Progressive bile duct injury after thiabendazole administration

GASTROENTEROLOGY 1987;93:245-9

Progressive Bile Duct Injury After Thiabendazole Administration J. CARLOS MANIVEL, JOSEPH R. BLOOMER, and DALE C. SNOVER Departments of Laboratory Medicine and Pathology and Medicine, University of Minnesota Hospital and Medical School, Minneapolis, Minnesota

A 27-yr-old man developed jaundice 2 wk after exposure to thiabendazole. Cholestasis persisted for 3 yr, at which time a liver transplant was performed. Two liver biopsy specimens and the hepatectomy specimen were remarkable for almost complete disappearance of interlobular bile ducts. Prominent fibrosis and hepatocellular regeneration were also present; however, the lobular architecture was preserved. This case represents an example of “idiosyncratic” drug-induced liver damage in which the primary target of injury is the bile duct. An autoimmune pathogenesis of the bile duct destruction is suggested. Fifteen cases of cholestasis associated with thiabendazole have been reported (l-8). In 8 cases the jaundice disappeared in a few months, 2 patients died shortly after the onset of jaundice, and no follow-up was provided in 5 cases. Therefore, the long-term effects of this condition have not been assessed. We have recently seen a patient with cholestasis induced by thiabendazole; observation of this patient for more than 3 yr allowed us to document progressive bile duct injury.

Case Report A 27-yr-old white man developed lower abdominal cramps and diarrhea in March 1983. No parasites were identified on stool examination, but he was treated by his local physician with thiabendazole (Mintezol, Merck Sharp & Dohme, West Point, Pa.), 1500 mgiday for 1 wk. Two weeks later he developed dark urine, light-colored stools, and jaundice, followed in a few days by pruritus, malaise, nausea, and vomiting. His past medical history was remarkable for asthma and allergy to multiple foods, grasses, pollens, and penicillin. There was no history of Received October 2, 1986. Accepted January 8, 1987. Address requests for reprints to: J. Carlos Manivel, M.D., Department of Laboratory Medicine and Pathology, Box 76 Mayo, University of Minnesota, 420 Delaware Street S.E., Minneapolis, Minnesota 55455. 0 1987 by the American Gastroenterological Association 0016-5085/87/$3.50

hepatitis, jaundice, inflammatory bowel disease, transfusions, or alcohol intake. He had no known contact with jaundiced persons. One sister had multiple allergies. In April 1983 he was admitted to a local hospital for evaluation. Laboratory examination revealed the following values: total serum bilirubin, 17 mgidl (normal 0.3-1); aspartate aminotransferase, 42 U/L (normal l-32); and alanine aminotransferase, 62 U/L (normal O-30). Alkaline phosphatase was “minimally elevated,” and serologic tests for viral hepatitis were negative. A presumptive diagnosis of non-A, non-B hepatitis was made, and the patient was treated symptomatically. Because of the persistence of his symptoms, the patient was admitted to another hospital in May 1983 for further evaluation. During this admission, total serum bilirubin was elevated to 17.2 mg/dl, alkaline phosphatase was 163 U/L (normal 81-290), aspartate aminotransferase was 42 U/L; coagulation tests and white blood cell and platelet counts were normal. The major diagnostic considerations were thiabendazole-induced cholestasis, cholestatic viral hepatitis, primary biliary cirrhosis, primary sclerosing cholangitis, and bile duct obstruction. Antimitochondrial antibodies and repeat tests for hepatitis virus infection were negative, and an abdominal sonogram was normal. A liver biopsy was performed. The patient was treated with cholestyramine. One month later, serum alkaline phosphatase had increased to 670 U/L, lactic dehydrogenase was 326 IUiL (normal 25-loo), and cholesterol was 492 mgdl (normal 135-283). An endoscopic retrograde cholangiopancreatogram showed no abnormalities. During the subsequent 3 yr there was a gradual increase in total bilirubin to 24.5 mgidl (14.3 mg/dl direct), alkaline phosphatase to 6280 U/L, and cholesterol to 1170 mgidl; aspartate aminotransferase ranged between 144 and 312 U/L, and alanine aminotransferase between 176 and 328 U/L. The pruritus became difficult to control, and the patient developed multiple cutaneous xanthomas, carotid bruits, malabsorption, easy bleeding, bone pain, and paresthesias. The patient was treated with cholestyramine, phenobarbital, cimetidine, parenteral vitamins, and reduced dietary fat. Because of the progression of his symptoms, the patient was referred to the University of Minnesota Hospital where liver biopsies were performed in June 1984 and December 1985. An orthotopic liver transplantation was performed in April 1986. The postoperative course was

246

MANIVEL ET AL.

GASTROENTEROLOGY Vol. 93, No. 2

Figu Nre 1. Liver biopsy specimen arrow).

complicated

by moderate

from 1983 shows marked canalicular

rejection

which

responded

to

therapy. However, disseminated herpes simplex virus infection developed. Despite treatment with acyclovir, the patient developed respiratory and renal failure, and died 6 wk after the transplant.

