Background and objectives: In 21 published case reports, the use of the herb Greater Celandine (GC) (Chelidonium majus L.) has been causally related to liver injury, but a variety of confounding variables were evident that might have offset causality. This study reanalyses causality levels in these cases with a liver-specific causality evaluation method.
Methods: All 21 cases were submitted to the liver-specific, standardized, structured, quantitative and updated scale of the Council for International Organizations of Medical Sciences. This scale considers, among other items, latency period, course of alanine aminotransferase after treatment discontinuation, risk factors, comedication and alternative causes.
Results: Using this method for assessment, causality for GC was highly probable in two and probable in six cases, with lower causality grading in the remaining 13 cases. In these patients, causality for GC was possible in 10 cases and excluded in three cases. On the basis of the eight cases with highly probable and probable causality gradings, GC hepatotoxicity represents an idiosyncratic reaction of the metabolic type, whereas immunologic or obligatory hepatotoxic features are lacking. In some cases, alternative diagnoses and poor data quality were confounding variables that reduced causality levels.
Conclusion: Confounding variables reduced causality levels for GC in reported cases of liver injury, but there is still striking evidence for herb-induced liver injury by GC with high causality gradings. GC hepatotoxicity is caused by an idiosyncratic reaction of the metabolic form, but there is uncertainty with respect to its culprit(s).