Herbal hepatotoxicity by Greater Celandine (Chelidonium majus): Causality assessment of 22 spontaneous reports

Toxic liver injury due to the herb Greater Celandine (GC) (Chelidonium majus L.) has been assumed in patients originating from various European countries and created concern. Based on regulatory and liver unspecific ad hoc causality assessments in 22 spontaneous cases of Germany, causality levels for GC were considered probable in 16 and possible in 6 cases. We now analyzed the data of these 22 cases regarding their causality levels employing the liver specific, standardized, structured and quantitative assessment method of the updated scale of CIOMS (Council for International Organizations of Medical Sciences). Causality for GC was found highly probable (n=2), probable (n=6), possible (n=10), unlikely (n=1), and excluded (n=3). Thus, causality could be upgraded in 2 cases to a highly probable causality level, but had to be down graded to excluded, unlikely, or possible causality levels in 3, 1, or 9 cases, respectively. GC hepatotoxicity shows a hepatocellular pattern of liver injury with female gender predominance. On average, age of the patients was 56.4years, treatment 36.4days, and latency period until first symptoms and jaundice 29.8 and 35.6days, respectively. This analysis therefore provides further evidence for the existence of GC hepatotoxicity as a distinct form of herb induced liver injury, but due to poor data quality the causal association between GC use and liver injury is less strong than hitherto assumed. We propose replacement of the regulatory organ unspecific by a liver specific causality assessment method in cases of herb induced liver injury as well as stricter pharmacovigilance strategies towards improvements of data quality. Toxicological studies are now warranted to elucidate the mechanism(s) of human GC hepatotoxicity that represents a European issue.