A comprehensive update in herbal and dietary supplement–induced liver injury

INTRODUCTION

In the United States, over 100,000 over-the-counter herbal and dietary supplements (HDS) are available both in stores and online, contributing to the potential for HDS-induced liver injury (HILI).¹ HDS includes a diverse spectrum of constituents like vitamins, minerals, proteins, herbs, and botanical essences and is now a well-recognized potential cause of both acute and chronic liver injury.^1–3 Unlike conventional medicines, HDS products are sometimes contaminated with unclear safety profiles due to adulteration and mislabeling.⁴

Cases of HILI reported to the US Drug-Induced Liver Injury Network have increased from 7% of all DILI cases in 2005 to 20% in 2014.^2,5 Diagnosing HILI is challenging due to scarce assessment tools, unclear injury patterns, and uncertainty about the culprit ingredients that cause liver damage. The aim of this review is to provide clinical insights and updates on the diagnosis and management of HILI.

CURRENT TRENDS IN HERBAL AND DIETARY SUPPLEMENT–INDUCED LIVER INJURY

The incidence of severe liver injury from HDS, as per the Acute Liver Failure (ALF) Study Group database, rose from 12.4% (1998–2007) to 21.1% (2007–2015).⁶ Despite a lower rate of comorbid conditions, patients with HDS-related ALF exhibit higher transplantation and mortality rates than cases due to prescription medication.⁶ Similarly, in a 2015 population study, over 18% of ALF cases were attributed to HDS, with 50% resulting in death or liver transplant.⁷

In the prospective study of DILI from the NIH-funded Drug-Induced Liver Injury Network, liver injury attributable to HDS increased from 7% in 2004–2005 to 20% by 2013–2014.^3,5 In Europe, studies also show a growing tendency for HDS use.³ Notably, despite widespread HDS usage in Asia, the incidence varies significantly, ranging from as high as 70% in Singapore⁸ to 2.5% in India.⁹

Amid the COVID-19 pandemic, interest in HDS spiked, propelling US sales to over $10 billion in 2020, a 17.3% surge from the prior year—the highest in 2 decades. Recent statistical data reveal that more than 50% of US adults use HDS daily.¹⁰ Use is more prevalent among females, non-Hispanic Asian and non-Hispanic White, individuals aged 70 or older, and those with conditions like stroke, obesity, arthritis, and respiratory problems.^5,11,12 Interestingly, use increases with income and education among adults.^5,11,12

In the United States, HDS is regulated by the Food and Drug Administration (FDA)’s Center for Food Safety and Applied Nutrition through the Dietary Supplement Health and Education Act of 1994. Per the Dietary Supplement Health and Education Act, HDS products are not subject to the same stringent premarket safety and efficacy testing as pharmaceuticals.¹³ Additionally, HDS products are exempt from mandatory investigational clinical trials for efficacy documentation. Federal law mandates HDS manufacturers to provide a comprehensive list of dietary ingredients. Regardless, many HDS products have been withdrawn from the market due to discrepancies between the listed dietary ingredients and the actual product formulation. Recent research has revealed a 51% mislabeling rate for dietary ingredients, notably suggesting that mislabeled HDS products could harbor undisclosed ingredients while omitting listed substances.⁴ At times, these products were found to be contaminated with heavy metals and pesticides.¹⁴ The FDA prohibits making explicit claims about the treatment or prevention of specific diseases. Nevertheless, manufacturers frequently draw consumer interest by using unsubstantiated statements and consumers may not understand the level of regulatory supervision imposed on HDS.

NAVIGATING CLINICAL CONUNDRUMS AND INSIGHTS IN DIAGNOSING HEPATIC INJURY FROM HERBAL AND DIETARY SUPPLEMENTS

There is no specific serum marker to aid in diagnosing HILI. The skill of history-taking holds paramount importance and entails obtaining information on supplement usage, initiation and cessation dates, dosage alterations, data pertaining to recovery after discontinuation (dechallange), and re-exposure (rechallenge)¹ (Figure 1).

FIGURE 1.

A stepwise approach to diagnosis of herbal and dietary supplement–induced liver injury. Adapted from Navarro et al³ and Fontana et al.¹⁵ Abbreviations: ALT, alanine amniotransferase; ALP, alkaline phosphatase; AST, aspartate aminotransferase; HDS, herbal and dietary supplements; INR, international normalised ratio; ULN, upper limit normal.

Causality assessment methods have been developed to evaluate the probability that a particular drug or HDS is responsible for liver injury¹ (Table 1). However, use is rather limited in HILI due to the use of multi-ingredient HDS, adulteration, mislabeling, and simultaneous intake of pharmaceuticals and other over-the-counter medications.¹ Detailed review of clinical data remains the gold standard for diagnosis and causality assessment.

