Does Milk Thistle Work? The Honest Evidence on Silymarin
Educational summary — not medical advice. Milk thistle is studied as an adjunct, not a treatment for liver disease; clear it with your clinician if you take antidiabetic drugs, are pregnant, or have a ragweed-family allergy.
The honest answer
Milk thistle's evidence is modest and specific. Start with the mortality question: a Cochrane review found no significant all-cause mortality benefit in cirrhosis, and the liver-mortality signal was not significant in high-quality trials (Rambaldi 2007). Its best-supported effect is reducing elevated ALT/AST in fatty liver disease (Li 2024) — statistically real, clinical relevance debated. A small fasting-glucose drop rests on low-quality trials (Voroneanu 2016). And the famous 93% Amanita survival is a hospital IV drug, not an oral capsule.
What "the active" is, and why the dose is always silymarin
Milk thistle's studied effects trace to silymarin, a complex of flavonolignans whose main component is silibinin (silybin). Trials dose in silymarin milligrams, not seed milligrams — which is why the number that matters is almost never the big figure on the front of the bottle (see the label trap). Every result below only means something if you actually get a real silymarin dose, which most products don't disclose.
What the trials show — by outcome
| Outcome | Design | Finding | The honest limit |
|---|---|---|---|
| Cirrhosis / mortality Rambaldi 2007 | Cochrane review (alcoholic + hepatitis B/C liver disease) | All-cause mortality RR 0.78 (not significant); liver-related mortality RR 0.57 (0.28–1.19) | No confirmed mortality benefit — the liver-mortality signal was not significant in high-quality trials. Lead here, not with enzymes |
| Liver enzymes / NAFLD Li 2024 · Zhong 2017 | Meta-analysis, 26 RCTs / 2,375 (Li) | ALT SMD −12.4, AST SMD −11.0 (Li 2024); concrete ALT −9.2 U/L, AST −6.6 U/L (Zhong 2017) | Statistically real, clinical relevance debated. An enzyme drop isn't a proven change in liver outcomes — the honest read is "reduces elevated ALT/AST," not "treats liver disease." |
| Type 2 diabetes Voroneanu 2016 | Systematic review + meta-analysis (adjunct to standard care) | Fasting glucose −26.9 mg/dL, HbA1c −1.07% | Low-quality trials — a promising adjunct signal, not established. Also why it can add to antidiabetic drugs |
| Amanita poisoning (IV) Mengs 2012 | Observational, 1,491 cases — intravenous Legalon SIL | ~93% survival with the IV silibinin protocol | This is a hospital IV drug, not an oral OTC capsule. A supplement does not reproduce it — don't treat a capsule as an antidote |
Read together: the case is modest and specific — a reliable-ish reduction in elevated ALT/AST whose clinical meaning is debated, a low-quality glucose signal, and no confirmed survival benefit in cirrhosis. A narrative review of 29 RCTs is consistent, finding enzymes fell in about 65.5% of studies but without a pooled effect size (Calderon Martinez 2023). It's a mild adjunct, not a liver treatment — and no trial supports a "detox" or "cleanse" claim.
Safety and interactions
Milk thistle is generally well tolerated; the main reported side effects are mild gastrointestinal upset (loose stools, nausea). Two cautions are worth respecting:
Who it's most reasonable for
- People wanting a mild adjunct for elevated liver enzymes — e.g. NAFLD — alongside, not instead of, the diet, weight, and medical management that actually move the disease.
- People who'll get a real silymarin dose — a disclosed 80%-standardized extract at ~200 mg silymarin 2–3×/day, not an undisclosed "1,000 mg" softgel. See the dosage guide.
- Not a substitute for prescribed treatment of liver disease or diabetes, and not an antidote for mushroom or any acute poisoning.
Skip or get medical advice first if you: take antidiabetic drugs (possible additive lowering), have a ragweed-family allergy, or are pregnant or breastfeeding.
Frequently asked questions
Does milk thistle lower liver enzymes?
Modestly — its best-supported effect. Silymarin reduced ALT/AST in NAFLD (SMD ~−12.4/−11.0, 26 RCTs; Li 2024), or ~−9.2/−6.6 U/L in concrete units (Zhong 2017). Statistically real, clinical relevance debated.
Does it help you live longer with cirrhosis?
No confirmed benefit. Cochrane found all-cause mortality RR ~0.78 (NS) and liver-mortality not significant in high-quality trials (RR 0.57, 0.28–1.19; Rambaldi 2007). Lead with this null.
Does it help blood sugar?
Maybe, on low-quality evidence: fasting glucose ~−26.9 mg/dL, HbA1c ~−1.07% as an adjunct in type 2 diabetes (Voroneanu 2016). Promising, not established — and a reason to monitor if on antidiabetic drugs.
Does it cure mushroom poisoning?
The 93% Amanita survival (Mengs 2012) is an intravenous hospital drug (Legalon SIL), not an oral capsule. A supplement doesn't reproduce it and is not an antidote. Mushroom poisoning is an ER emergency.
Related
- Milk thistle: what it is & who it's for
- The silymarin label trap — why the big milligram number isn't the dose
- Milk thistle dosage guide — the ~420 mg/day silymarin target + a calculator
- Best milk thistle — every product by disclosure and cost per 100 mg silymarin
Sources
- Rambaldi A, et al. "Milk thistle for alcoholic and/or hepatitis B or C virus liver diseases." Cochrane Database Syst Rev. 2007. PMID: 17943794 (all-cause mortality RR 0.78 NS; liver-mortality not significant in high-quality trials).
- Li S, et al. "Administration of silymarin in NAFLD/NASH: A systematic review and meta-analysis." Ann Hepatol. 2024. PMID: 38579127 (26 RCTs, 2,375; ALT SMD −12.4, AST SMD −11.0).
- Zhong S, et al. "The therapeutic effect of silymarin in the treatment of nonalcoholic fatty disease: a meta-analysis." Medicine (Baltimore). 2017. PMID: 29245314 (ALT −9.2 U/L, AST −6.6 U/L).
- Voroneanu L, et al. "Silymarin in type 2 diabetes mellitus: a systematic review and meta-analysis of RCTs." J Diabetes Res. 2016. PMID: 27340676 (fasting glucose −26.9 mg/dL, HbA1c −1.07%; low quality).
- Mengs U, et al. "Legalon SIL: the antidote of choice in patients with acute hepatotoxicity from amatoxin poisoning." Curr Pharm Biotechnol. 2012. PMID: 22352731 (intravenous silibinin; 1,491 cases, ~93% survival).
- Calderon Martinez E, et al. "Milk thistle (silymarin) and liver enzymes: a systematic review." Cureus. 2023. PMID: 38021897 (29 RCTs; enzymes fell in 65.5% of studies, no pooled effect size).
- Fenclová M, et al. "Poor chemical and microbiological quality of commercial milk thistle supplements." Sci Rep. 2019. PMID: 31366891 (content variance 35–125% of declared).
- NCCIH, "Milk Thistle" (nccih.nih.gov) and MedlinePlus, "Milk thistle" (medlineplus.gov) — Asteraceae/ragweed allergy caution; generally well tolerated, mild GI effects. Consumer-health references, not clinical trials.