Liver Supplements: What Actually Works, What's Hype, and What I'd Skip Entirely

Why I Stopped Trusting the Liver Supplement Aisle

The $2 billion ‘detox’ problem

Here’s the core tension nobody selling these products wants to name. Your liver is a detox organ. That’s its job description. It runs roughly 500 metabolic functions, filters about 1.4 litres of blood per minute, and it does this whether or not you paid $39.99 for a bottle of dandelion powder. So a capsule marketed as a “liver detox” is selling you a biological function you were born with.

The category is enormous anyway. Liver and detox supplements pull in well over a billion dollars annually in the US alone, and the shelf space keeps growing.

What this article covers (and what it won’t)

My position, stated up front so you know where I’m coming from: a handful of ingredients have genuine human randomized trial support, most have rodent data dressed up as proof, and a few actively damage the organ they claim to protect.

That last part isn’t hyperbole. Navarro and colleagues, working with the US Drug-Induced Liver Injury Network, have documented that herbal and dietary supplements now account for roughly 20% of DILI cases in the registry, up from about 7% in the early 2000s. Read that again. One in five cases of drug-induced liver injury in a national registry traces back to something bought without a prescription, often something bought for liver health.

The framework I use throughout: human trials beat mechanism, every time. And hard endpoints (liver fat measured by MRI-PDFF, fibrosis staging, histology) beat enzyme numbers alone. A compound that drops your ALT by 8 points while your liver fat sits unchanged has done something to a blood marker, not necessarily to your liver.

What I won’t do is pretend everything works or that nothing does. Both positions are lazy.

Your Liver 101: What You're Actually Trying to Support

Phase I and Phase II detoxification, explained simply

Think of Phase I as unwrapping a package and Phase II as bagging it for pickup.

Phase I (mostly cytochrome P450 enzymes) chemically opens up a toxin, and here’s the awkward part: the intermediate product is often more reactive than what you started with. Phase II then conjugates that intermediate, attaching sulfate, glucuronide, or glutathione so it can leave via bile or urine.

A supplement that accelerates Phase I without supporting Phase II is just leaving trash on the counter. That’s not a theoretical worry, it’s the reason certain “activating” botanicals make me nervous.

Bile, glutathione, and fat metabolism

Glutathione is the liver’s master antioxidant, and the rate-limiting piece is almost always cysteine availability, not glutathione itself. Which matters enormously for how you’d supplement (more on the oral glutathione question later, because it’s a mess).

Bile acids carry fat-soluble waste out through the gut. Sluggish bile flow, cholestasis in clinical language, is the actual mechanism behind several supplement claims that otherwise sound like folklore.

What liver enzymes (ALT, AST, GGT, ALP) actually tell you

ALT is the most liver-specific of the four. Typical lab ranges run up to 40-55 U/L, though there’s a solid argument that the true healthy upper limit sits closer to 30 for men and 20 for women. AST is less specific (muscle releases it too, so a hard leg day can bump it). GGT tracks alcohol and biliary irritation. ALP points toward bile duct issues.

The caveat that matters more than any reference range: normal enzymes do not rule out fatty liver or fibrosis. Plenty of people with significant steatosis, and some with advanced fibrosis, have a perfectly boring ALT.

Which brings up the elephant. MASLD, metabolic dysfunction-associated steatotic liver disease (renamed from NAFLD in 2023 by a multi-society consensus), affects roughly 30% of adults worldwide according to Younossi’s global epidemiology work. Nearly a third of all adults. That’s the population these products are really aimed at.

Do You Actually Need a Liver Supplement?

Who has a legitimate case

Some people have a real reason to be here:

• Elevated ALT or GGT on bloodwork, confirmed on a repeat draw
• Diagnosed MASLD or MASH
• Metabolic syndrome, type 2 diabetes, or visceral obesity
• Heavy alcohol history, current or past
• Long-term use of hepatotoxic medications (methotrexate, amiodarone, some antiretrovirals, high-dose isotretinoin)
• High-dose anabolic steroid use, especially oral 17-alpha-alkylated compounds

Who’s wasting money

Healthy adults with normal labs buying “insurance.” I understand the impulse. But you’re paying to support an organ that isn’t struggling, and you’re adding compounds to a system that has to metabolize them.

And then there’s the crowd taking two capsules before a night out to “protect” the liver. That’s not how any of this works. Nothing on the shelf meaningfully changes acetaldehyde exposure or hepatocyte stress from a heavy drinking session.

Red-flag symptoms that mean labs, not capsules

Jaundice. Dark urine. Right-upper-quadrant pain. Unexplained itching. Easy bruising or bleeding gums. Any of these means bloodwork and imaging, not shopping.

Now the position I’ll defend hardest in this entire article: a supplement is a rounding error compared with body weight reduction. Vilar-Gomez and colleagues, publishing in Gastroenterology in 2015, followed 293 patients with biopsy-proven steatohepatitis through 52 weeks of lifestyle intervention. Among those who lost 10% or more of body weight, 90% had resolution of steatohepatitis and 45% showed fibrosis regression. No capsule in this article comes within a mile of that.

