Butterfly Pea: Antioxidant and Metabolic Claims Examined

"Packed with antioxidants" is on the front of nearly every bag of blue tea sold. It is the single most common health claim made for butterfly pea, and it deserves a page of its own — not because it is a lie, but because it is a true statement about a test tube being used as if it were a true statement about a person.

Two separate corrections are needed, and they compound. The first is about the measurement: an antioxidant score is a chemistry reading, the USDA formally withdrew its own ORAC food database in 2012 because those readings were being misused in exactly this way, and the large randomized antioxidant-supplement trials did not deliver the benefit the hypothesis predicted. The second is about where the molecule goes: anthocyanins are among the least bioavailable dietary polyphenols. Only a small percentage is absorbed intact; the majority reaches the colon and is dismantled by gut bacteria into simpler phenolic acids. The compound in the cup is largely not the compound in your blood.

Then there is the metabolic claim, which is a different and more interesting case — because here there is real human data. It is small, it is acute, and it is honest work that should be reported at its actual tier rather than inflated or dismissed.

The Claim and Where It Comes From
What ORAC, DPPH, FRAP and TEAC Actually Measure
The 2012 ORAC Withdrawal
Bioavailability: The Cup Is Not the Bloodstream
The Antioxidant Trials That Did Not Deliver
What Anthocyanins May Actually Do: Signalling, Not Scavenging
Blocking Starch Digestion: The In-Vitro Case
The Human Post-Meal Glucose Trial
The High-Fat Meal Trial
If You Take Diabetes Medication
Dose Reality: What Is Actually in a Cup
Evidence Tier Summary
Key Research Papers
Connections

The Claim and Where It Comes From

The chain of reasoning behind "antioxidant-rich blue tea" runs like this, and every link in it is worth inspecting separately:

Butterfly-pea flowers contain anthocyanins and other flavonoids. True, and well characterised.
Those compounds score highly in laboratory antioxidant assays. True, and reproducible.
Therefore the tea is "high in antioxidants". Partly true, and dose-dependent in a way the phrase conceals.
Therefore drinking it reduces oxidative damage in your body. Not established.
Therefore it prevents the diseases attributed to oxidative damage. Not established, and the class of claim that failed when tested directly.

Steps one and two are ordinary phytochemistry. Step three quietly changes the subject from concentration in a plant to dose in a serving. Steps four and five are the leap, and they are the leap that a generation of supplement marketing made on behalf of vitamins C and E, beta-carotene, selenium and every "superfruit" of the 2000s.

Butterfly pea inherited that entire rhetorical structure ready-made. Its deep blue colour makes the claim feel self-evidently right, which is precisely why it needs checking.

What ORAC, DPPH, FRAP and TEAC Actually Measure

These acronyms appear on packaging and in supplement literature as though they were clinical measurements. They are not. Each is a specific chemical reaction run in a cuvette, and knowing what the reaction is tells you exactly how far the number can travel.

ORAC — Oxygen Radical Absorbance Capacity. A fluorescent probe is degraded by a chemically generated peroxyl radical; the sample's ability to delay that degradation is measured against a Trolox standard. The radical species is not one your cells generate in that form, and the reaction happens in buffer at a pH and concentration nothing like a cell.
DPPH — a stable violet synthetic radical that turns yellow when it accepts an electron or hydrogen atom. Convenient, cheap and extremely common in the phytochemistry literature. DPPH does not exist in biology; it is a laboratory reagent chosen because it is coloured and stable.
FRAP — Ferric Reducing Antioxidant Power. Measures the sample's ability to reduce iron from Fe(III) to Fe(II) at acidic pH. It is a reducing-power assay, not an antioxidant assay in any physiological sense, and it is affected by anything reducing in the sample — including, notably, uric acid, which dominates FRAP readings in human plasma.
TEAC — Trolox Equivalent Antioxidant Capacity. Expresses whatever a given assay measured in units of a water-soluble vitamin E analogue, so that different samples can be compared. It standardises the reporting, not the biological relevance.

Three consequences follow, and they apply to every "antioxidant score" you will ever read:

The assays disagree with each other. Rank a set of foods by ORAC and by FRAP and you get different orders, because the underlying chemistry differs. A single number described as "the antioxidant capacity" of something is a choice of assay presented as a fact of nature.