TABLE 1.

Causality assessment instruments^16–19

	RUCAM (Roussel Uclaf Causality Assessment Method	RECAM (Revised Electronic Causality Assessment Method	The Maria‐Victorino (CDS)	The Digestive Disease Week–Japan 2004 (DDW‐ J)
Criteria	Time to onset from the beginning of the drug/herb Course of ALT after cessation of the drug/herb Percentage difference between ALT peak and ULN Risk factors Concomitant drug(s)/herb(s) Exclusion of Alternative Causes Previous hepatotoxicity of the drug/herb Response to unintentional re-exposure	Onset after drug start and onset after drug stop Dechallange or washout Literature supporting liver injury Exclusion of competing diagnoses	Temporal relationship between drug intake and onset of clinical picture Exclusion of alternative causes Extrahepatic manifestations Intentional or accidental re-exposure to the drug Previous report in the literature of cases of DILI associated with drug	The score is a modification of the RUCAM with the inclusion of DLST results and peripheral eosinophilia.
Total score and resulting causality grading	≤ 0, excluded; 1–2, unlikely; 3–5, possible; 6–8, probable; ≥ 9, highly probable	< −4, unlikely; 3 to −3, possible; 7–4, probable; ≥ 8, highly likely;	< 6, excluded; 6–9, unlikely; 10–13, possible; 14–17, probable; > 17 definite;	≤ 0, excluded; 1–2, unlikely; 3–4, possible; 5–8, probable; > 17 definite;
Specific operational information	Prospective use, retrospective use is less accurate; Calculated individually for each co-administered product; For idiosyncratic, not for intrinsic reactions; Applicable only for acute liver injury, not for pre-existing chronic liver disease; Used in HDS-HILI;	Better than RUCAM at diagnostic extremes Has not been validated in HILI	Not used widely in clinical practice because it was shown to be inferior to the RUCAM	Not currently used outside of Japan because of the lack of widely available and reproducible DLST assays.

Abbreviations: ALT, alanine aminotransferase; CDS, Clinical Diagnostic Scale; DLST, drug‐lymphocyte stimulation test; HDS, herbal and dietary supplements; HILI, HDS-induced liver injury; ULN, upper limit normal.

LiverTox, created in 2012 by the National Institute of Diabetes and Digestive and Kidney Diseases, is a great resource and tool for physicians, researchers, and patients. It offers up-to-date insights into the potential hepatoxicity of over 1000 medications and HDS with likelihood scales for liver injury.²⁰ LiverTox incorporates the Drug-Induced Liver Injury Network’s 5-point categorization as “likelihood scores” as part of its comprehensive approach to assess the likelihood of HILI. This classification relies heavily on published literature, making it more precise for medications that have been used widely but less so for newer drugs or herbals with limited use.

Several in vitro assessment tools have been developed to help aid in the diagnosis of DILI²¹ (Table 2). Genetic and immunologic susceptibility plays an important role in the development of HILI.¹ A strong link has been observed between liver injury related to green tea extract consumption and the HLA allele B*35:01.²² This allele was identified in 72% of the patients with HILI due to green tea extract and in over 90% of those in whom the injury was categorized as highly probable or certain.²² Allele carriers are younger at presentations and have shorter latency periods, higher median alanine aminotransferase values, and more severe presentations.²² Notably, HLA-B*35:01 has also been linked to the liver injury induced by Polygonum multiflorum,²³ Garcinia cambogia,²⁴ and Turmeric.²⁵ A missense variant (rs2476601) in PTPN22, which has been correlated with various autoimmune disorders, poses a risk factor for all-cause DILI in diverse racial and ethnic populations.^26,27

TABLE 2.

Investigational in vitro assessment tools to help aid in the diagnosis of HILI

In vitro test	Process	Comment
LTT	PBMCs are incubated with questionable hepatotoxin. Lymphocyte proliferation is measured by interferon-gamma release or 3H-thymidine	Low sensitivity and specificity; Has not been standardized or validated; Integral component of DDW-J 2004 scale
Cluster of differentiation 69 (CD69)	Incubated with implicated hepatotoxin. CD69 expression monitored	Validation studies not published
Monocyte-derived hepatocyte-like cells	Incubating MH cells with questionable hepatotoxin. Toxicity is analyzed by measuring the release of lactate dehydrogenase	92.3% sensitivity and 100% specificity; Has not been externally validated;

Note: Weber S et al.²¹

Abbreviations: LLT, lymphocyte transformation test; PBMC, peripheral blood mononuclear cells.