Before you spend a dollar, get the baseline: a CMP with ALT, AST, ALP and bilirubin, plus GGT, ferritin, fasting insulin, and a calculated FIB-4 score. FIB-4 is free, uses age, ALT, AST and platelets, and gives you a rough fibrosis risk read in about ten seconds.

The Ingredients With Real Human Evidence (Ranked)

Tier 1: Strong human RCT support

Vitamin E (800 IU/day RRR-alpha-tocopherol). The strongest evidence in this entire category isn’t an exotic botanical. It’s a vitamin your grandmother took.

The PIVENS trial, published in the New England Journal of Medicine in 2010 by Sanyal and colleagues, randomized 247 non-diabetic adults with biopsy-proven NASH to vitamin E, pioglitazone, or placebo for 96 weeks. Histological improvement: 43% on vitamin E versus 19% on placebo. Statistically significant, biopsy-confirmed, and it beat the pharmaceutical arm on the primary endpoint.

So why isn’t everyone on it? Two safety signals at that dose. The SELECT trial (Klein et al., JAMA 2011) found a 17% relative increase in prostate cancer incidence with 400 IU/day in healthy men. And a 2010 meta-analysis in the BMJ by Schürks and colleagues found a small but real increase in hemorrhagic stroke risk with vitamin E supplementation. Neither finding is enormous in absolute terms. Both are enough that I wouldn’t take 800 IU indefinitely without a biopsy-confirmed reason and a physician tracking it.

Omega-3 EPA/DHA. Multiple meta-analyses converge here. Parker and colleagues pooled the early trials in the Journal of Hepatology (2012) and found consistent reductions in hepatic fat content. The WELCOME trial (Scorletti et al., Hepatology 2014) gave 4 g/day of DHA-rich EPA+DHA for 15 to 18 months and saw improvement in liver fat percentage, though not in fibrosis markers.

The pattern across trials: you want 2 to 4 g/day of combined EPA plus DHA, and you want at least six months. The 1 g fish oil capsule with 300 mg of actual omegas isn’t running the same experiment.

Silymarin (milk thistle). It earns Tier 1 placement for MASLD specifically and Tier 3 for almost everything else, which is a weird enough situation that I gave it its own section below.

NAC (N-acetylcysteine). Here’s my favourite argument-ender: NAC is a hospital antidote. When someone arrives in the emergency department after an acetaminophen overdose, intravenous NAC is standard of care, because it restores hepatic glutathione and it demonstrably prevents liver failure. Nothing else in the supplement aisle can make that claim.

For chronic use, the data is smaller. Khoshbaten et al. (Hepatitis Monthly, 2010) gave 600 mg twice daily to NAFLD patients for three months and reported ALT improvement. Doses in the literature run 600 to 1,200 mg/day. Cheap, well tolerated, and mechanistically coherent since cysteine is the glutathione bottleneck.

Tier 2: Promising but incomplete

Berberine. The metabolic route does the work here. Think of AMPK as your cells’ fuel gauge, and berberine as a thumb pressing on it. Yan and colleagues published a randomized trial in PLoS One in 2015 comparing berberine plus lifestyle intervention against lifestyle alone in patients with NAFLD, and the berberine group showed significantly greater reductions in hepatic fat by imaging, alongside better insulin sensitivity and lipids. Typical dose: 500 mg two or three times daily with meals. GI upset is common at first.

Choline. Underappreciated, and I’ll say that plainly. Zeisel’s group at UNC has shown repeatedly in controlled feeding studies that choline deprivation reliably produces hepatic steatosis, and that repletion reverses it. Choline is needed for VLDL export, which is literally how fat leaves the liver. Adequate intake is 425 mg/day for women and 550 mg/day for men, and most people, especially postmenopausal women and anyone avoiding eggs and organ meats, sit below it. If you eat two eggs a day you’re probably fine. If you don’t, this is worth a look before anything exotic.

Artichoke leaf extract. Panahi and colleagues ran a randomized trial published in Phytotherapy Research in 2018: 600 mg/day of artichoke leaf extract for two months in NAFLD patients produced reduced liver fat on ultrasound and lower ALT and AST versus placebo. One trial, modest size, but a clean design and a plausible bile-flow mechanism.

Curcumin (bioavailability-enhanced only). Pooled RCT data show modest ALT/AST and liver fat reductions, with Rahmani et al. (2016) reporting a notably large liver fat improvement using an amorphous dispersion formulation. The irony you should know about: curcumin itself has accumulated DILI case reports, particularly with piperine-boosted and phytosome products in genetically susceptible people. A compound with liver injury reports being sold for liver health is exactly the kind of thing this category does.

Probiotics and synbiotics. I wanted these to work. Then the INSYTE trial (Scorletti et al., Gastroenterology 2020) gave a synbiotic to 104 NAFLD patients for a year, successfully changed the gut microbiome, and changed liver fat by essentially nothing. Meta-analyses of smaller trials look friendlier, which usually tells you something about small-trial bias. Tier 2 at best, and only just.

