Berberine: Blood Sugar, Cholesterol, Traditional Use and Medicine Interactions

What is berberine?

Berberine is a bitter yellow alkaloid found in several unrelated medicinal plants, including barberry (Berberis vulgaris), Indian barberry or tree turmeric (Berberis aristata), Chinese goldthread (Coptis chinensis), Oregon grape (Mahonia aquifolium) and goldenseal (Hydrastis canadensis). A capsule of isolated berberine HCl is not the same preparation as a root or bark tea. It is not a nutrient: the body does not need it, and food does not supply a meaningful amount. A botanical review^[1] documents these sources and their traditional use.

People take it for blood sugar, cholesterol, and weight. What is established: in people with type 2 diabetes it lowers blood sugar modestly, and in placebo-controlled trials it lowers LDL cholesterol, often called “bad” cholesterol, and triglycerides, another type of blood fat, a little. It also changes how the body clears several prescription drugs, which makes it more like a medicine than a vitamin.

Trials used 500 mg two or three times a day with meals, 1,000 to 1,500 mg a day in total. Most capsules contain 500 mg of berberine hydrochloride, shortened to HCl; the label’s 500 mg is compared directly with the 500 mg used per dose in these trials. There is no official daily amount.

Where berberine's traditional uses came from

Plants containing berberine appear in Chinese and Ayurvedic medical traditions, especially for digestive complaints. The named plant, part used and preparation matter: Chinese goldthread rhizome, barberry bark and today's isolated berberine tablet are not interchangeable doses. European and North American traditions also used barberry or goldenseal preparations. LiverTox^[2] describes the long digestive-use history; the older practice explains why the compound was studied but is not itself a trial of a modern supplement.

The research emphasis has shifted. The older human diarrhea trials tested particular infections; recent studies mostly test blood glucose, blood lipids, PCOS or fatty-liver markers. The article compares each of those results on its own terms.

What people take berberine for

Nobody needs it. It is sometimes called a natural metformin, but it is not a swap for the drug. Metformin has long-term trials showing fewer diabetes complications; berberine has trials of blood sugar numbers only, most of them small and short.

If you have high blood sugar or type 2 diabetes, trials found a modest fall in blood sugar, and it should be tried only with the clinician who manages your diabetes. If you want lower cholesterol, the drop in LDL was small, about 18 mg/dL (milligrams per deciliter, the unit on US lab reports), and no trial has tested whether berberine prevents heart attacks.

PCOS and fatty liver disease are also studied, but those studies have mostly measured laboratory values and imaging over short periods. They do not give berberine a proven role in preventing the outcomes those conditions can cause.

When to ask a doctor before taking berberine

• You take diabetes medicine. Berberine may lower blood sugar further. The effect depends on the medicine and your treatment, so your clinician should review the combination and glucose monitoring.
• You take ciclosporin, or several prescription drugs. In a six-person transplant study, overall ciclosporin exposure—measured from blood levels over time—rose by about a third after berberine. Other human studies found changes in drug handling, explained below.
• You are pregnant or breastfeeding, or giving it to a baby. The National Institutes of Health says berberine can be harmful to newborns and that goldenseal, which contains it, should not be used in pregnancy or breastfeeding.

Amounts used in trials

There is no recognized recommended daily amount or official upper limit for berberine.

The outcome table above separates the diabetes and cholesterol regimens. In the small 13-week diabetes trial, HbA1c fell from 9.5 to 7.5 percent; the 37-trial review found a much smaller average reduction, about 0.6 percentage points. A within-group fall is not the same as the treatment effect compared with a control group.

The trials divided the daily total into separate 500 mg doses taken with meals. Stomach upset is the most common side effect; the studies tested that schedule, not one large daily dose.

All forms at a glance

Compare products by milligrams of berberine, usually berberine HCl. Skip “natural metformin” claims and blends that hide the active amount. If you take ciclosporin or another prescription drug, have a doctor or pharmacist check the combination before using berberine.

What to pay attention to on the label

The number to find is milligrams of berberine per capsule, and the thing to check is whether that number is berberine or a blend.

Which number on the label is the amount?

Look for berberine or berberine HCl in milligrams on the Supplement Facts panel. One product lists a 1,000 mg “berberine HCl complex” per two capsules. The complex mixes berberine HCl with a barberry root extract, so the label does not say how much is berberine. The same capsules add a 300 mg blend of cinnamon, bitter melon, and gymnema. If the label does not give berberine in milligrams, you cannot compare it with the trials.

Is dihydroberberine better?

Not shown. No blood-sugar trial of dihydroberberine has matched the participant numbers and study duration of the berberine trials, and no one has shown how much dihydroberberine matches 500 mg of berberine.

