Methods: Data were included from a number of sources. Laboratory and adverse event clinical trial data as of 09 December 2005 (>29,000 patients) were analyzed. Post-marketing spontaneous reports possibly related to hepatic events reported as of 2 August 2006 were categorized as to clinical relevance and probability of being related to duloxetine. Adverse events recorded in the FDA Adverse Event Reporting System (AERS) database for duloxetine were compared to those for all drugs and other antidepressants. Hepatic related events for duloxetine were compared to those for venlafaxine using an insurance claims database. National or institutional review boards at each study site approved the protocols and all patients provided signed informed consent prior to study participation.

Results: Clinical trial data revealed that duloxetine was associated with transient elevations of transaminases >3 times the upper limit of normal (usually peaked at 8 weeks, not associated with bilirubin elevations, not more likely in patients already having pre-existing elevations) in about 1% of patients. Among >5 million patients who have taken duloxetine (>1.5 million patient years), the reporting frequency of hepatic failure cases possibly (not probably) related to duloxetine use, is 1 per 500,000 persons exposed, or 0.6 per 100,000 person-years. AERS data revealed that virtually all hepatic related adverse events disproportionately reported with duloxetine are already known and described in the package insert. The claims database revealed that the hepatic profile of duloxetine is similar to that of venlafaxine.

Conclusion: Duloxetine has an effect on the liver, which in the vast majority of cases is clinically benign. More serious cases, including liver failure have been reported, but these appear to be similar to the background rate in the general population.