Tier 3: Mechanistically interesting, clinically unproven

Vitamin D. Deficiency correlates with MASLD severity in observational data. Supplementation trials mostly fail to move liver fat. Correct a deficiency because deficiency is bad, not because you’re treating your liver.

TUDCA. Interesting mechanism, thin human data, dedicated section below.

Choline-adjacent methyl donors, betaine, silybin phytosomes at unusual doses: all plausible, all under-tested in humans with hard endpoints.

Quick-reference comparison table

Ingredient	Typical studied dose	Best evidence	Effect size	Who it’s for	My confidence (1-5)
Vitamin E (RRR-alpha)	800 IU/day	PIVENS, NEJM 2010	43% vs 19% histological improvement	Biopsy-proven MASH, non-diabetic, physician-supervised	4
EPA + DHA	2-4 g/day, 6+ months	WELCOME trial; multiple meta-analyses	Small-to-moderate liver fat reduction	MASLD with high triglycerides	4
Silymarin	140 mg 2-3x/day, or 200-400 mg phytosome	MASLD meta-analyses (positive), Cochrane (negative)	ALT down ~5-10 U/L	MASLD, mild enzyme elevation	3
NAC	600-1,200 mg/day	Acetaminophen antidote data; small MASLD RCTs	Modest ALT reduction	Glutathione support, oxidative stress	3.5
Berberine	500 mg 2-3x/day	PLoS One 2015 RCT	Meaningful liver fat + insulin sensitivity gains	Metabolic syndrome, prediabetes, MASLD	3.5
Choline	425-550 mg/day	Zeisel depletion-repletion studies	Corrects deficiency-driven steatosis	Low-egg diets, postmenopausal women	4 (for deficiency)
Artichoke leaf	600 mg/day	Phytotherapy Research 2018 RCT	Reduced ultrasound steatosis, lower ALT	MASLD, sluggish bile symptoms	2.5
Curcumin (enhanced)	500-1,500 mg/day equivalent	Pooled RCTs	Modest ALT/AST and fat reduction	MASLD, with DILI awareness	2.5
Vitamin D	1,000-4,000 IU/day	Mixed RCTs	No clear liver-specific effect	Documented deficiency only	1.5
Probiotic/synbiotic	Varies	INSYTE (null), smaller positive RCTs	Minimal on liver fat	Nobody, for liver purposes	1.5
TUDCA	250-1,000 mg/day	Kars et al., Diabetes 2010	Insulin sensitivity yes, liver fat no	Cholestatic patterns, oral AAS users	2

The Ingredients I'd Skip (And Why the Marketing Is Better Than the Data)

‘Liver detox’ and cleanse blends

The word “detox” on a liver product is my single fastest filter. If a formulator understood hepatic biochemistry, they’d talk about conjugation capacity, bile flow, or lipid export. “Detox” is a marketing word, not a physiological one.

Oral glutathione

Reduced glutathione is a tripeptide, and the gut is exceptionally good at hydrolyzing tripeptides into their component amino acids. Which means most of what you swallow becomes glycine, glutamate and cysteine before it reaches your bloodstream. Richie and colleagues (European Journal of Nutrition, 2015) did find increased body glutathione stores at 250 and 1,000 mg/day over six months, so I won’t call absorption zero. But cysteine delivery via NAC or whey protein costs less and has a cleaner rationale. Paying a premium for the assembled molecule when the limiting brick is cheap makes no sense to me.

Dandelion root, burdock, and the folk-medicine tier

These appear in nearly every blend on the market. Controlled human trials with liver outcomes? Essentially none. Dandelion has diuretic activity and a long traditional use record, burdock and yellow dock similar, and that’s the entire case. Traditional use is a reason to study something. It isn’t evidence of effect.

Proprietary blends and fairy dusting

Let me show you the arithmetic, because it’s the most useful thing in this section.

A popular formula lists a “Liver Support Complex, 500 mg” with 12 ingredients. Milk thistle needs roughly 420 mg/day of standardized extract to match trial dosing. NAC needs 600 mg. Artichoke needs 600 mg. Berberine needs 1,000 mg or more.

Even if milk thistle took the entire 500 mg blend (it doesn’t, it’s listed third), you’d still be short. Divide 500 mg across 12 ingredients and the average is 42 mg each. That’s not a dose. That’s a label decoration.

More ingredients usually means a worse product, not a better one. Twelve items on a panel signals that the formulator optimized for the ingredient list photograph, not for the amount of anything reaching your hepatocytes.

Oh, and the olive-oil-and-lemon-juice “gallstone flush”? The stones people proudly photograph in the toilet were analyzed and reported in the Lancet in 2005 by Sies and Brundert: they were saponified fatty acid soap, formed from the olive oil in the gut. Not gallstones. No cholesterol, no bilirubin, no calcium.

Milk Thistle: The Most Studied Liver Supplement, Honestly Assessed

What silymarin and silybin actually are

Silymarin isn’t one molecule. It’s a flavonolignan complex from Silybum marianum seeds, containing silybin (also written silibinin), silychristin, silydianin and isosilybin. Silybin is the most active fraction. Standardized extracts are typically 70 to 80% silymarin, which is the number you should be checking on a label.