EXPLORING THE COMMON HEPATOTOXIC CULPRITS: HERBAL AND DIETARY SUPPLEMENTS UNVEILED THROUGH PATTERNS OF LIVER INJURY

HDS generally falls into 2 categories: single ingredient (Table 3), involving 1 plant product, or multi-ingredient formulations (Table 4). Multi-ingredient HDS poses diagnostic challenges due to complex mixtures, nuanced pharmacology with potential synergies, high rate of mislabeling, adulteration, substance omission, and even pesticide contamination, all contributing to diverse host risks.^4,14 Green tea, an ancient drink, has a good safety profile when consumed as a beverage but may pose a risk for liver injury when consumed in the extract form.²⁸ The primary component, epigallocatechin gallate, is known to cause potential hepatotoxicity, especially with prolonged use, fasting intake, and doses exceeding 800 mg.²⁸ The use of organic solvents in extraction, like methanol or chloroform, adds another layer of risk.²⁸ Even caffeine, commonly found in weight loss or performance-enhancing supplements, can amplify the hepatotoxicity of stimulants.²⁸ Yohimbine extract, prevalent in weight loss supplements, has been linked to a higher incidence of severe outcomes.²⁸ This caution extends beyond individual components, emphasizing the potential synergistic toxicity of additives.

TABLE 3.

Commonly used hepatotoxic HDS in the market

Ingredient	Chemical structure	Plant and region	Marketed Use	Purported benefit	Latency	Adverse effects	Pattern of liver injury	outcome	DILIN Class
Ashwagandha, “Indian ginseng”	Steroidal lactone	Withania Somnifera India and Southeast Asia	Energy enhancement	Adaptogen, anxiolytic, anti-diabetic, anti-inflammatory, aphrodisiac and anti-aging;	2–12 wk	Gastrointestinal upset, diarrhea, nausea, and vomiting	Cholestatic or mixed pattern of injury, jaundice, and pruritus	No deaths or transplants	Category C likelihood score
Black cohosh	Triterpenes glycosides and polyphenols	Actaea racemosa/Cimicifuga racemose North America	Menopause symptom relief	Malaise, gynecological disorders, kidney disorders, malaria, rheumatism, and sore throat;	2–12 wk	Hypotension, bradycardia, central nervous system effects, nausea, and vomiting	Hepatocellular, autoimmune hepatitis	Liver failure, transplant, and death reported	Category A likelihood score
Curcumin/Turmeric	Polyphenol	Curcuma longa India Southern Asia and Central America	Indigestion	Anti-inflammatory, antioxidant, antimicrobial, and even antineoplastic effects;	4–12 wk	Dermatitis and gastrointestinal upset	Hepatocellular or mixed or autoimmune hepatitis	Death reported	Category B likelihood score
Garcinia cambogia/Garcinia gummi-gutta	(−)−Hydroxycitric acid	Malabar tamarind tree India, Nepal and Sri Lanka	Weight loss	Anti-inflammatory and appetite suppressant;	12–48 wk	Serotonin syndrome, rhabdomyolysis	Mostly hepatocellular pattern, rarely mixed	Liver failure, transplant, death, chronic HILI reported	Category B likelihood score
GTE epigallocatechin-3-gallate (EGCG)	Catechin-Polyphenol	Camellia sinensis South and Southeast Asia	Weight loss	Chemoprotective antiproliferative and antioxidant;	1–64 wk	Headache, dizziness and nausea	Hepatocellular	Most recovery, Liver failure, transplant, and death reported	Category A likelihood score
Kratom	Tetracyclic indole and pentacyclic oxindole alkaloids	Mitragyna speciosia, Southeast Asia (Thailand, Myanmar, and Malaysia)	Analgesic	Anxiolytic, antitussive, antidiarrheal, stimulant, analgesic, opiate withdrawal;	1–8 wk	Respiratory depression, aggression, hallucinations, delusions, insomnia, vomiting, severe withdrawal, hypertension, dyspnea, confusion, seizure, coma, death	Mixed or cholestatic injury	No deaths from acute liver failure or transplants	Category B likelihood score

Note: DILIN class, that is, likelihood of causing clinically significant liver injury and the pattern of resultant liver injury. The likelihood score was obtained from the DILI Network LiverTox database. It is a 5‐point scale (A–E) that estimates whether a medication is a cause of liver injury: A = well‐known cause; B = highly likely cause; C = probable cause; D = possible cause; E = unlikely cause; E* = suspected but unproven cause; X = unknown.

Abbreviations: GTE, green tea extract; HILI, HDS-induced liver injury.

TABLE 4.