Misleading labels

“Natural metformin”

Berberine is not metformin. In one small trial the two lowered HbA1c by similar amounts, but a 2012 review of 14 trials found berberine alone was not better than diabetes drugs and that the trials were of low quality.

Once-a-day capsules

The trials gave berberine two or three times a day. A single daily capsule, even a “time-release” one, is not the pattern that was tested.

Confusing servings and items per serving

Some manufacturers put 180 capsules on the front, and the back says 3 capsules per serving and 60 servings. The amount on the Supplement Facts panel is for the whole 3-capsule serving.

Common Myths about Berberine

Does berberine reduce belly fat?

Not by much, and the reviews disagree. A 2020 review of 10 studies found no change in body weight, though waistlines were about 2.75 cm (about 1 inch) smaller. A 2026 review of 23 trials found weight about 0.9 kg (2 lb) lower, with no change in waist-to-hip ratio. The 2026 review flagged incomplete reporting of berberine purity and dose, and a lack of blinding in some studies—meaning participants or researchers knew which treatment was given.

A “metabolism” or “belly fat” label claim may draw on those pooled markers. In a six-month randomized trial^[3] of people with obesity and fatty liver disease but without diabetes, 1 g/day did not reduce measured visceral fat or liver fat versus placebo. That directly limits the claim that berberine reliably targets abdominal fat.

Is berberine harmless with other medicines because it comes from a plant?

No. In Guo’s study of 18 men^[4], 300 mg three times a day for two weeks raised overall exposure to midazolam, a sedative, by about 40 percent, as measured from blood levels over time. The researchers also measured urine ratios of the cough medicine dextromethorphan and the blood-pressure medicine losartan to their breakdown products. Those ratios changed, showing altered drug metabolism.

In a separate six-person transplant study^[5], overall ciclosporin exposure, measured from blood levels over time, rose by 34.5 percent after berberine. These are measured interactions with named drugs. They justify checking other medicines that use the same pathways, but they do not predict the size or clinical effect of every combination.

Drawbacks and side effects

Understanding these risks

There is no official upper limit. Stomach and gut upset is the most common side effect: in one diabetes trial, about a third of people had short-lived stomach effects, and in placebo-controlled cholesterol trials, gut side effects were reported by 2 to 23 percent of people on berberine and 2 to 15 percent on placebo. No serious side effects were reported in those trials.

Check with a doctor or pharmacist before taking berberine with ciclosporin, diabetes medicines, or drugs broken down by the same liver enzymes as midazolam, dextromethorphan, or losartan. A pharmacist can tell you whether yours are among them.

As for taking it every day or long term: trials lasted 4 to 24 weeks. There are no long-term safety trials and no trials of heart attacks or diabetes complications.

The symptoms behind “gut upset” include nausea, abdominal pain, bloating, constipation and diarrhea^[6]. Constipation and diarrhea are both reported; they do not have to occur together.

The science behind berberine

The evidence shows a modest drop in blood sugar and cholesterol in short trials, mostly from China, and clear effects on how the body clears some prescription drugs.

The older digestive use and what diarrhea trials found

Berberine's medicinal background includes digestive complaints and diarrhea in Ayurvedic and Chinese traditions, as LiverTox describes^[2].

In Rabbani's 1987 trials^[7], berberine sulfate reduced stool output in adults with diarrhea caused by a particular toxin-producing strain of E. coli. Its effect in cholera was limited, and adding it to tetracycline did not improve stool output over tetracycline alone. A 1985 trial^[8] also found no benefit over placebo in its non-cholera group.

The results give some support to a specific digestive use, with conflicting findings across infections. Severe or persistent diarrhea needs assessment and fluid replacement.

Blood sugar

In a 2008 trial, 36 people newly diagnosed with type 2 diabetes took 500 mg of berberine or metformin three times a day for 13 weeks. On berberine, HbA1c fell from 9.5 to 7.5 percent, and the authors called the effect similar to metformin. A 2022 review of 37 trials with 3,048 people found:

• An average drop of about 0.6 points in HbA1c.
• An average drop of about 15 mg/dL in fasting glucose.
• The drop was larger in people whose blood sugar started higher.
• There was no rise in side effects or low blood sugar.

In a 2021 placebo-controlled trial of 297 people with newly found high blood sugar, fasting glucose fell about 9 mg/dL more than on placebo over 16 weeks.

“Supports healthy blood sugar” on a bottle often refers to these trials in people with raised glucose or type 2 diabetes. It does not tell you what the same dose would do for someone whose glucose is already normal.

Cholesterol

A 2023 review of 18 placebo-controlled trials with 1,788 adults found:

• LDL cholesterol was about 18 mg/dL lower.
• Triglycerides were about 30 mg/dL lower.
• HDL cholesterol, often called “good” cholesterol, rose slightly in women but not in men.
• Most trials were done in mainland China or Hong Kong.