Where the evidence is strong

Two places, and they’re specific.

MASLD populations. Meta-analyses of randomized trials in fatty liver disease show consistent, if modest, ALT and AST reductions, with some showing improvement in ultrasound steatosis grade. Not dramatic. Reproducible.

And Amanita phalloides poisoning, where intravenous silibinin (Legalon SIL) is standard supportive care in Europe and available through expanded access in the US. It blocks OATP1B3-mediated uptake of amatoxin into hepatocytes. That’s a genuine, mechanism-specific, life-saving application. It also has nothing to do with your daily capsule.

Where it disappoints

Now the uncomfortable part.

The Cochrane review by Rambaldi and colleagues examined milk thistle across 13 randomized trials in alcoholic and hepatitis B or C liver disease. No convincing effect on mortality, complications, or liver histology. When they restricted analysis to trials with adequate methodology, the effects shrank further.

Then the SyNCH trial, published in JAMA in 2012 by Fried and colleagues. This one deserves attention because it was designed to give silymarin its best shot: 154 hepatitis C patients who had failed interferon therapy, randomized to placebo, 420 mg, or 700 mg of silymarin three times daily for 24 weeks. Doses far above typical retail. Result: no meaningful difference in ALT or HCV RNA versus placebo.

So milk thistle appears to help metabolic fatty liver and not to help viral or alcoholic liver disease. That’s a narrower claim than any bottle makes.

Dose, form, and the bioavailability problem

Bioavailability is the recurring theme in this whole category, and silymarin is the poster child. Plain silymarin is poorly water-soluble and poorly absorbed, with oral bioavailability estimated in the low single digits. Phytosome forms (silybin bound to phosphatidylcholine) achieve substantially higher plasma silybin levels, on the order of several-fold in pharmacokinetic comparisons.

Practical stuff. Either 140 mg of a standardized 80% extract, two to three times daily, or 200 to 400 mg/day of a phytosome form. Take it with a meal containing fat.

My take: worth including in a stack, not worth expecting miracles from. It’s the aspirin of liver supplements. Broadly reasonable, occasionally useful, never the answer on its own.

TUDCA and Bile Acids: The Bodybuilder Favorite

How bile acid supplements are supposed to work

TUDCA is tauroursodeoxycholic acid, the taurine-conjugated form of UDCA. And UDCA is a real drug, FDA-approved for primary biliary cholangitis, which is exactly why TUDCA sells so well. The supplement rides on the prescription’s reputation.

The proposed mechanism is genuinely interesting: shifting the bile acid pool toward less cytotoxic species, reducing endoplasmic reticulum stress in hepatocytes, and improving bile flow. That’s a cholestasis mechanism. Notice it isn’t a fatty liver mechanism.

What the human data actually shows

I want to be careful here because the online enthusiasm has badly outrun the evidence.

The most-cited human trial is Kars et al., published in Diabetes in 2010. Twenty obese participants took roughly 1,750 mg/day of TUDCA for four weeks. Hepatic and muscle insulin sensitivity improved. Liver fat content? Unchanged. That’s the headline trial, and it’s a null result on the endpoint most buyers care about.

Almost everything else in circulation is UDCA extrapolation or rodent work. Which isn’t nothing, but it isn’t a reason to spend $60 a month either.

Who it might genuinely help

Two groups, in my read.

People running oral 17-alpha-alkylated anabolic steroids, where cholestatic injury is the classic pattern. There’s no RCT here, just mechanism and community experience, but the mechanism fits the injury type better than anything else on the shelf.

And people with a cholestatic biochemical picture: high ALP, rising bilirubin, unexplained itching. Though if that’s you, the answer is a hepatology referral first and a discussion about prescription UDCA, not a capsule you found on Amazon.

Doses in trials and practice run 250 to 1,000 mg/day, usually split. Expect $40 to $70 a month for a legitimate, third-party-tested product, and expect a lot of underdosed or adulterated options below that price point.

Fatty Liver (MASLD): The Condition Most People Are Actually Trying to Fix

Let me name the elephant. When someone searches for liver supplements, roughly nine times out of ten they’ve either been told they have a fatty liver on ultrasound, or they’ve seen an ALT of 62 on a routine panel and started worrying. Metabolic dysfunction-associated steatotic liver disease (MASLD, formerly NAFLD) affects around 30% of adults worldwide. That’s the actual market.

Why insulin resistance is the real driver

Fatty liver isn’t a liver problem that happens to involve metabolism. It’s a metabolic problem that happens to show up in the liver.

Here’s the sequence. Insulin resistance in adipose tissue means fat cells stop holding onto their triglycerides properly, so free fatty acids flood the portal circulation. Meanwhile, hyperinsulinemia switches on de novo lipogenesis, the liver’s own fat-manufacturing pathway. The liver receives more fat than it can export as VLDL, and the excess parks itself in hepatocytes. Add oxidative stress and you get inflammation, then fibrosis.