They measure the sample, not the eater. A high value tells you the extract can donate electrons in a cuvette. Whether any of it survives digestion, gets absorbed, reaches a tissue where oxidation matters, and arrives at a concentration high enough to compete with the body's own enzymatic defences — superoxide dismutase, catalase, glutathione peroxidase, all of which are catalytic and vastly more efficient than a stoichiometric small molecule — is a completely separate question that the assay cannot address.

A rise in "plasma antioxidant capacity" after a drink is not the outcome it sounds like. This is the most important subtlety, and it applies directly to the butterfly-pea trials. FRAP and similar assays applied to blood plasma are heavily influenced by urate, ascorbate and albumin thiols. A measurable rise after consuming a polyphenol-rich beverage has been shown for coffee, tea, cocoa and fruit juice, and can occur through mechanisms as banal as a fructose-induced change in urate. It demonstrates that something absorbable happened. It does not demonstrate that oxidative damage to your tissues went down, and it has repeatedly failed to predict clinical benefit.

The 2012 ORAC Withdrawal

This is the single most useful fact to carry into any conversation about antioxidant claims, and it is rarely mentioned by the people quoting the numbers.

The United States Department of Agriculture's Agricultural Research Service published an ORAC database of selected foods, and it became the reference table behind the entire "superfood" marketing era — the source of every "ten times more antioxidants than blueberries" comparison. In 2012 the USDA withdrew it.

The stated reasoning was direct: the values were being used by food and supplement companies to make health claims that the data could not support, and the assay's relevance to human health had not been demonstrated. The agency's position was that the biological meaning of an ORAC value in a food was unproven and that many of the polyphenols responsible for high readings have no antioxidant function in the body after they have been metabolised.

That last clause is the crux, and it is why this history sits in a butterfly-pea article. The compounds that generate the high score are often not the compounds present in your bloodstream after digestion. The next section is about exactly that.

The database's withdrawal did not stop the numbers circulating. They persist in marketing, in blog posts and in supplement literature, usually without attribution, more than a decade after the body that produced them disowned them. If you see butterfly pea assigned an antioxidant rank or score, that is where the genre comes from. USDA Agricultural Research Service.

Bioavailability: The Cup Is Not the Bloodstream

Anthocyanins have a distinctive and, for marketing purposes, inconvenient pharmacokinetic profile. Tier: this is established human research, and it is one of the better-studied corners of nutrition science.

The broad picture, from the polyphenol bioavailability literature reviewed by Manach and colleagues and refined substantially since:

Only a small percentage of ingested anthocyanin is absorbed as the intact parent molecule. Estimates from urinary recovery of the unchanged compound have historically come out at around or below one percent of the dose — among the lowest of any dietary polyphenol class. Peak plasma concentrations of intact anthocyanins after a normal food serving are typically in the nanomolar range, orders of magnitude below the micromolar and higher concentrations at which cell-culture experiments demonstrate effects.
Most of it reaches the colon. There, gut microbiota cleave the sugar off, break the pigment's central ring, and convert it into smaller phenolic acid metabolites — protocatechuic acid, vanillic acid, hippuric acid, ferulic acid and related compounds, depending on which anthocyanidin you started from.
Those metabolites are then conjugated by your own tissues — glucuronidated, sulfated and methylated in the gut wall and liver — producing a circulating profile made almost entirely of modified fragments.
Total recovery is far higher than the intact-parent figure suggests. This is the crucial refinement. The carbon-13 tracer study by Czank, Cassidy and colleagues followed a labelled anthocyanin through the whole body and found that when metabolites are counted, a substantial fraction of the dose is accounted for — far more than the classic "one percent" — and that these metabolites persist in circulation much longer than the parent compound. de Ferrars and colleagues mapped the same picture pharmacokinetically.

Why this matters for butterfly pea specifically. There are two distinct implications and they pull in different directions, which is why the honest answer is neither "it does nothing" nor "it is packed with antioxidants".

The first implication is deflationary. If the ternatin you drank is mostly not in your blood, then the ORAC or DPPH value of the ternatin is not the relevant number for anything happening in your body. Whatever butterfly pea does, it does not do it by putting ternatins into your tissues at antioxidant-relevant concentrations. That reasoning is simply unavailable.

The second implication is more interesting. If the biologically active species are the microbial phenolic acid metabolites, then the effect of an anthocyanin depends partly on your gut microbiome — which varies enormously between people — and the active compound may be shared across many different anthocyanin sources, since different pigments converge on overlapping metabolite sets. That is a live research area for anthocyanins as a class, and it is a reason the field has moved on from radical-scavenging explanations.

