Drug-Induced Autoimmune Hepatitis: Robust Causality Assessment Using Two Different Validated and Scoring Diagnostic Algorithms

Abstract

Drug-induced autoimmune hepatitis (DIAIH) is a relatively new subtype of idiosyncratic drug-induced liver injury (iDILI), but the features of DIAIH have been variably described due to the inhomogeneity of assessed study cohorts. The aim of this analysis is to harmonize DIAIH cohorts by unifying causality assessments, which may help characterize the features of DIAIH. Methods: Published reports of DIAIH cases were evaluated for the causality assessment methods used to verify the diagnosis of DIAIH. This disorder consists of two parts, i.e., the iDILI part and the autoimmune (AIH) part, whereby each part needs a specific diagnostic algorithm. The validated and scoring Roussel Uclaf Causality Assessment (RUCAM) is privileged for assessing the iDILI part, and the validated, simplified AIH score is the perfect choice for evaluating the AIH part. The analysis of DIAIH publications revealed that 12/20 reports (60%) presented cases assessed by both the RUCAM and the simplified AIH score, providing 49 drugs and drug combinations as causative drugs in up to 25 cases of DIAIH. Serum alanine aminotransferase activities of up to 3489 UL and high titers of autoimmune parameters such as anti-nuclear antibodies, anti-smooth-muscle antibodies, and soluble liver antigen antibodies supported DIAIH diagnosis. In contrast, 4/20 reports (20%) applied only RUCAM, and 2/20 reports (10%) used only the simplified AIH score; these 6 reports therefore provided insufficient criteria for a valid DIAIH diagnosis. Moreover, 2/20 reports (10%) did not use any causality algorithm, providing elusive features of DIAIH. While DIAIH is clearly restricted to drugs as responsible agents, this term is erroneously used to refer to disease induced by non-drugs such as herbs, green tea, dimethoate (an organophosphate insecticide), dietary supplements, biologics, herbal remedies, different viruses, and bacteria, as well as vaccines. For diseases induced by these agents, a better term could be, for instance, non-drug-induced autoimmune hepatitis. Drug cessation and immunotherapy with corticosteroids and azathioprine comprise the treatment of choice. The characteristics of DIAIH can best be described if both the RUCAM and the simplified AIH score are used concomitantly.

Keywords: DIAIH; autoimmune hepatitis; idiosyncratic DILI; simplified AIH score; updated RUCAM.