Nothing in that chain is fixed by an antioxidant blend. Think of the liver as the drain at the bottom of a sink that’s overflowing. You can scrub the drain all you like. Somebody needs to turn off the tap.

What moves liver fat on an MRI

I like ranking things by effect size on hepatic fat fraction (measured by MRI-PDFF, the only endpoint I really trust), because it puts supplements in their proper place. Roughly, from strongest to weakest:

Intervention	Typical effect on liver fat
10%+ body weight loss	40 to 80% relative reduction; NASH resolution in ~90% (Vilar-Gomez, Gastroenterology 2015)
GLP-1 agonists / bariatric surgery	Large, comparable to major weight loss
Resmetirom (prescription)	~30% relative reduction, plus fibrosis benefit
5 to 7% weight loss	~30% relative reduction
Resistance or aerobic training without weight loss	10 to 20% relative reduction
Cutting sugar-sweetened beverages	20 to 30% in controlled feeding studies
Omega-3 at 3 to 4 g/day	~10 to 15% relative reduction
Vitamin E 800 IU (non-diabetic NASH)	Histologic improvement in 43% vs 19% placebo
Silymarin, berberine, choline	Small, inconsistent, or context-dependent

See the shape of that list? Every supplement sits below every behavior.

The supplement stack I’d build for MASLD

If you’ve handled the tap and you want the extra 10 to 15%, here’s what I’d defend in front of a hepatologist.

Omega-3s, 2 to 4 g/day EPA+DHA. The WELCOME trial (Scorletti et al., Hepatology 2014) gave 103 patients around 4 g of DHA-rich omega-3 for 15 to 18 months and found a modest but real reduction in liver fat, with DHA enrichment in erythrocytes predicting response. Meta-analyses agree the effect is small and reliable. Small and reliable beats big and imaginary.

Vitamin E, 800 IU/day, non-diabetics, biopsy-proven NASH, time-limited. PIVENS (Sanyal et al., NEJM 2010) is still the best evidence any supplement has in this space: 43% of the vitamin E arm improved histologically versus 19% on placebo over 96 weeks. It failed in diabetics. And I’d cap it at 12 months, for reasons I’ll get to in the safety section.

Berberine, 500 mg two or three times daily, if the metabolic markers are ugly. Yan and colleagues randomized 184 patients with NAFLD in 2015 and found berberine added to lifestyle changes reduced hepatic fat content beyond lifestyle alone, alongside improvements in HbA1c and lipids. It’s working upstream, on insulin sensitivity, which is exactly where I want a supplement working.

Choline, if your intake is low. Most people don’t hit the adequate intake of 425 to 550 mg/day, and choline depletion causes steatosis directly by impairing VLDL export. Fischer’s controlled depletion work in AJCN (2007) showed healthy adults developed fatty liver on low-choline diets and reversed it on repletion. Three whole eggs a day handles this for free.

Standardized silymarin, 420 to 700 mg/day. Wah Kheong et al. (Clinical Gastroenterology and Hepatology, 2017) used 700 mg three times daily for 48 weeks and saw fibrosis improvement without significant NAS change. Not spectacular. Not nothing.

Coffee: the free intervention nobody sells

I’ll be blunt: the strongest liver-protective agent in this entire article costs about forty cents a cup and nobody’s running Instagram ads for it.

Pooled cohort data consistently link 2 to 3+ cups of coffee daily with lower liver fibrosis, lower cirrhosis risk, and roughly 30 to 40% lower hepatocellular carcinoma incidence. Poole’s umbrella review in BMJ (2017) mapped this across dozens of outcomes and liver endpoints came out among the most consistent findings in the entire coffee literature. Decaf appears to help too, which points away from caffeine and toward chlorogenic acids and diterpenes.

If you drink coffee, keep drinking it. If you don’t, this isn’t a reason to start, but it should recalibrate how much you expect from a $60 bottle.

And one more piece of context: resmetirom (Rezdiffra) received FDA approval in March 2024 for NASH with moderate-to-advanced fibrosis, based on MAESTRO-NASH. Real drugs exist now. Supplements are adjuncts to a plan, not the plan.

How to Choose Liver Supplements Without Getting Fleeced

The ingredient matters less than the product. I’ve seen excellent evidence attached to garbage capsules more times than I can count.

Third-party testing that actually means something

Three seals I take seriously: NSF Certified for Sport, USP Verified, and Informed Choice. Each confirms the product contains what the label says at roughly the stated amount, screens for heavy metals and common contaminants, and (for the sport programs) checks against banned substance lists.

What none of them do is verify the supplement works. A USP seal on a useless ingredient means you’re getting an accurately dosed useless ingredient. Certification is a purity guarantee, not an efficacy one.

Absent a seal, ask for the Certificate of Analysis. A real company emails it within a day. A drop-shipper goes quiet.

Standardization and label literacy

This is where most money gets wasted.

“Milk thistle seed powder, 1,000 mg” tells you nothing about silymarin content, which can range from under 2% to over 80% depending on extraction. “Milk thistle extract, standardized to 80% silymarin, 250 mg” tells you exactly what you’re getting: 200 mg of active. The first bottle looks like more. It’s almost certainly less.

