cholangiocellular carcinoma in man. The high incidence of cholangiocarcinomas and particularly of the mucin-producing type in man and animals infected with Clonorchis sinensis

(1 5, 16) concurs with this point of view. However, such an association has not been reported so far in human schistosomiasis mansoni. It is worth noting that in clonorchiasis, fascioliasis and opis- thorchiasis, parasites are in intimate contact with the bile duct epithelium, which does not happen in schistosomiasis. However, when there is a close contact between the parasite and the epithelium, as observed between Schistosoma haematobium and the urinary bladder epithelium, a high incidence of bladder carcinoma is observed (17, 18).

It appears, therefore, that some parasite infestations can stimulate bile duct epithelial proliferation, occasionally with intestinal metaplasia. Ex- perimental evidence shows that the aminoacid proline might be the active agent inducing bile-duct hyperplasia in fascioliasis (19), and, on the other hand, it is known that Schistosoma mansoni also releases large amounts of proline into the hepato- enteric circulation (3). However, isolation and cul-

INTRAHEPATIC BILE DUCTS IN HUMAN SCHlSTOSOMlASlS MANSONI 107

Fig. 7. Septa1 bile duct showing new glands formation due to the invagination of the lining epithelium (arrow). The bile duct has an irregular contour, with focal areas of moderate epithelial hyperplasia. HE x 100.

ture of intrahepatic bile ducts of rats (20) and the addition to the culture medium of putative cholan- giotrophic factors, among them proline, failed to result in any alteration of epithelial replication.

The role played by the antigenic components of the S. mansoni egg and/or worm in the morphological changes of the bile ducts is so far merely speculative. Antigen deposits were detected in altered bile duct cells in murine experimental schistosomiasis mansoni (2). Furthermore, common antigens were isolated and partially characterized from adult Schistosoma mansoni, Fasci- ola hepatica, Trichinella spiralis and Biornphalaria glabrata (21-23). We were unable to detect schistosomal antigen in human intrahepatic bile duct, but it must be pointed out that on morphological grounds the experimental model does not repro- duce entirely the human infestation and that the antigenic load is usually heavier in the experimental than in the human infestation. As only one polyclonal serum was used in this study, further research is needed, employing monoclonal and other polyclonal sera before excluding the pres-

ence of antigen deposits in the hyperplastic human bile duct epithelium.

Ductular proliferation and degeneration of the epithelial ductal lining were related to the actual presence of the parasite in the host. Bile-ductular proliferation is a non-specific finding, commonly seen in enlarged portal tracts and also in fibrous septa in various hepato-cellular and biliary diseases (24), and may be the result of necro-inflammatory conditions of different etiologies, mechanical obstruction and developmental lesions (25).

Periductal fibrosis with an onion skin pattern was a frequent (97.2%) but also non-specific fea- ture in our study. It is described also in chronic intra- and extrahepatic (including gallbladder) disorders of the biliary tree (26) and its presence in schistosomiasis, sometimes around atrophic bile ducts, is further evidence of the chronic bile duct disease in this parasitic infestation.

Acknowledgements We thank Miss Edna Aparecida Leick for technical assistance, Mr Justin0 Silva for the photographic work

108 VIANNA ET AL.

.I. 100

:: 7 0

I WYPERPLASIA

- MARKED - MODERATE

- S L I G H T

0 - A B S E N T

so

4 0

30

20

10

0

PATTERN a PATTERN b PATTERN c I O + l ) x ( 2 + 3 )

p = 0.0019

Fig. 8. Relationship between the distribution patterns of MPS and the variable epithelial hyperplasia.

and Miss Maria Eli P. de Castro for assistance in prepa- ring the manuscript.

References 1. GONNERT R V. Schistosomiasis Studien. IV. Zur Pa-

thologie der Schistosomiasis der Maus. 2 Tropen- med Parasit 1955: 6 279-336.

2. DA SILVA L C, ALVES V A F, ABRANTES C P, LIMA D M, CHRISTO C H, DE BRITO T. Effects of chemo- therapy on mice submitted to multiple Schistosoma mansoni infections. A controlled randomized prospective study. Trop Med Parasit 1985: 36 150-154.

3. BEDI A J K, ISSEROFF H. Bile-duct hyperplasia in mice infected with Schistosoma mansoni. Int J Para- sitol 1979: 9 401404.

4. SYMMERs w ST c. Note on a new form of liver cirrhosis due to the presence of the ova of Bilharzia haematobia. J Pathol Bacteriol 1903: 9: 237-239.

5. STERNBERCER L A, HARDY P H, CUCULITS J J, ME- YER H J. The unlabelled antibody-enzyme method of immunohistochemistry. Preparation and proper- ties of soluble antigen-antibody complex (Horse- radish peroxidase-antihorseradish peroxidase) and its use in identification of spirochetes. J Histochem Cytochem 1970: 18: 315-333.