There is a further wrinkle unique to ternatins. The polyacylation that makes butterfly pea's pigment so stable in a cup also makes it a large, heavily decorated molecule — larger and more substituted than the simple monoglucosides on which most anthocyanin pharmacokinetics has been measured. There is no reason to assume ternatins are absorbed like cyanidin-3-glucoside, and to the best of the published record their human pharmacokinetics have not been characterised in their own right. That is an open question, not a hidden advantage.

The Antioxidant Trials That Did Not Deliver

The antioxidant hypothesis — that supplementing radical-scavenging compounds would reduce chronic disease — was not dismissed on theoretical grounds. It was tested, at scale, in large randomized controlled trials, and it largely failed. Tier: randomized clinical trials, large, long, with hard clinical endpoints. This is the highest tier of evidence discussed on this page, and its verdict is negative.

An honest caveat first, stated plainly because it matters: these trials tested isolated vitamins and minerals at high doses, not anthocyanins, and not butterfly pea. They do not prove that a flower tea does nothing. What they demonstrate is that the reasoning — high antioxidant capacity in vitro, therefore disease prevention in vivo — is not reliable, because when that reasoning was tested with the best tools medicine has, it did not hold.

Beta-carotene in smokers. The Alpha-Tocopherol Beta-Carotene study in Finland, reported in the New England Journal of Medicine in 1994, and the CARET trial reported in the same journal in 1996, both tested beta-carotene in populations at high lung-cancer risk. Neither found the expected protection, and both raised concern in the opposite direction. This is the historical moment the field's confidence broke.
Vitamin E and selenium for prostate cancer. The SELECT trial, reported in JAMA in 2011, was a large, long, well-conducted test of the hypothesis and did not show the anticipated benefit.
Antioxidant supplements and mortality overall. The systematic review and meta-analysis by Bjelakovic and colleagues in JAMA in 2007 pooled randomized trials of antioxidant supplements for primary and secondary prevention and did not find the mortality reduction the hypothesis predicted.

What the field concluded, and why it is relevant here. Oxidative signalling turns out to be physiologically necessary, not merely damaging — reactive oxygen species are messengers in insulin signalling, immune defence and the adaptive response to exercise. Blanket suppression is not obviously desirable. Meanwhile, the observational finding that diets rich in colourful plant foods track with better health has held up well. The gap between those two observations is now generally explained by something other than radical scavenging: the fibre, the food matrix, the displacement of worse foods, and specific signalling effects of individual compounds and their metabolites.

So: eat the plants. Do not buy the mechanism.

What Anthocyanins May Actually Do: Signalling, Not Scavenging

The modern account of anthocyanin activity has almost nothing to do with mopping up radicals, and it is genuinely more plausible. Tier: preliminary, mostly cell and animal work, with some supportive human data on intermediate endpoints for anthocyanin-rich foods generally — not for butterfly pea.

The leading candidate mechanisms:

Nrf2 activation. Rather than neutralising oxidants directly, some polyphenols and their metabolites mildly stress the cell's redox sensors, which activates the Nrf2 transcription factor and upregulates the cell's own antioxidant and detoxification enzymes. This is a catalytic, amplified response, and it works at the low concentrations that are actually achievable — unlike stoichiometric scavenging, which would need far more compound than is ever present.
Endothelial function. Anthocyanin-rich foods have been associated with improvements in measures of vascular function in human studies. The mechanistic proposals involve nitric oxide availability rather than antioxidant chemistry.
Enzyme inhibition at the gut surface. Polyphenols can inhibit digestive enzymes and sugar transporters in the intestinal lumen, which requires no absorption at all — the compound acts where it already is. For butterfly pea this is the most credible route by a wide margin, and it is the subject of the next section.
Microbiome interaction. Polyphenols reaching the colon are substrates for bacteria and can shift microbial composition. Since the metabolites those bacteria produce are the compounds that reach your circulation, this is both a mechanism and a source of person-to-person variability.
Anti-inflammatory signalling in cell models. The work by Nair and colleagues in the Journal of Agricultural and Food Chemistry tested butterfly-pea ternatins and quercetin glycosides against lipopolysaccharide-stimulated macrophages and reported protective effects. This is cell culture at concentrations chosen by the experimenter; read it as a hypothesis about mechanism.