Same logic for “extract ratio 30:1” with no marker compound named. Thirty parts of what, concentrated to what specification? Without a standardized marker, that ratio is decoration.

For NAC, look for the actual milligram figure. For omega-3, ignore “fish oil 1,000 mg” and find the EPA and DHA numbers, which are often a pathetic 300 mg combined.

Price per effective dose

Do this calculation once and you’ll never shop the same way.

Say a bottle of silymarin is $24 for 120 capsules of 250 mg standardized to 80%. That’s 200 mg silymarin per capsule. The trial dose I’d aim for is 420 mg/day, so roughly two capsules daily, giving 60 days of use. That’s $0.40 per day, about $12 a month.

Now the “liver complex” next to it: $39 for 60 capsules, proprietary blend of 22 ingredients totaling 800 mg, milk thistle listed fourth. You cannot calculate the dose because the blend hides it. That’s the point of a proprietary blend.

Five label red flags

1. Proprietary blends. If they won’t tell you the doses, it’s because the doses are embarrassing.
2. “Detox” or “cleanse” in the product name. A marketing decision, not a physiological one.
3. 25+ ingredient formulas. Fairy dusting. Each ingredient is present in an amount too small to matter.
4. No COA available on request. Walk away.
5. Claims about curing hepatitis, cirrhosis, or “reversing” liver damage. Illegal claims signal a company that doesn’t fear the FDA, which tells you what else they’re willing to do.

One more reason to be paranoid: Tucker and colleagues analyzed FDA data in JAMA Network Open (2018) and identified 746 dietary supplements adulterated with unapproved pharmaceutical ingredients between 2007 and 2016. Weight loss and sexual enhancement products dominated, but “cleanse” and detox categories showed up repeatedly. Sibutramine in a product marketed for liver health is not a hypothetical.

Dosing, Timing, and Stacking

Label directions are written by marketing departments. Trial doses are written by researchers. Use the second set.

Evidence-based dose ranges by ingredient

Ingredient	Trial-supported dose	Notes
Silymarin (80% standardized)	420 to 2,100 mg/day, split	Higher end used in the 2017 NASH trial
Silybin phytosome	188 to 376 mg/day	Better absorbed, lower dose needed
NAC	600 to 1,800 mg/day, split	Away from meals
TUDCA	250 to 1,000 mg/day	Cholestatic contexts mainly
Vitamin E (RRR-alpha-tocopherol)	800 IU/day	Non-diabetic NASH, time-limited
Omega-3 (EPA+DHA)	2,000 to 4,000 mg/day	With a fat-containing meal
Berberine HCl	500 mg, 2 to 3x daily	Before meals
Choline (bitartrate or CDP-choline)	425 to 550 mg/day total intake	Diet counts toward this

With food or without?

Fat-soluble compounds need fat. Vitamin E, curcumin, silybin phytosome, and omega-3 all absorb substantially better with a meal containing 10 to 15 grams of fat. Taking your fish oil with black coffee at 6am is close to throwing it away.

NAC is the exception I flag most often. It absorbs better on an empty stomach, and splitting the dose (morning and evening) suits its short half-life. Berberine goes 20 to 30 minutes before meals, since a chunk of its benefit involves postprandial glucose.

Cycling: necessary or nonsense?

Mostly nonsense. Nobody cycles their choline intake. Omega-3s work by getting incorporated into membranes, which takes months, so cycling actively defeats the mechanism.

The exception is high-dose vitamin E. I’d run it in 6 to 12 month blocks with lab rechecks, not indefinitely, given the mortality signal I’ll cover next. Same instinct for high-dose green tea extract, though my honest recommendation there is to skip it entirely.

What not to stack together

Combinations I like: omega-3 plus choline (both support VLDL export and membrane phospholipids), silymarin plus vitamin E (different antioxidant mechanisms, and they were combined in several European trials), berberine plus a fiber source (blunts the GI complaints).

Combinations I don’t: three separate products each containing milk thistle, NAC, and artichoke, stacked on top of each other. You’re now taking 1,800 mg of NAC you didn’t plan for and paying triple. Read every label in your cabinet and total up the overlap. Most people are shocked.

Set a retest date at 8 to 12 weeks. ALT, AST, GGT, ALP, a lipid panel, HbA1c, and a FIB-4 calculation. That’s your decision point. Not how you feel.

Safety: When Liver Supplements Damage Your Liver

The irony writes itself, but the numbers aren’t funny.

Supplement-induced liver injury by the numbers

Herbal and dietary supplements now account for roughly 20% of drug-induced liver injury cases in the US DILI Network registry, up from about 7% in the early 2000s. Navarro and colleagues have documented this shift repeatedly. Supplement-related DILI also tends to be more severe than prescription-drug DILI, with higher rates of transplant listing and death, partly because people don’t suspect the bottle and keep taking it.

The worst offenders

Green tea extract / EGCG. Dose-dependent, worse when taken fasted, and worse in people carrying HLA-B*35:01. Doses above roughly 800 mg EGCG daily carry documented risk. Drinking green tea is fine. Concentrated capsules are the problem.

