Herbal and Dietary Supplement-Induced Liver Injury Compared With Conventional Drug-Induced Liver Injury: A Retrospective Cohort Study.

Abstract

Introduction:  Herbal and dietary supplements (HDS) are widely used globally and across the Middle East, yet their hepatotoxic potential remains underrecognized in this region. Comparable Bahrain-specific prevalence data are limited. This study aimed to quantify the proportion of drug-induced liver injury (DILI) attributable to HDS in a Bahraini cohort and to compare the clinical features and outcomes of HDS-related versus conventional medication-related liver injury, with severe liver injury as the primary outcome. We hypothesized that HDS-related DILI would be associated with more severe clinical outcomes than conventional medication-related DILI.

Methods:  We conducted a retrospective cohort study of adults admitted with suspected DILI to Bahrain Government Hospital in Manama, Bahrain, between January 2018 and December 2024. Cases underwent structured causality assessment using the updated Roussel Uclaf Causality Assessment Method (RUCAM), and patients with RUCAM scores ≥6 for at least one implicated agent were included. Patients were classified into HDS-related or conventional medication-related liver injury groups according to the agent with the highest RUCAM score; cases with equal scores or unresolved competing exposures were excluded. Alternative causes of liver injury, including viral hepatitis and autoimmune liver disease, were excluded through routine clinical evaluation and available laboratory and imaging data. The primary outcome was severe liver injury, defined as coagulopathy, hepatic encephalopathy, or liver transplantation. Multivariable logistic regression was performed to identify independent predictors of severe liver injury.

Results:  A total of 712 patients met the inclusion criteria, of whom 168 (23.6%) were attributed to HDS and 544 (76.4%) to conventional medications. Patients with HDS-related liver injury were younger (median age 34 vs. 52 years, p = 0.002) and more commonly female (63.2% vs. 41.1%, p < 0.001). Hepatocellular injury was more frequent in the HDS group (68.5% vs. 45.2%, p = 0.02). Severe liver injury occurred in 49/168 (29.2%) of HDS cases versus 79/544 (14.5%) of medication-related cases, corresponding to an absolute risk difference of 14.7% (95% CI 7.2-22.2). On multivariable analysis, HDS exposure remained independently associated with severe liver injury (adjusted OR 2.3, 95% CI 1.1-5.1, p = 0.03). Liver transplantation (7.7% vs. 1.7%, p = 0.08) and mortality (10.7% vs. 4.8%, p = 0.21) were numerically higher in the HDS group, although these differences did not reach statistical significance.

Conclusions:  In this retrospective Bahraini cohort, HDS accounted for nearly one-quarter of RUCAM-confirmed DILI cases and was associated with a higher likelihood of severe liver injury than conventional medications. These findings support routine, structured inquiry about HDS exposure in patients presenting with liver injury and suggest that standardized medication and supplement history tools may improve early recognition. Larger prospective studies with product-level characterization are needed to better define the risks associated with HDS use in the region.