Pathologic Findings Sections of the needle biopsy from May 1983 showed marked canalicular and centrilobular hepatocellular cholestasis. There was evidence of hepatocellular regeneration (Figure 1). Inflammation was minimal. No portal tracts were available for evaluation. Sections of a liver biopsy specimen from June 1984 were remarkable for an almost complete absence of interlobular bile ducts despite the fact that the specimen contained more than 10 portal tracts. The rare bile ducts that were present were abnormal, having missing cells and cytoplasmic vacuolization (Figure 2). Occasional portal tracts contained proliferating ductules and a mild lymphocytic infiltrate. Portal fibrosis extended into the lobules in a stellate-shaped pattern; periportal hepatocytes showed “pseudoxanthomatous” change. There was mild centrilobular cholestasis. Numerous hepatocytes displayed “feathery” degeneration. Scattered necrotic hepatocytes were surrounded by polymorphonuclear leukocytes. Evidence of hepatocellular re-

cholestasis

(small arrows] and multinucleated

hepatocytes

(lasrge

generation continued to be present, and there was mild, diffuse sinusoidal fibrosis. A liver biopsy performed in December 1985 disclosed features similar to the previous biopsy; however, the number of portal lymphocytes and periportal pseudoxanthomatous hepatocytes had increased. The degree of portal and sinusoidal fibrosis was essentially unchanged. As in the previous biopsy, the number of bile ducts was markedly reduced. No ductular proliferation was seen in this specimen; mild centrilobular cholestasis was present. Stain for copper (rubeanic acid) was negative. The hepatectomy specimen weighed 2335 g, measured 32 X 24 X 18.5 cm, and was green. Histologically, the lobular architecture was still preserved, but there was prominent portal and centrilobular fibrosis with portal-toportal and portal-to-central bridging. No normal interlobular bile ducts were present; numerous portal tracts showed ductular proliferation (Figure 3). Evidence of hepatocellular regeneration was similar to that seen in previous specimens. Periportal pseudoxanthomatous hepatocytes and cholestasis were slightly more prominent than in the previous specimen, and focal periportal copper deposition was observed. The large trabecular ducts and major bile ducts were unremarkable. The gallbladder was normal.

August 1987

Figure 2. Liver biopsy magnification

THIABENDAZOLE-INDUCED

LIVER DAMAGE

specimen from 1984. Portal space shows damaged bile duct (arrow) and lymphoid infiltrate. of bile duct shows missing epithelial cells and cytoplasmic swelling and vacuolization.

Discussion Fifteen cases of cholestasis associated with thiabendazole have been described (l-8). In 8 patients the jaundice disappeared within 3 mo after In 2 of the 8 withdrawal of this drug (1,2,4,6). patients, serum alkaline phosphatase remained elevated for 6 and 9 mo, respectively, but eventually returned to normal values. Two patients died shortly after the onset of jaundice; one in hepatic failure (l), and the other as a complication of a diagnostic procedure (8). No follow-up was available in 5 cases (3,5,7). In addition to cholestasis, 3 patients developed the sicca complex (keratoconjunctivitis sicca and xerostomia). This syndrome resolved after 1 yr in 1 patient and 1.5 yr in the second patient (6), but was still present after 1 yr in the third patient (2). In the latter case, serum levels of pancreatic amylase were increased. The patient we describe developed cholestasis 2 wk after administration of the drug. His symptoms consisted predominantly of pruritus, jaundice, and hypercholesterolemia with secondary development of cutaneous xanthomas and accelerated atherosclerosis. Despite cessation of the drug, the cholestasis

247

Inset: hi@;her

worsened. Therefore, rechallenge with the drug to prove an etiologic association could not be performed. However, other causes of hepatitis were ruled out by appropriate laboratory and radiologic procedures. Although most previous cases have been selflimited, our case demonstrates that this is not invariably true. Persistence of symptoms for 3 yr and almost complete disappearance of bile ducts indicate that the process would not have resolved. Reversibility of the cholestatic injury probably depends on the extent of epithelial damage and the regenerative reserve of the affected cells. Liver tissue has been examined in seven instances of thiabendazole-induced liver injury, including the present case. Prominent cholestasis was a common denominator; parenchymal degeneration was minimal in all except 1 case (3). At autopsy, the changes in the liver in 1 case were described as “subacute cholangiohepatitis” (1). Eosinophilic inflammation was prominent in several cases. In the only biopsy specimen from a patient with sicca syndrome, bile duct damage was reported to be absent (2). Cholestasis induced by thiabendazole has not been described in toxicologic studies or animal mod-

248

MANIVEL

ET AL.