Ashwagandha. Björnsson’s 2020 case series in Liver International described a cholestatic injury pattern with a latency of two to twelve weeks. More cases have accumulated since. This one surprised a lot of people, myself included.

Garcinia cambogia. Multiple transplant cases. There is no benefit here worth this risk.

Black cohosh and kava. Both well-documented, both with case fatalities.

Turmeric with piperine. Halegoua-DeMarzio et al. published ten DILIN cases in The American Journal of Medicine (2023), with HLA-B*35:01 again over-represented. Piperine boosts curcumin absorption dramatically, which appears to be part of the problem. Culinary turmeric is not implicated.

Notice the pattern? Every single one of these is sold as a health product.

Drug interactions worth knowing

• Vitamin E above 400 IU/day: Miller’s meta-analysis in Annals of Internal Medicine (2005) pooled 19 trials and found a dose-dependent increase in all-cause mortality above that threshold. It’s a contested finding, but combined with the bleeding risk on warfarin or DOACs, it’s why I time-limit it.
• Milk thistle: in vitro inhibition of CYP2C9 and UGT enzymes. Clinically modest, but if you’re on warfarin or phenytoin, tell your prescriber.
• NAC and nitroglycerin: potentiates vasodilation, meaning hypotension and severe headache.
• Berberine: inhibits CYP3A4 and P-glycoprotein, which is a long list of affected drugs (statins, calcium channel blockers, cyclosporine). Combined with metformin or insulin it can push glucose too low.
• Omega-3 at 3 g+: additive with antiplatelets and anticoagulants.

Who should avoid these entirely

Pregnancy. Active viral hepatitis without hepatology supervision. Transplant recipients on tacrolimus or cyclosporine (CYP3A4 interactions here can be catastrophic). Anyone on four or more CYP-metabolized medications. Anyone with decompensated cirrhosis.

And know the warning signs. Fatigue plus nausea. Dark urine. Pale, clay-colored stools. Right upper quadrant pain. Yellowing of the eyes or skin. Itching without a rash. If any of those appear, stop everything non-essential and get an LFT panel within 48 hours. Don’t wait to see if it passes.

How Long Before You See Anything?

Realistic timelines by marker

ALT and AST: 8 to 12 weeks. That’s the window used in essentially every trial that found a signal. Enzyme half-lives and hepatocyte turnover set the floor here, and no supplement overrides biology. Anyone advertising a two-week transformation is selling something.

Liver fat on imaging: 3 to 6 months. MRI-PDFF changes are detectable earlier under aggressive intervention (the fructose restriction studies saw movement in nine days), but for supplement-level effects, give it a full season.

Fibrosis: 12 to 24 months. Where it’s reversible at all, collagen remodeling is slow. This is a two-year project, not a supplement cycle.

Symptoms: unreliable at any timepoint. Here’s my least popular opinion in this article. The “amazing energy” people report in week one is placebo, or a coincidental diet change, or the caffeine hiding in the formula. The liver has no pain fibers in its parenchyma and doesn’t produce the sensations people attribute to it. Trust the labs. They don’t have a marketing budget.

The Free Stuff That Outperforms Any Bottle

Alcohol, fructose, and visceral fat

Alcohol is dose-responsive with no established safe threshold for liver outcomes, and the interaction with metabolic risk is multiplicative, not additive. That’s why the 2023 nomenclature update created a whole category, MetALD, for people with both fatty liver and moderate alcohol intake. If you have a fatty liver and drink four beers on Saturday, you’re not adding two risks together. You’re multiplying them.

Fructose deserves its own paragraph. It bypasses the phosphofructokinase checkpoint and feeds de novo lipogenesis almost directly. Schwarz and colleagues showed in Gastroenterology (2017) that nine days of isocaloric fructose restriction in children cut liver fat substantially, with calories held constant. Nine days. Cutting soda outperforms every capsule reviewed here.

Resistance training and the liver

This is the finding I wish more people knew. Hallsworth’s team reported in Gut (2011) that eight weeks of resistance training reduced hepatic lipid by around 13% in NAFLD patients with no weight loss at all.

No weight loss. That matters enormously for anyone who has tried and failed to lose weight repeatedly, which is most people. You get liver benefit from the training itself, through improved skeletal muscle glucose disposal. Two sessions a week, compound lifts, progressive load.

Sleep and circadian timing

Short sleep (under six hours) and late-night eating both worsen insulin resistance and correlate with higher hepatic fat in cohort data. Your liver runs a strong circadian program, and eating at midnight asks it to do daytime work on a night shift.

Honestly? Fixing a chronic 5.5-hour sleep habit will do more for your ALT than every capsule in this article combined. I know that’s not what a supplement guide is supposed to say.

Frequently Asked Questions About Liver Supplements

What do liver supplements actually do? Most liver supplements supply antioxidants, glutathione precursors, or bile acids that support existing liver processes. They don’t remove toxins (your liver does that on its own) and they don’t repair scarring. The evidence-backed effects are modest: small reductions in ALT and AST, minor improvements in liver fat, and in some cases better insulin sensitivity.

