Narrative reviewindependent cross-checkfull text checked★ key study I-SUP-10 in-depth review
Which supplements harm the liver?
Liver injury from herbal and dietary supplements.
Overview
Navarro VJ et al.. Hepatology 2017 · PMID 27677775 · DOI 10.1002/hep.28813
Mainly bodybuilding steroids, green tea extract and multi-ingredient products.
Does the study support the specific claim? Partly · Methodological quality: moderate
Plain-language summary
Study explained simply
- What did the researchers want to find out?
Supplement liver injury.
- How was the study done?
Review.
- What was actually found?
In the US DILIN registry, 16-20% of drug-induced liver injury cases were attributed to HDS, mainly anabolic steroids, green tea extract and multi-ingredient products; simple vitamins/minerals were 6 of 85 non-steroid cases.
- How large was the observed effect?
About 3 HDS-related liver injury cases per 100,000 persons per year (Iceland estimate)
- What do the results mean in everyday life?
Avoid risky product types.
- What are the study's limitations?
Narrative.
- What does it say about Elawady's specific claim?
Partly supports; limits generalisation.
- Do the results apply to Germany and Europe?
Applicable.
- What can explicitly not be concluded from this study?
That all supplements, or standard vitamins, whey, creatine or omega-3, harm the liver; risk per user.
Scientific in-depth review
1. Full reference
- Authors
- Navarro VJ et al.
- Journal
- Hepatology
- Year
- 2017
- Identifier
- PMID 27677775 DOI 10.1002/hep.28813
- Verification
- verified – PubMed efetch 2026-10-25
- Retracted
- no
- Corrections
- none
- Extent checked
- full text
- Conflicts of interest and funding
Authors report no commercial conflicts of interest. Symposium funded by NIDDK/NIH with AASLD, with support from NIH ODS, NCCIH, NIEHS, USDA, CDC and the National Center for Natural Products Research (University of Mississippi).
2. Research question
Liver injury from herbal and dietary supplements
3. Study design and population
- Study design
- Narrative review – Narrative review / workshop summary of an AASLD-NIH research symposium (Bethesda, May 4-5, 2015), drawing mainly on the US DILIN Prospective Study and other registries
- Population
US cases of hepatotoxicity
- Country
USA
- Period
Symposium May 2015; DILIN data 2004-2014; published 2017
- Sample size
Not a primary study. Cited DILIN data: 130 HDS-related cases in the first 8 years (45 bodybuilding/anabolic steroid; 85 other HDS attributed to 116 products)
- Inclusion and exclusion criteria
Not applicable (narrative review; no systematic search or selection criteria reported)
4. Methods and endpoints
- Exposure / intervention
HDS
- Comparator
None
- Endpoints
Proportion of hepatotoxicity cases attributed to HDS, implicated product types, clinical phenotype, diagnosis, regulation
- Statistical methods
Narrative
5. Actual results
- Results
DILIN: HDS accounted for 16% of DILI cases overall, rising from 7% (2004-2005) to 19% (2010-2012) and 20% (2013-2014). Of 130 HDS cases, at least 45 were bodybuilding agents (anabolic steroid phenotype). Of the 85 non-steroid HDS cases: 14 (16%) single/multiple named herbal products, 7 (8%) traditional botanical mixtures, 6 (7%) simple vitamins/minerals/dietary supplements (e.g. niacin, multivitamins, levocarnitine), 58 (68%) multi-ingredient nutritional supplements (MINS); green tea was the suspected agent in 24 instances. Of 97 tested products, 49 contained green tea catechins, 29 of them not labelled as such. Spain: HDS share 2% (2006) to 13% (2010-2013); Asia 2.5% (India) to 73% (Korea). OxyELITE Pro outbreak: 36 cases in Hawaii, 1 death, 2 transplants.
- Absolute risks
No population-based US estimate. Best cited estimate (Iceland 2011-2012): overall DILI incidence 19 per 100,000 persons, 16% attributed to HDS, i.e. about 3 per 100,000 persons for HDS-related acute liver injury. Case fatality unknown because the number of users is unknown.
- Authors' conclusion
HDS liver injury is a growing problem; priorities are better epidemiology, accurate identification of injurious products, chemical/toxicological testing of ingredients, and stronger regulation (registration, full ingredient listing, adverse-event reporting, no unproven structure/function claims).
6. Strengths
- Expert symposium summary drawing on the prospective, adjudicated DILIN registry
- Detailed breakdown of implicated product categories
- Transparent funding and COI statement
7. Weaknesses
- Narrative review without systematic search, selection or appraisal methods
- Case-series/registry proportions only; no denominator of users, so no risk per user
- Causality for multi-ingredient products often uncertain; contamination and mislabelling common
- Proportions from referral registries are subject to referral and reporting bias
- US focus
8. Risk of bias
- Orienting tool
- none (narrative review; no standard tool fits)
- Type of assessment
- structured assessment based on the full text by a single reviewer (no formal double review)
- Full formal assessment
- no
- Overall risk of bias
- not assessable
- Rationale
Narrative review; no standard risk-of-bias tool applies. Structured single-reviewer judgement from full text; a formal assessment would require two independent reviewers and a protocol.
| Domain | Judgement | Rationale |
|---|---|---|
| Search and selection of evidence | high | No systematic search or selection criteria; content reflects symposium presentations. |
| Validity of underlying data (DILIN registry) | some concerns | Prospective adjudicated registry, but referral-based case series without user denominators; causality assessment for HDS is acknowledged to be unreliable. |
| Attribution of causality to specific ingredients | high | 68% of non-steroid cases due to MINS with unknown culprit; undeclared ingredients found in tested products. |
| Interpretation | low | Authors explicitly state the lack of population-based frequency estimates and do not overgeneralise to all supplements. |
Full-text review (25 October 2026): Added DILIN case numbers and product-category breakdown (simple vitamins/minerals 6/85 = 7%), time trend 7%->20%, Iceland population incidence (~3/100,000), funding/COI; structured narrative-review bias judgement added (overall not assessable).
9. Assessment of the statistical interpretation
Figures are shares of liver-injury cases attributed to HDS, not risk among users. The population incidence estimate (about 3 per 100,000 persons, Iceland) indicates HDS-related liver injury is rare in absolute terms; simple vitamins/minerals made up only 6 of 85 non-steroid HDS cases.
10. Relevance to the specific claim
11. Independent contrary findings
12. Transferability to Germany and Europe
Similar products are sold in Europe; the Spanish registry shows a rising HDS share (13% in 2010-2013). EU regulation of food supplements differs from US DSHEA.
13. Overall methodological judgement
Credible expert summary that supports product-specific liver risk (anabolic steroids, green tea extract, multi-ingredient products, adulterated products); it does not support the idea that all supplements harm the liver or that liver enzymes rise after any supplement.
| Multidimensional assessment | |
|---|---|
| Authenticity of the source | verified |
| Methodological quality | moderate |
| Risk of bias | not assessable |
| Precision of results | not applicable |
| Directness | direct |
| Reproducibility | replicated |
| Consistency with the overall evidence | consistent |
| Regional transferability | high |
| Adequacy of the interpretation (by the archive) | not applicable |
14. Open questions
Risk per user for specific product classes; responsible ingredients in MINS; extent of adulteration in EU-marketed products.