Commonly encountered hepatotoxic multi-ingredient HDS

Supplement	Marketed Use	Latency	Suspected hepatotoxins	Pattern of liver injury	Likelihood score
Herbalife	Weight loss, digestive health, boosting immunity, heart health, energy and fitness	8-36w	Saw palmetto Green tea extract	Hepatocellular (most common), mixed, cholestatic, sinusoidal obstruction syndrome(rare)	A
OxyELITE Pro	Weight loss, body building, performance enhancement aid	2-20w	Aegeline	Hepatocellular, submassive or massive necrosis, autoimmune hepatitis-like	A
Hydroxycut	Weight loss, body building “fat burning”	2-12w	Ashwagandha Ephedra Green tea extract	Hepatocellular, submassive or massive necrosis, cholestatic or mixed hepatitis(rarely)	B
Slimquick	Weight loss	3-12w	Green tea extract, Turmeric	Hepatocellular	C

DILIN class, i.e., likelihood of causing clinically significant liver injury; and the pattern of resultant liver injury. Likelihood score obtained from the DILI Network LiverTox database.It is a 5‐point scale (A to E) that estimates whether a medication is a cause of liver injury: It is a 5-point scale (A to E) that estimates whether a medication is a cause of liver injury: A = well‐known cause; B = highly likely cause; C = probable cause; D = possible cause; E = unlikely cause; E* = suspected but unproven cause; X = unknown.

Multi-ingredient supplements like OxyELITE-pro, OxyELITE-Pro New Formula (OxyELITE-pro-NF), and Hydroxycut led to a myriad of complications that mandated their removal from the market. Chemical analysis often identified potentially hepatotoxic ingredients like caffeine and yohimbine, alongside green tea extract and Garcinia cambogia, as potential culprits.²⁸ Significantly, notwithstanding the removal of original Hydroxycut formulations from the market in 2009, reformulations are presently available. New, albeit rare, cases of liver injury linked to Hydroxycut are still reported, leading to cholestatic liver injury and even vanishing bile duct syndrome.²⁸

STRATEGIES OF MANAGEMENT AND PROGNOSIS OF HERBAL AND DIETARY SUPPLEMENT–INDUCED LIVER INJURY

HILI is typically idiosyncratic in nature, with a latency period of 2–24 weeks.^1,15,29 The pattern of liver injury is more commonly hepatocellular and depends on the interplay between host, environmental factors, and culprit ingredients.^1,15,29 Suspected agents should be promptly discontinued and reported through the FDA’s medical product safety reporting program (MedWatch https://www.fda.gov/safety/medwatch-fda-safety-information-and-adverse-event-reporting-program).

Approximately 80% of patients fully recover without long-term sequelae.^1,15 However, patients with ALF characterized by coagulopathy and encephalopathy require urgent referral to a liver transplant center due to their limited likelihood (~25%) of spontaneous recovery. Alarmingly, despite similar disease severity (Model for End-Stage Liver Disease scores), patients with HDS-induced ALF show lower transplant-free survival and higher transplantation rates at 21 days compared to allopathic medicine–induced ALF cases.³⁰ Patients with pre-existing liver disease, lower serum albumin levels upon presentation, and higher total bilirubin and INR levels are at increased risk of adverse hepatic outcomes.¹ Conversely, patients exhibiting granulomas and eosinophil infiltrates on biopsy tend to have a higher likelihood of spontaneous recovery.¹

General supportive care is the standard of care for HILI. In cases of ALF with early-stage encephalopathy, a 3-day N-acetylcysteine regimen has been shown to significantly improve transplant-free survival from 27% to 58% in adults.¹ Corticosteroids for 1–3 months may benefit patients with autoimmune phenotype and plasma cell predominance on biopsy, but empirical use is discouraged, especially in those with cholestatic liver injury patterns, as it may cause harm.¹ Like idiosyncratic DILI, there is no specific antidote for liver injury caused by HDS.

SUMMARY

HILI accounts for 20% of DILI cases in the United States. Identifying and preventing HILI poses substantial challenges due to diverse presentations, diagnostic complexities, ingredient identification hurdles, limited treatment options, and lack of regulatory oversight of HDS products to confirm ingredients and ensure safety. Further advances in the identification of at-risk genetic phenotypes, greater oversight from regulatory organizations, and specific treatments for HILI are needed.

CONFLICTS OF INTEREST

Dina Halegoua-DeMarzio consults for and received grants from Novo Nordisk and Madrigal. She received grants from Akero, Viking, Intercept, and Galectin. The remaining authors have no conflicts to report.

EARN CME FOR THIS ARTICLE

https://cme.lww.com/browse/sources/222