Drug-induced liver injury is a major problem of pharmacotherapy and is also frequent with antidepressive psychopharmacotherapy. However, there are only few studies using a consistent methodologic approach to study hepatotoxicity of a larger group of antidepress ants. We performed a quantitative signal detection analysis using data from the Uppsala Monitoring Centre from the WHO that records adverse drug reaction (ADR) data from worldwide sources; we retrieved substance- and country-specific (Australia, France, Germany, Italy, Spain, the United Kingdom, and the United States) ADR data and calculated reporting odds ratios as measures for disproportionality within a case/noncase approach. To allow for identification of agents that cause severe forms of hepatotoxic ADRs, we used 2 terms of the MedDRA ("drug-related hepatic disorders-comprehensive search" [DRHD-CS] and "… -severe events only" [DRHD-SEO]). Distribution of signals was heterogeneous throughout the different data sets, and consistent findings were present for only a few substances: agomelatine (AGM) and tianeptine as well as both positive control agents (amineptine, nefazodone) generated signals related to DRHD-CS and DRHD-SEO in all analyzed data sets. Tri- and tetracyclic antidepressants (here amitriptyline, clomipramine, mianserin, mirtazapine, trimipramine) were associated with hepatotoxicity in several data sets. Using 2 MedDRA terms did not allow for detection of agents that cause severe hepatotoxic ADR. Our results support the findings of previous, primarily literature-based, systematic analyses of hepatotoxicity related to antidepressive psychopharmacotherapy. No new safety information could be generated. Application of 2 MedDRA terms did not increase the substance-specific safety information.
Keywords: adverse drug reaction; antidepressant; data mining; disproportionality analysis; drug safety; liver enzymes; pharmacovigilance.
© 2015, The American College of Clinical Pharmacology.