This is the evidence a “supports healthy cholesterol” label may be referring to. In laboratory and animal studies, berberine has increased liver LDL-receptor expression and affected PCSK9 and lipid-synthesis pathways. Those mechanisms could help explain the blood measurements, but the human trials have not tested whether they prevent heart attacks. One mechanistic study^[9] used liver cells and hamsters, so its cellular detail should not be mistaken for a demonstrated human pathway.

Polycystic ovary syndrome

PCOS trials have measured insulin resistance, blood lipids, androgen levels, ovulation and pregnancy outcomes. A 2020 review of randomized trials^[10] reported improvements in several metabolic markers, but live-birth rates were similar to placebo or metformin in the available comparisons and lower than with letrozole. Study designs and co-treatments varied. Better insulin markers cannot be read as proof that berberine restores fertility, and pregnancy is a reason to discuss treatment with a clinician.

Fatty liver disease

Small trials have tested standard berberine for liver and metabolic measures. A separate 100-person phase 2 trial^[11] found less liver fat with berberine ursodeoxycholate, a compound containing berberine and ursodeoxycholic acid, than with placebo. That result belongs to the tested compound, not an ordinary berberine HCl capsule. More recently, a six-month randomized trial^[3] in adults with obesity and fatty liver disease but without diabetes found that 1 g/day of ordinary berberine did not reduce visceral fat or liver fat compared with placebo. These findings make the formulation and population central to the answer.

What berberine does in the body

The clinical blood-sugar and blood-fat changes are measurable; the chain of events producing them in people is less certain. In cells and animals, berberine affects AMPK, an enzyme that responds to a cell's energy state, as well as liver LDL-receptor expression and other pathways. Calling it a “metabolic master switch” makes the result sound more direct and complete than the human evidence allows. Human glucose studies have not shown that AMPK activation alone explains the change in HbA1c.

Berberine is poorly absorbed into the bloodstream, leaving much of an oral dose in the gut. The 297-person hyperglycemia trial^[12] measured changes in the gut microbiota alongside glucose outcomes. It did not prove that a particular bacterium caused the glucose change, and it does not make a berberine capsule an established microbiome treatment. Reports that berberine directly kills harmful gut microbes largely come from laboratory work.

What is still open

No trial has tested whether berberine prevents diabetes complications, heart attacks, or deaths, or how safe it is beyond six months.

Because berberine is often compared with metformin, people who train sometimes ask whether it could blunt exercise gains. Two small metformin trials in older adults found smaller improvements from exercise training, but no trial has tested berberine this way. Keep prescribed metformin in place; these findings concern metformin rather than berberine.

What you may have heard on podcasts

Andrew Huberman (Huberman Lab, 31 October 2023) said he does not take berberine, which he called “poor man’s metformin,” because it gave him severe headaches that he put down to lower blood sugar.

A 2022 review of 37 trials did not find more low blood sugar on berberine, but those trials were in people with diabetes, not in people without it.

Labs

No test measures berberine. HbA1c and fasting glucose track the blood-sugar reason people take it. Liver and kidney tests are not an established routine monitoring program for berberine; a clinician may order them based on your health, symptoms, and other medicines.

HbA1c usually helps track average blood sugar, but iron-deficiency anemia, changes in red-cell lifespan and some inherited hemoglobin variants can distort it. If it disagrees with glucose measurements, a clinician should investigate the mismatch. ^[13]

Blood glucose testing

If blood sugar is the reason you are considering it, the numbers that matter are HbA1c and fasting glucose.

HbA1c shows average blood sugar over about three months; fasting glucose shows one point in time. Neither measures berberine. The Labs section lists the available panels.

Before the test if you already supplement

There is no established berberine-stop interval for the HbA1c and glucose tests that we know of. Ask the clinician whether the aim is to follow your current regimen or assess a change; briefly stopping cannot recreate an untreated three-month HbA1c. If a fasting glucose test is ordered, follow the specific fasting instructions and ask how to handle your morning berberine dose and any diabetes medicines rather than changing them yourself.

The bottom line

Berberine can lower blood sugar and cholesterol measurements, but its weight-loss effect is small or inconsistent. Calling it a natural version of metformin makes the comparison sound much more settled than it is.

Whether those changes would be useful depends on your starting results, treatment and tolerance. Berberine can also change how some medicines are handled by the body, so adding it to an existing regimen needs a medicine-specific check.

Help us improve this article

Spotted an error or found a source we should consider? Tell us which part of this article needs another look and include a link if you have one. Please leave out personal health details and lab results; we can’t answer individual medical questions.

Your name

Email address

What would you like to share?Select a topicCorrectionSource or research suggestionArticle requestAccessibility issueBusiness inquirySomething else

Your message

Continue to email →Email us directly