Objective: Review nonclinical and clinical trial data for hepatic effects of duloxetine. Methods: Review studies of toxicology, metabolism, mitochondrial effects, and clinical trials. Results: Nonclinical studies revealed no treatment-related transaminase elevations and no effects of duloxetine on mitochondrial beta-oxidation in rat hepatocytes. In patients with a normal baseline alanine transaminase (ALT), duloxetine was associated with elevated transaminases > 3X ULN in about 1% of patients. ALT and aspartate transaminase values peaked at 8 weeks, alkaline phosphatase steadily increased to maximum value at Week 52 and mean total bilirubin values were not increased. Hepatic-related treatment-emergent adverse events were uncommon. Seven of 23,000 duloxetine- and 2/6000 placebo-treated patients met criteria for modified Hys rule (significant elevation of both ALT and total bilirubin) but were complicated by contributing factors such as excessive alcohol consumption (n=3), gall stones, common bile duct calculus, hepatitis C, and liver adenocarcinoma (n=1 each). Conclusions: Duloxetine has an effect on the liver, manifested by transient, self-limiting transaminase elevations. Rare events characterized as hepatocellular injury, cholestatic injury, or mixed type of hepatic injury have been reported. The pattern of liver effects was different from that in laboratory animals.