Milk thistle supplement guide: forms, labels, testing and price

Start with what the label actually measures

Most of the confusion in this category comes from one number: the milligrams printed on the front. Raw milk thistle seed contains a modest and variable amount of silymarin, the flavonolignan complex (led by silibinin) that the research is about. That is why the human trials used standardized extracts rather than ground seed.

A label reading 'milk thistle extract 1,000 mg' with no percentage tells you the weight of the material, not its concentration. A 250 mg capsule standardized to a high silymarin percentage can deliver more of the active complex than a 1,000 mg capsule with no stated standard. Look for a line such as 'standardized to 80% silymarin,' then do the multiplication yourself.

For background on the plant and what is known about it, see our milk thistle topic page.

The dose worth measuring against

Several human trials used 420 mg a day of silymarin. That figure appears in a small trial in people with diabetic kidney disease, in a trial in breast cancer patients on chemotherapy, and in a drug-interaction study in healthy volunteers. It is typically split into three 140 mg doses.

That is a reference point, not a recommendation, and the evidence behind it is mixed. A 2017 pooled analysis in people with liver disease found silymarin lowered ALT and AST by an amount the authors considered clinically negligible. A 2-year trial of 450 mg a day in alcoholic cirrhosis found no effect on survival. A larger trial in chronic hepatitis C found that doses up to 2,100 mg a day did not beat placebo on liver enzymes or viral levels.

Some pooled analyses are more encouraging, particularly in fatty liver disease, but their authors call for confirmation. The practical point for shopping is narrower: if a product claims to be 'research-based,' check whether a serving comes anywhere near the amounts studied, expressed as silymarin.

Forms compared

Standardized extract capsules and tablets are the form used in nearly all human research. Everything else is an extrapolation, which is not the same as being bad, but it does change how much you can verify.

A 2021 systematic review found that silymarin's active compounds are poorly absorbed by mouth, largely because the body processes and excretes them quickly. That limit applies to every form. The only format engineered around it is the phytosome, which a 2005 review found improved silibinin absorption compared with plain silymarin extract.

Powders, capsules and gummies, in practice

Powders. 'Powder' can mean two things. A capsule filled with powdered standardized extract is functionally the standard form. A bag of ground seed is closer to tea: you are getting whole seed with an unstated and variable silymarin content, and no trial describes what a spoonful delivers.

Capsules. These can concentrate a meaningful amount of extract in a small volume, which is why they dominate the research. Check how many capsules make up one serving. A bottle of small capsules taken three at a time empties faster than the capsule count suggests.

Gummies. A gummy has to stay chewable and palatable, which limits how much extract it can hold. Most carry less than the amounts used in trials, and many state a total extract weight without a silymarin percentage. Products vary, so read the specific label rather than assuming. Check the gelling agent too: gelatin is not vegan, while pectin typically is.

Gummies are not necessarily useless, but with no trial behind the format and a lower dose per serving, they are the hardest form to defend on evidence.

What 'third-party tested' can and cannot tell you

The phrase has no single legal definition on a supplement label. It might mean a lab confirmed silymarin content, heavy metals and microbial counts. It might also mean one lab ran one test once. Labeled and actual ingredient amounts do not always match across the supplement category, which is why the document behind the badge matters more than the badge.

• A batch-specific certificate of analysis (COA) that names what was tested (identity, silymarin content, metals, microbes) and shows a number against a limit, not just a pass logo.
• A named public standard. USP publishes monographs for milk thistle raw material, capsules, tablets and powdered extract. The raw material must contain at least 2.0% silymarin (as silybin) on a dried basis. Finished capsules and tablets must contain 90% to 110% of the silymarin amount on their own label.
• Not the same as organic. USDA organic certification covers growing and handling. It says nothing about silymarin content or potency.

A product can be tested without referencing USP, so absence of the name is not an automatic failure. Absence of any document is a different matter. For more detail on seals, see third-party tested milk thistle.

Price per effective dose, not per bottle

Divide the bottle price by the number of servings at the label's directed dose. Then divide again by the silymarin each serving delivers. A $15 bottle of 60 capsules taken one at a time is $0.25 a serving. A $28 bottle of 90 capsules taken three at a time is 30 servings, or about $0.93 each. The cheaper bottle is only cheaper if the capsules contain comparable silymarin.

Price differences usually track standardization, extraction method, phytosome formulation and testing. No dose-response trial has compared milk thistle at different amounts, so a higher price is not evidence of a stronger effect. Our price-per-serving guide walks through more examples.

Label red flags

• No silymarin percentage. A bare milligram figure cannot be converted into a dose.
• Proprietary blends. If milk thistle shares a capsule with dandelion, artichoke, turmeric or other botanicals, each should list its own amount. A combined total hides what you are getting.
• 'Detox' or 'cleanse' language. No trial has measured either in a general population. The human research looked at specific lab markers in people with diagnosed liver or kidney conditions.
• Milk supply claims. An NIH reference monograph reports no valid clinical trial support for milk thistle increasing milk supply.
• Absorption claims with no mechanism. Piperine raised silibinin availability by 146% to 181% in rats, but that is an animal and cell-line result, not a human trial. Treat 'enhanced with black pepper' as unproven for people.
• A testing badge with nothing behind it. If a COA cannot be requested or found, the claim cannot be checked.

Organic and non-GMO claims are agricultural, not potency claims. They are fine to have, but they do not substitute for a standardization percentage.

Safety checks before you buy

A 2019 safety review found silymarin generally well tolerated in humans, including at high doses for 24 weeks, with gastrointestinal upset the main limiting side effect. The NIH monograph notes occasional GI or skin reactions and rare allergic cross-reactivity in people sensitive to ragweed-family plants.

On medications, the picture is low risk but not zero. A standardized extract taken three times daily for 14 days had no significant effect on four major drug-metabolizing liver enzymes in healthy volunteers. A separate trial of 420 mg a day for 14 days changed losartan blood levels, with the effect depending on each person's CYP2C9 genotype. If you take prescription drugs or have a liver condition, mention the supplement to a pharmacist or doctor before starting.

Also read the other-ingredients line. Capsule shells, flow agents and carrier oils (often soybean oil in softgels) matter if you avoid a specific ingredient. Gelatin capsules are not vegan, while HPMC or cellulose capsules are.