Hepatitis is one of the most common adverse effect from anti-tuberculosis. Silymarin showed its efficacy to decreased serum alanine transaminase enzyme in animal models from recent study. No confirmed this efficacy was performed in human.

A prospective, double-blind, placebo-controlled trial was carried out according to Good Clinical Practice Guideline. This study is to define the efficacy of silymarin to prevent hepatotoxicity from anti-tuberculosis drugs. Informed consent is obtained prior to the study. New patients diagnosed with tuberculosis are enrolled. Patients with liver diseases, current alcohol drinking more than 20 g/day, regular use of herbal or other potential hepatotoxic drugs are excluded. Patients are treated with a standard regimen of four anti-tuberculosis therapy. They will randomize to receive either placebo or silymarin (140 mg) thrice daily. Liver function test (LFT) and clinical changes are assessed at 2- and 4-week after initiation of the treatment. DILI from anti-tuberculosis drugs ('atb-DILI') is defined as: i) a rise of alanine aminotransferase (ALT) to 2 times above normal upper limit, or ii) an elevation of total bilirubin more than 2 mg/dl with or without ALT elevation. The study endpoints are the level of ALT by week 4 and the number of patients who developed atb-DILI.

Statistical analysis is used to compare the differences in ALT and number of atb-DILI

- Prevention of antituberculosis-related drug induced liver injury with silymarin is investigated.

Inclusion Criteria:

• tuberculosis cases
• treated with isoniazid, rifampicin, ethambutol and pyrazinamide

Exclusion Criteria:

• no known liver disease (HBV, HCV), and HIV infection
• normal ALT level before enrollment
• refuse to participate