Berberine and Fatty Liver: What MASLD Trials Show

Berberine fatty liver research has produced mixed results. Earlier trials reported improvements in liver fat or selected metabolic markers, but a six-month 2026 trial in 337 diabetes-free adults with obesity and MASLD found no significant benefit for its primary liver-fat or visceral-fat outcomes. That result belongs near the beginning of any current evidence review. ^{[1] [2]}

It does not erase every earlier finding. It changes the question from “Does berberine cleanse the liver?” to “Which patients, preparations and endpoints have actually shown benefit?” Liver enzymes, liver fat and fibrosis are different outcomes.

NAFLD, MAFLD and MASLD: related names, different enrollment rules

Older studies commonly used non-alcoholic fatty liver disease, or NAFLD. Later research may use metabolic dysfunction-associated fatty liver disease, MAFLD, or metabolic dysfunction-associated steatotic liver disease, MASLD. These labels reflect changing diagnostic frameworks and should not be assumed to produce identical trial populations.

When comparing studies, look at the actual eligibility criteria: diabetes status, glucose impairment, obesity, alcohol exposure, medicines and how fat or fibrosis was measured. A trial in people with abnormal glucose regulation may not predict the same response in people without diabetes.

The 2015 trial: encouraging, with important qualifications

The 2015 multicenter study enrolled 184 adults and compared lifestyle intervention alone with lifestyle plus pioglitazone or lifestyle plus berberine. The berberine regimen was 500 mg three times daily for sixteen weeks. ^[1]

The reported relative reduction in hepatic fat was approximately 52.7% with berberine plus lifestyle and 36.4% with lifestyle alone. Those figures are relative changes, not percentage-point changes in liver fat and not the proportion of people cured.

The trial was open-label and included active lifestyle care. The additional improvement is interesting, but the design is less protected against bias than a well-blinded placebo comparison. A later lipid-profiling publication arose from the same research program and should not be counted as a wholly independent clinical trial. ^[3]

The 2024 placebo-controlled trial

A smaller randomized double-blind trial enrolled 70 people with MAFLD and used 1,500 mg/day for twelve weeks. Selected enzyme and cholesterol measures favored berberine, while many other outcomes did not differ significantly. ^[4]

This illustrates why “improved fatty liver” can conceal several different results. A lower ALT is potentially useful, but it is not direct proof of a substantial liver-fat reduction, reduced fibrosis or fewer future complications. Changes within both groups also need to be distinguished from differences between groups.

BRAVO: the newer trial that changes the conversation

BRAVO randomized 337 adults at eleven centers in China. Participants had obesity and MASLD but did not have diabetes. They received berberine HCl 500 mg twice daily or placebo for six months, alongside lifestyle advice. CT assessed visceral adipose tissue and liver fat. ^[2]

Neither coprimary outcome significantly favored berberine. The between-group estimate for absolute liver-fat change was 0.9 percentage points, with a 97.5% confidence interval from -0.4 to 2.1; this did not establish a benefit. The visceral-fat result was likewise not significant.

The study reported modest improvements in secondary LDL, apoB and inflammation measures. These should be discussed as secondary signals, especially because multiple analyses increase the chance of apparently positive findings. They should not be repackaged as success for the primary liver-fat claim.

Why might the studies disagree?

Population differences are one possibility: glucose dysregulation, diabetes and other metabolic abnormalities may influence response. Dose, duration, background care, preparation and measurement methods also differ. These are plausible explanations to investigate, not established excuses for dismissing a null trial.

The older and newer studies should not be arranged as a simple dose ladder proving that 1.5 g works while 1 g fails. Without an appropriate direct comparison in the same population, that conclusion is not justified.

Systematic reviews can also become dated when new trials appear. A favorable meta-analysis published before BRAVO cannot be treated as if it had already incorporated its result. ^{[5] [6]}

Liver enzymes, fat and fibrosis are separate questions

ALT and AST are markers of hepatocellular injury, not direct measurements of the amount of fat or the stage of scarring. They can improve without demonstrating fibrosis reversal. Conversely, clinically important liver disease can exist without strikingly elevated enzymes.

An ultrasound report of steatosis, a calculated fibrosis score, elastography and a biopsy each contribute different information. A supplement trial measuring one endpoint should not be credited with improvements in all of them.

