Abstract: Objective To observe clinical characteristics of herb-induced liver injury (HILI).Methods A total of 187 inpatients in our hospital between September 2015 and June 2016 were prospectively enrolled in the study, whose general condition, medical history, clinical manifestations, biochemical indices, prognosis and diagnosis differences using integrated evidence chain-based causality identification algorithm (iEC), structured expert opinion process (SEOP) and the Roussel Uclaf Causality Assessment Method (RUCAM) were analyzed.Results Among the patients, 137 (73.3%) were females aging from 18 to 81 years old, with majority in 40~59 years old. The median time from the application of Chinese herbs to the occurrence of liver injury was 60 days. Moreover, 139 cases (74.3%) were classified as hepatocellular injury. The most important 5 diseases, herbs for which were most likely to cause HILI, were gastritis (8.0%), disc herniation (5.3%), insomnia (4.8%), hair loss (4.8%), and psoriasis (4.3%). Among the Chinese herbs and traditional patent medicines that lead to HILI, the top 5 Chinese herbal with hepatotoxicity were Polygonum multiflorum (n=21), Corydalis Rhizome (n=10), Rhei Radix et Rhizome (n=9), Bupleuri Radix (n=8) and Psoraleae corylifolia (n=7). In the HILI patients, there were 29 (15.5%) with cirrhosis, 30 (16.0%) with chronicity and 3 (1.6%) died. The clinical diagnosis rates using iEC and SEOP were 47.6% and 29.9%, respectively. The ratio of HILI/DILI diagnosed with iEC, RUCAM and SEOP were 23.2%, 48.7% and 14.6%, respectively.Conclusion Attention should be paid to herbal hepatotoxicity. Our study suggests that iEC method could improve the clinical diagnosis rate of HILI and decrease the radio of HILI to DILI.

Key words: Herb-induced liver injury, Herbs, Integrated evidence chain-based causality identification algorithm, Clinical manifestations, Prognosis