Milk thistle benefits: what human trials actually found

Liver disease: mixed results, and negative where claims are loudest

The liver is where milk thistle's reputation was built, and it is also where the trials disagree most. The active complex, silymarin (led by silibinin), has been tested almost entirely in people who already had a liver diagnosis. Nobody has shown it protects a healthy liver. For background, see our milk thistle topic page.

A 2017 meta-analysis of randomized trials in diagnosed liver disease pooled six studies and found silymarin lowered ALT by 0.26 IU/mL and AST by 0.53 IU/mL against control. The authors judged that clinically negligible, and gamma-GT did not change. The trials varied widely in dose and extract, and quality was low.

The two largest placebo-controlled trials were flatly negative. In 200 people with alcoholic cirrhosis, 450 mg/day of silymarin for up to two years had no effect on survival or the clinical course of disease. In 154 people with chronic hepatitis C, silymarin at up to 2,100 mg/day for 24 weeks did not beat placebo on liver enzymes or viral levels.

Fatty liver is the brighter spot. A 2024 meta-analysis of 26 trials in NAFLD or NASH found improvements in lipid markers and liver enzymes. In a subset of trials with biopsy data, it also found better liver histology. The authors still asked for further confirmation, and a 2018 review reached a similar verdict about liver disease overall: stronger trials are needed.

Combination products add little clarity. A 4-week pilot in 15 people with a liver condition reported better enzyme and lipid markers with silymarin and curcumin, but its abstract describes no placebo group. A 3-month placebo-controlled trial of a six-ingredient mixture in 113 people found no difference from placebo.

Diabetic kidney disease and blood sugar: two trials that disagree

A 2012 placebo-controlled trial in 60 people with type 2 diabetes and kidney disease gave 420 mg/day of silymarin for three months. A urine-protein marker of kidney damage fell more than with placebo. Markers of inflammation and oxidative stress were lower too.

A larger 2017 trial in 102 people with the same condition found no significant difference in kidney function decline, dialysis need or death. Hospitalizations were lower with silymarin. The primary outcomes of the two trials do not agree, and a surrogate marker moving is not the same as kidneys lasting longer.

On blood sugar, the 2024 pooled analysis of 33 trials (1,943 people) reported lower fasting glucose, fasting insulin and HbA1c with silymarin than with control. The populations were mostly people with diabetes or another metabolic condition, and the authors want larger, longer trials. None of this supports using milk thistle in place of diabetes medication.

Cholesterol, blood pressure and weight

The same 33-trial analysis is the main source here. Against control, total cholesterol fell by about 14 mg/dL, triglycerides by about 26 mg/dL and LDL by about 17 mg/dL. HDL did not change significantly.

Blood pressure showed a small signal: diastolic pressure was about 1.25 mmHg lower, while systolic pressure did not differ significantly. That is a very modest gap. Body weight and BMI did not change either, so the weight-loss claim has no support from silymarin alone.

One 52-week trial in 136 obese adults with type 2 diabetes did find less abdominal fat and lower BMI. It used a berberine-plus-silymarin product, and berberine has its own metabolic effects, so the result cannot be credited to milk thistle.

Chemotherapy side effects: small trials, no cancer outcomes

No trial has tested milk thistle against cancer itself. Two small randomized trials asked whether silymarin, given alongside chemotherapy, eases its side effects.

• Breast cancer, 30 people: 420 mg/day for one month lowered chemotherapy-related liver damage graded on ultrasound. Blood-based liver tests showed no significant difference.
• Colorectal cancer, 70 people: silymarin around chemotherapy cycles was linked to less diarrhea and nausea, with no difference in liver toxicity.

The two trials measured different things and found different effects. Neither measured tumor response or survival.

Inflammation and heart surgery

The broad inflammation marker, C-reactive protein, did not differ between silymarin and control across the 33 pooled trials. Narrower results exist in specific groups. The 2012 diabetic kidney trial found lower urinary inflammatory and oxidative-stress markers.

A 2024 study of 160 people undergoing coronary bypass surgery also found a signal. Those given 400 mg of silymarin beforehand had fewer cases of postoperative atrial fibrillation (14 versus 30) than those who got none. The study does not state that assignment was randomized or blinded, which limits how far the result can be trusted.

Mood, stress, anxiety and sleep

A 2023 systematic review of 17 studies found that silymarin and silibinin shift brain chemistry, inflammation markers and neurogenesis. These are measurements typical of animal and lab work. The authors themselves say clinical studies are needed. We found no human trial of milk thistle for depression, anxiety or sleep.

The closest human data are negative. A placebo-controlled trial in 43 adults with gambling disorder gave 300-600 mg/day of silymarin for eight weeks. It did not separate from placebo on gambling, depression or anxiety scores. Both groups improved by a similar amount, which the authors attributed to a placebo response.

Skin: acne, eczema and rosacea

Skin trials mostly test creams and serums, not the capsules sold as milk thistle supplements. They are small, short, and often lack a placebo.

• Acne: a 4-week open-label study of a 0.5% silymarin serum in 22 people reported less acne severity, with no placebo group. Industry-reported studies of a silymarin serum (56 and 40 people) were also uncontrolled. In a 3-month split-face trial of 30 people, acne improved on both the 1.4% silymarin-cream side and the salicylic-acid-peel side.
• Oral silymarin for acne: one 60-person trial compared it with doxycycline. It is the only oral-form trial we found, and we do not draw a conclusion from it here.
• Rosacea: a one-month placebo-controlled trial in 46 people found a silymarin-plus-MSM cream improved redness, papules, itching and hydration. The two-ingredient formula hides silymarin's own contribution.
• Eczema: a 2-week trial in 40 people compared a two-herb cream with a prescription steroid, which cannot show what silymarin does alone. A mouse study is the only silymarin-specific evidence.

Absorption, safety and claims without human data

A 2021 systematic review found silymarin's active compounds are poorly absorbed orally, because the body processes and excretes them quickly. A 2005 review found that binding silybin to phosphatidylcholine improves absorption compared with plain extract. In rats, piperine raised silybin availability by 146-181%, but that is an animal finding, not a human one.

Safety data are reassuring. A 2019 review found silymarin generally well tolerated, even at high doses for 24 weeks, with gastrointestinal upset the main limit. An NIH monograph notes rare allergic cross-reactivity with ragweed-family plants. For drugs, a 14-day study of an extract found no significant effect on four major metabolizing enzymes. A separate trial of 420 mg/day found it changed losartan levels in a genotype-dependent way. See our guide to milk thistle interactions and side effects.

Several popular claims have little or no human support:

• Detox or cleanse: not a measurable trial outcome, and the best test of a cleansing effect, the high-dose hepatitis C trial, was null.
• Iron overload: in 8 pooled trials of people with thalassemia, silymarin lowered blood ferritin, but the smaller liver-iron analysis showed no significant reduction. The authors found no evidence it reduces iron load.
• Infections: only lab-dish studies, at concentrations well above what a swallowed dose is known to reach.
• Menopause: only rat and bench studies; no human trial of any symptom.
• IBS and bowel disease: only a three-ingredient product in a retrospective study, and a review that lists silymarin without a specific result.
• Milk supply: no valid clinical trial support.