For an individual, the clinically relevant next step may be fibrosis risk assessment and management of diabetes, lipids, weight or alcohol exposure. A lower laboratory number should not delay appropriate evaluation.

What about berberine ursodeoxycholate and HTD1801?

HTD1801 is berberine ursodeoxycholate, an investigational compound studied in metabolic and liver settings. It is not simply another brand of a standard HCl supplement. Trials of this compound cannot be used to claim that an ordinary retail capsule has the same clinical effect. ^{[7] [8] [9]}

The distinction is especially important when an article quotes a favorable phase-2 result without naming the intervention. Formulation identity is part of the evidence, not a technical detail to remove from the headline.

Does milk thistle make it better for fatty liver?

Berberine–silymarin combinations have been studied for metabolic markers, and a meta-analysis reported favorable glucose and lipid changes. That does not establish that adding any milk-thistle capsule to any berberine product reverses liver fibrosis or outperforms the exact single ingredient in a direct trial. ^[10]

Our milk-thistle guide addresses those combinations. Here, the central issue remains whether the trial actually measured a liver outcome and whether its preparation matches the proposed use.

Safety and treatment decisions

People with liver disease often take other medicines. Berberine can affect drug-metabolizing enzymes, and safety findings from selected trial participants may not apply to cirrhosis or complex treatment regimens. ^[11]

Do not assume that a product marketed for liver health cannot cause an adverse reaction or interaction. Seek medical assessment for jaundice, dark urine, severe pain or other significant new symptoms. The side-effects guide covers safety separately.

The dosage guide describes study exposures without turning them into a fatty-liver prescription. More capsules are not an established solution to a null response.

Relative liver-fat change vs percentage-point change

Consider an illustrative liver-fat measurement falling from 20% to 10%. That is a ten-percentage-point absolute reduction and a 50% relative reduction. It is not a 50-percentage-point change. The distinction matters when older study summaries quote large relative effects.

The 2015 study's reported 52.7% relative reduction should therefore not be rewritten as losing 52.7 percentage points of liver fat or curing 52.7% of participants. Its comparator group also improved with lifestyle intervention. ^[1]

BRAVO used a different statistical framework and population. Comparing the largest number in each abstract without aligning the endpoint can manufacture an apparent contradiction or exaggerate a treatment effect.

Preparation characterization can change the interpretation

One randomized fatty-liver report described a boiled preparation by dried-material weight, with unclear standardized active-berberine content. A figure of 6.25 g/day in that context should not be copied as 6,250 mg of isolated berberine or used to recommend an unusually large capsule intake. ^[12]

This is a source-quality issue as well as a formulation issue. If the active amount is uncertain, the study is harder to compare with a 500 mg HCl capsule. Being explicit about that uncertainty is more useful than assuming every botanical preparation is chemically equivalent.

A companion report is not another independent trial

The lipid-profiling report linked to the 2015 trial adds mechanistic and metabolic detail. It does not create a new independent population demonstrating liver-fat efficacy. ^[3]

The same caution applies when a vascular-function paper follows a fatty-liver intervention. Different endpoints can be worth reporting without multiplying the number of independent trials. The Data Center links related reports so readers can trace the underlying cohort.

When a liver claim should change clinical decisions

The key issue is not merely whether an enzyme moved. A clinician may need to establish fibrosis risk, identify another cause of liver disease or manage diabetes and cardiovascular risk. A supplement result should be considered within that larger assessment.

Do not interpret a normal or improved ALT as permission to ignore recommended follow-up. Likewise, a fatty-liver ultrasound cannot on its own supply the same information as a fibrosis assessment. A precise outcome prevents a favorable biomarker from carrying more clinical meaning than it actually has.

Where a supplement fits

Berberine's liver research remains worth following, particularly as better controlled trials clarify who may benefit. At present, it should not be presented as a reliable liver-fat remover, fibrosis-reversal treatment or replacement for clinical management.

If product selection becomes relevant after clinical review, our DoNotAge Berberine review separates current manufacturer facts from ingredient research. Biohack Blueprint may earn commission from qualifying purchases. The recommendation does not transfer HTD1801 or combination-specific findings to that bottle.

Explore the individual studies and linked companion reports in the Berberine Data Center. A complete account includes the encouraging older findings and the newer result that challenged the broadest claims.