6. LILIENFIELD A M, LILIENFIELD D E. Foundation of epidemiology, 2nd edn. N.Y.: Oxford University Press, 1980.

7. ABDUL-AAL G, ATTALLAH A M. Immunopathology of experimental Schistosoma mansoni: immunohistochemical localization of parasite antigens in the

host tissue. Int Arch Allergy Appl Immunoll987: 82

8. DE BRITO T, HOSHINO-SHIMIZU S, YAMASHIRO E, DA SILVA L C. Chronic human schistosomiasis mansoni: schistosomal antigen, immunoglobulin and complement C3 detection in the liver. Liver 1985: 3 64-70.

9. CHOU S T, GIBSON J B. The histochemistry of biliary mucins and the changes caused by infestation with Clonorchis sinensis. J Pathol 1970: 101: 185-197.

10. Hou P C. The pathology of Clonorchis sinensis infestation of the liver. J Pathol Bacteriol 1955: 7 0 53-64.

11. ACOSTA-FERREIRA W, VERCELLI-RELTA J, FALCONI L M. Fasciola hepatica human infection. Histopatho- logical study of sixteen cases. Virchows Arch [A]

12. THAMAVIT W, KONGKANUNTN R, TIWAWECH D, MOORE M A. Level of Opisthorchis infestation and carcinogen dose - dependence of cholangiocarcino- ma induction in Syrian golden hamsters. Virchows Arch [B] 1987: 5 4 52-58.

13. KOSUKA S, KURASHINA M, TSUBONE M, HACHISUKA K, YASUI A. Significance of intestinal metaplasia for the evolution of cancer in the biliary tract. Cancer

14. KURASHINA M, KOZUKA S, NAKASIMA N, HIRABAY- ASI N, ITO M. Relationship of intrahepatic bile duct hyperplasia to cholangiocellular carcinoma. Cancer 1988: 61: 2469-2474.

15. Hou P C. The relationship between primary carcinoma of the liver and infestation with Clonorchis sinensis. J Pathol Bacteriol 1956: 72: 239-246.

16. Hou P C. Primary carcinoma of bile duct of the liver of the cat (Felis catus) infested with Clonorchis sinensis. J Pathol Bacteriol 1964: 87: 239-244.

17. SMITH J H, CHRISTIE J D. The pathobiology of Schis- tosoma haematobiurn infection in humans. Hum Pa-

18. FUKUSHIMA S, ITO N, EL-BOLKAINY M N, TAWFIK H N, TATEMOTO Y, MORI M. Immunohistochemical observation of keratins, involucrin, and epithelial membrane antigen in urinary bladder carcinomas from patients infected with Schistosoma haematobium. Virchows Arch [A] 1987: 411: 103-115.

19. ISSEROFF H, SAWMA J T, REINO D. Fascioliasis: role of proline in bile duct hyperplasia. Science 1977:

20. GALL J A M, BHATHAL P S. Isolation and culture of intrahepatic bile ducts and its application in asses- sing putative inducers of biliary epithelial cell hyperplasia. Br J Exp Pathol 1987: 68: 501-510.

21. RASMUSSEN K R, HILLYER G V, KEMP W M. Iso- lation and partial characterization of an antigen shared between Schistosoma mansoni, Fasciola hepatica and Biomphalaria glabrata. J Parasitol 1985:

22. HILLYER G V. Induction of immunity in mice to Fasciola hepatica with a Fasciola/Schistosoma cross-

89-94.

1979: 383 319-327.

1984: 54: 2277-2285.

tho1 1986: 17: 333-345.

198: 1157-1 159.

71: 792-798.


reactive defined immunity antigen. Am J Trop Med

23. ARONSTEIN W S, LEWIS S A, NORDEN A P, DALTON J P, STRAND M. Molecular identity of a major antigen of Schistosoma mansoni which cross-reacts with Trichinella spiralis and Fasciola hepatica. Parasit-

24. NAKAMURA Y, OHTA G. Immunohistochemical study on bile ductular proliferation in various hepa- tobiliary diseases. Liver 1986 6 206211.

25. International group. Histopathology of the intrahepatic biliary tree. Liver 1983: 3 161-175.

Hyg 1985: 34: 1127-2231.

ology 1986: 92: 133-151.

26. POPPER H. General pathology of the liver: light microscopic aspects serving diagnosis and interpret- ation. Sem Liver Dis 1986: 6 175-184.

Address: Thales B De Brito, M.D. Faculdnde de Medicina da USP Departamento a'e Patologia Av. Dr. Arnaldo - 455 01246 - S6o Paul0 - S.E! Brad

intrahepatic bile duct changes in human hepatosplenic schistosomiasis mansoni

Documents