How do liver supplements work in the body? Through four main mechanisms. Silymarin stabilizes hepatocyte membranes and reduces oxidative stress. NAC supplies cysteine to rebuild glutathione, the liver’s primary detoxification antioxidant. TUDCA shifts the bile acid pool toward less toxic species and reduces cellular stress. Berberine and omega-3s work upstream on insulin resistance and fat handling, which is where fatty liver actually starts.

Are liver supplements safe, and can they damage your liver? Some can, absolutely. Herbal and dietary supplements now cause about 20% of drug-induced liver injury cases in the US registry. The documented offenders include high-dose green tea extract, ashwagandha, garcinia cambogia, black cohosh, kava, and turmeric combined with piperine. Silymarin, NAC, and omega-3s have strong safety records at standard doses.

What is the best dosage for liver supplements? Use trial doses, not label doses. Silymarin: 420 to 700 mg/day standardized to 80%. NAC: 600 to 1,800 mg/day split. Omega-3: 2,000 to 4,000 mg combined EPA+DHA. Berberine: 500 mg two to three times daily before meals. Vitamin E: 800 IU/day for non-diabetics with confirmed NASH, limited to 6 to 12 months.

How long do liver supplements take to work? Expect 8 to 12 weeks for measurable changes in ALT and AST, 3 to 6 months for liver fat changes on imaging, and 12 to 24 months for any fibrosis improvement. Products promising results in days or two weeks are making claims no trial has ever supported.

Do liver supplements lower elevated ALT and AST levels? Silymarin and NAC show modest ALT reductions in meta-analyses, typically in the range of a few units per liter, which is statistically real but clinically small. Weight loss of 7 to 10% and cutting alcohol reduce ALT far more. If your ALT is elevated, find the cause first rather than reaching for a bottle.

Can liver supplements reverse fatty liver disease? Not on their own. Weight loss of 10% or more produces NASH resolution in roughly 90% of patients, which is a scale of effect no supplement approaches. Omega-3s, vitamin E, and berberine can add a small incremental benefit on top of diet and exercise. Used as a substitute for those changes, they do essentially nothing.

Is milk thistle or TUDCA better for liver health? It depends on the injury pattern. Milk thistle (silymarin) has more human data in fatty liver and general hepatoprotection, and it costs far less. TUDCA fits cholestatic problems better: elevated ALP, high bilirubin, oral steroid use, itching. For a typical person with fatty liver and a slightly high ALT, standardized silymarin is the more sensible starting point.

Do you need a liver detox or cleanse? No. Your liver detoxifies continuously through phase I and phase II enzyme pathways, and no juice, tea, or capsule accelerates the process. “Detox” products are marketing language, and the category has been repeatedly flagged for adulteration with unapproved pharmaceutical ingredients. Reducing alcohol, sugar, and excess body fat is the only real cleanse available.

Can you take liver supplements with medications like statins or metformin? Sometimes, but check the interactions. Berberine inhibits CYP3A4 and can raise statin levels while amplifying metformin’s glucose-lowering effect. Milk thistle mildly affects CYP2C9, relevant for warfarin. Omega-3 above 3 g adds to antiplatelet effects. Transplant recipients on immunosuppressants should avoid herbal liver supplements entirely.

My Verdict: What I'd Actually Take

The short list

Normal labs, no risk factors, just curious. Nothing. Drink your coffee, lift twice a week, keep alcohol modest. Buying liver supplements to protect a healthy liver is like buying a fire extinguisher for a house that isn’t on fire and then never checking the smoke alarms.

Elevated ALT with metabolic syndrome. This is the group with a real case. I’d run 2 to 3 g/day EPA+DHA with a fat-containing meal, 420 mg standardized silymarin, and berberine at 500 mg twice daily if HbA1c or triglycerides are poor. Cut sugar-sweetened drinks to zero. Start resistance training. Retest at twelve weeks.

Heavy alcohol history. Silymarin has the longest track record in alcohol-related liver disease, and NAC at 1,200 mg/day supports glutathione that alcohol depletes. But I’ll say the obvious part: the supplements are worth maybe 5% of the outcome. Stopping drinking is the other 95%, and you need a hepatologist, not a stack.

The bottom line

One ingredient only? Standardized silymarin at 420 mg/day. Not because it’s dramatic (it isn’t) but because it has the deepest human evidence base, the cleanest safety profile, and it costs about $12 a month. Omega-3 is a close second if your triglycerides are high.

Here’s the position I keep coming back to after reading through this literature. Supplements are a 10 to 15% lever sitting on top of an 85% foundation of body composition, alcohol, sugar, movement, and sleep. Any product implying otherwise is lying to you, and the ones with “detox” on the label are usually lying about more than that.

So don’t buy anything yet. Get an LFT panel, a lipid panel, HbA1c, and calculate your FIB-4. Change one behavior for twelve weeks. Then retest, and decide what a bottle actually needs to do.

Frequently Asked Questions