Causality assessment of herb-induced liver injury using multidisciplinary approach and Roussel Uclaf Causality Assessment Method (RUCAM).

Objective: To evaluate an algorithmic approach involving a multidisciplinary team for causality assessment of suspected herb-induced liver injury (HILI) cases and to compare the causality score using this multidisciplinary approach and Roussel Uclaf Causality Assessment Method (RUCAM).

Methods: A team consisting of hepatologist, clinical toxicologist, analytical toxicologist, and Chinese medicine (CM) pharmacist was formed to do causality assessment based on a protocol for suspected HILI cases. The likelihood of the diagnosis of individual case was first assessed systematically by a hepatologist and clinical toxicologist independently after collecting information about four aspects: (1) clinical course, (2) exclusion of alternative causes, (3) quality of the prescription and herbal product by examining the CM prescriptions and analysis of biological and herb samples, (4) scientific support on comprehensive literature review on English and Chinese medical database, and subsequently concluded in a consensus meeting held by the multidisciplinary team. The final causality score of each patient was compared with the likelihood of causality as assessed by RUCAM.

Results: Between 2005 and 2007, 48 consecutive patients with suspected HILI were enrolled and 21 patients were excluded due to the establishment of an alternative cause of liver impairment or the lack of any information on the herbs taken. Twenty-seven patients were recruited, among them 15 consumed Chinese herbal medicines, 10 used proprietary Chinese medicinal products, and 2 used both. The concordance between the causality assessment of the hepatologist and clinical toxicologist was moderate (weighted κ = 0.48, 95%CI 0.30-0.66). The causality assessment process concluded that the likelihood of HILI was "highly probable" in 5 cases and "probable" in 12, whereas there were 5 "highly probable" and 16 "probable" cases as assessed by RUCAM. The causality assessment by the multidisciplinary approach and RUCAM also showed moderate agreement (weighted κ= 0.51, 95%CI 0.22-0.81).

Conclusion: A multidisciplinary approach using defined algorithms is a scientific approach in causality assessment for HILI. Further study is needed to assess its accuracy and applicability.