Isoniazid-Associated Hepatotoxicity: A Clinicopathologic Study of 8 Cases
JS Do, MM Yeh. University of Washington, Seattle
Background: Tuberculosis (TB) remains an ongoing major health problem in the developed and developing world. Isoniazid (INH) has been one of the advances in treating TB; however, it has known hepatotoxicity, ranging from asymptomatic elevations of liver function tests, to acute hepatitis, or fulminant liver failure. A limited number of published data precludes chronic liver diseases as established risk factors for INH hepatotoxicity during TB therapy. We examined the pathologic features and clinical outcome of INH associated hepatotoxicity.
Design: 8 cases of documented (including 4 definite and 4 suspicious) INH-associated hepatotoxicity with available surgical pathology specimens were identified from our file. H&E and available special stained slides, clinical records, and laboratory data were reviewed.
Results: The 8 patients ranged from 23-69 years of age (mean: 44; M:F= 4:4). Time from initiation of INH therapy to development of symptoms/elevated LFT ranged from 1 to 6 months. AST ranged from 41 to 2451 (mean: 608); ALT ranged from 26 to 1591 (mean 500); alkaline phosphatase ranged from 99 to 530 (mean 213); total bilirubin ranged from 1.1 to 17.8 (mean 7.2). 3 definite and 1 suspicious cases for INH hepatotoxicity had no underlying liver disease. Histologically, 5 cases showed cholestasis, bridging necrosis and marked bile ductular reaction, whereas the remaining 3 cases showed mild lobular and focal portal inflammation. Acidophil bodies were common. In the 4 cases suspicious for INH hepatotoxicity, 3 had known underlying liver disease (1 had HCV/HIV and 2 were post liver transplant for HCV and/or HBV). 2 showed hepatitis and 2 showed extensive bridging necrosis and cholestasis on histology. Clinically, 2 definite cases for INH hepatotoxicity completely recovered after INH withdraw, whereas the other 2 definite cases for INH hepatotoxicity required liver transplant (1 died 1 day after transplant due to downhill hepatotoxicity, who also had concomitant non-alcoholic steatohepatitis and the other had no known adverse outcome after transplant).
Conclusions: While there is a spectrum of histology of INH-associated hepatotoxicity including cholestasis in addition to hepatitis and extensive liver necrosis, these observations may be obscured by concomitant liver disease. The common findings of marked bile ductular reaction is typically associated with extensive necrosis, suggesting activation of the regeneration compartment. As the outcome may be severe, caution should be exercised when prescribing INH, especially in patients with underlying liver disease.
Category: Liver & Pancreas
Monday, March 22, 2010 1:00 PM
Poster Session II # 179, Monday Afternoon