Figure

GASTKOENTEROLOGY

3. Portal

area from hepatectomy

specimen

els (9). The pathogenesis of this idiosyncratic reaction is not established, but hypersensitivity and metabolic aberrations have been proposed (5). Skin rash, low-grade fever, eosinophilia, and angioneurotic edema in some cases support hypersensitivity mechanisms (2). In 2 cases, challenge with thiabendazole readministration resulted in more rapid recurrence of hepatic injury (1,~). Lack of correlation between dosage or duration of therapy and the development of liver injury are well-known features of hypersensitivity-mediated idiosyncratic reactions to drugs (1,lO). It has been suggested that the drug may act as a hapten that binds to proteins and induces autoantibody production against biliary epithelium, resulting in cholestasis. In some patients it may also produce antibodies against lacrimal duct and salivary gland duct epithelium, resulting in the sicca syndrome (2). The almost complete absence of interlobular bile ducts in our case is reminiscent of the long-term complications of chronic graft-versushost disease, liver allograft rejection, and the endstage of primary biliary cirrhosis (11-14). In these disorders, immune-mediated destruction of the bile ducts has been postulated. Biliary cirrhosis induced by chloropromazine has been described (15); the pathogenesis may be similar.

shows

fibrosis

and ductular

proliferation

Vol. 93, No. 2

(arrows).

Nonimmune pathogenesis of this damage is less favored, but cannot be excluded. The mechanism could be direct toxic damage to the bile duct epithelium, or accumulation of toxic metabolites through some metabolic aberration (16). Damage to the bile excretory mechanism of the hepatocytes may result in loss of bile ducts presumably due to atrophy, as seen in Alagille’s syndrome (17). Treatment with phenobarbital was considered useful in 1 case (4). Cholestyramine, prednisone, and phenobarbital controlled pruritus in another case, but the cholestasis persisted (7). Our case showed no response to these drugs. The most important therapeutic decision is withdrawal of the drug and avoidance of reexposure to it (1,2,6). If autoimmune mechanisms are proved to be the cause of the hepatic injury, immunosuppressive therapy early in the disease may also be indicated. References Bayles MAH. Chromomycosis treatment with thiabendazole. Arch Derm 1971;104:476-85. Rex D, Lumeng L, Eble J, Rex L. Intrahepatic cholestasis and sicca complex after thiabendazole. Gastroenterology 1983; 85:718-21. Hennekeuser HH, Pabst K, Poeplau W, Gerok W. Thiaben-

August

4.

5.

6.

7. 8. 9.

10.

11.

THIABENDAZOLE-INDIJCED

1987

dazole for the treatment of trichinosis in humans. Texas Rep Biol Med 1969;27:581-96. Goyos MP, Guinzberg AL, Berumen AH, Degollado JR. Cervantes LY. Colestasis intrahepatica por thiabendazol tratada con fenobarbital. Prensa Med Mex 1976:41:167-71. Zimmerman HJ. Agents employed in the treatment of infection and parasitic diseases. In: Zimmerman HJ, ed. Hepatotoxicity. New York: Appleton-Century-Crofts, 1978:468-509. Fink Al, MacKay CJ, Cutler SS. Sicca complex and cholangiostatic jaundice in two members of a family probably caused by thiabendazole. Ophtalmology (Kochester) 1979: 86:1892-6. Bloomer JR, Boyer JL. Phenobarbital effects in cholestatic liver disease. Ann Intern Med 1975;82:310-7. Jalota R, Preston JW. Severe intrahepatic cholestasis due to thiabendazole. Am J Trop Med Hyg 1974;23:676-8. Robinson HJ, Silber RH, Graessle OR. Thiabendazole: toxicological, pharmacological and antifungal properties. Texas Rep Biol Med 1969;27:537-60. Popper H. Problems in the pathogenesis of intrahepatic In: Gentilini P, ed. Second International Sympocholestasis. sium on Problems in Intrahepatic Cholestasis. Florence, Italy. Basel, Switzerland: S. Karger AG, 1978:1-7. Shulman HM. McDonald GB. Liver disease after marrow

transplantation. ogy of bone 1984:104-35. 12. Portmann Grune 13. Snover

In: Sale GE, Shulman marrow transplantation.

B. Neuberger

lesions.

LIVEK DAMAGE

In: Caine

JM. Williams

RV, ed.

Rr Stratton,

Liver

249

HM, eds. The patholNew York: Masson, R. lntrahepatic

bile duct

transplantation.

London:

1983279-87.

DC. Preese

DK, Sharp

Ascher

NL. Liver allograft

biopsy

in determining

HI,, Bloomer

rejection:

outcome

JR. Najarian

an analysis

of rejection.

JS,

of the use of

Am J Surg Pathol

1987:11:1-10. 14. Popper

H. Schaffner

cholangitis

and

F. Non-suppurative

chronic

hepatitis.

destructive Prog

Liver

chronic

Dis

1970;3:

336-54. 15. Walker

CO. Combes

promazine. 16. Zimmerman

JH. Drug-induced

etl. International Academic,

Symposium

MJ. Possible vations 4:691-8.

cirrhosis

induced

by chloro-

1966;51:631-40. hepatic

disease.

In: Eliakim

on Hepatotoxicity.

New

M.

York:

1974:92-103. P, M’eber J, Cutz

17. Valencia-Mayoral hepatic

B. Biliary

Gastroenterology

defect

dysplasia on

the

E. Edwards

in the bile secretory (Alagille’s

syndrome):

ultrastructure

of

liver.

apparatus a review

VD, Phillips in arteriowith

Hepatology

obser1984;