Note the honest shape of this section: the plausible mechanisms are mostly ones that do not require the pigment to be absorbed and do not depend on antioxidant chemistry. The best case for butterfly pea is a local, gut-surface effect. Which is, conveniently, what the human trials actually looked at.

Blocking Starch Digestion: The In-Vitro Case

Tier: in vitro. Well established as a chemical fact, and the mechanistic basis for everything in the next two sections.

When you eat starch or sucrose, two enzyme families do the work. Pancreatic α-amylase chops long starch chains into shorter oligosaccharides in the small-intestinal lumen. α-Glucosidases anchored in the brush border of the intestinal wall — including sucrase-isomaltase and maltase-glucoamylase — then cut those into free glucose, which is what actually crosses into the blood.

Inhibit either step and glucose enters the circulation more gradually. This is a real, licensed pharmacological strategy, not a wellness invention: acarbose, miglitol and voglibose are α-glucosidase inhibitors used clinically in type 2 diabetes, and their characteristic side effects — flatulence, bloating, loose stools from undigested carbohydrate reaching the colon — are the direct signature of the mechanism working.

Butterfly-pea flower extract inhibits both enzymes in the test tube. Adisakwattana and colleagues, in BMC Complementary and Alternative Medicine in 2012, screened plant-based foods and their combinations against intestinal α-glucosidase and pancreatic α-amylase; the same Bangkok group went on to run the human work below. Escher and colleagues reported a broader in-vitro panel on butterfly-pea petal bioactives in Food Research International, including enzyme inhibition and effects on LDL oxidation.

Three constraints on how far this travels:

Potency. Botanical enzyme inhibitors are typically weak by pharmaceutical standards. The concentration needed for half-maximal inhibition in a cuvette is usually far above what a cup of tea puts into an intestinal lumen full of food. "Inhibits the enzyme" is a statement about a chemical interaction, not about a clinically meaningful slowdown.
The test tube is not the gut. Real intestinal contents contain bile, protein, fat, fibre, mucus and a large excess of substrate, and polyphenols bind non-specifically to protein. In-vitro inhibition routinely fails to survive that environment.
It is not a general "blood sugar" effect. This mechanism can only blunt the rise after a carbohydrate-containing meal. It cannot lower fasting glucose, it cannot improve insulin sensitivity, and it has no bearing on glucose entering the blood from the liver. Butterfly pea taken away from food, on this mechanism, should do nothing at all to glucose — and in the human trial, it did not.

The Human Post-Meal Glucose Trial

Tier: randomized clinical trial — small, acute, healthy participants, no clinical endpoint. This is the best study that exists on butterfly pea and it is worth understanding exactly, because it is routinely cited as proof of something much larger than it examined.

Chusak, Thilavech, Henry and Adisakwattana published it in BMC Complementary and Alternative Medicine in 2018, from Chulalongkorn University in Bangkok. The design was a randomized crossover: a small group of healthy young men, each of whom consumed several different 400 mL beverages on separate occasions — sucrose alone, butterfly-pea flower extract alone at two doses, and sucrose combined with each dose of extract — with blood sampled over the following three hours.

What was reported:

When the extract was consumed with sucrose, post-meal glucose and insulin were lower than with sucrose alone, with the effect most visible in the early part of the curve.
The extract on its own did not lower fasting glucose. This is an important finding in both directions: it is consistent with a gut-surface enzyme mechanism rather than a systemic glucose-lowering drug effect, and it is reassuring on safety, because it means the extract did not push glucose below normal in people who had not eaten.
Plasma antioxidant capacity measures rose after the extract, and a marker of lipid oxidation fell.

What it does not show, stated plainly:

The population is the narrowest possible. Healthy, slim, young men. Not people with diabetes, prediabetes or insulin resistance — precisely the people the finding is marketed to. Metabolically healthy young men handle a sugar load well; a small blunting of their glucose curve has no established clinical meaning.
The exposure was a single drink and the follow-up was three hours. No repeat dosing, no HbA1c, no fasting insulin over time, no weight, no clinical outcome of any kind. Acute postprandial studies are how this kind of research starts, and they are not evidence of durable metabolic benefit.
The intervention was concentrated extract, not tea. Gram quantities of extract in 400 mL, taken with the sugar. See the dose-reality section below for how far that is from a cup of flowers.
The antioxidant-capacity finding is the assay problem from earlier in this page, appearing in a real trial. A rise in plasma FRAP after a polyphenol drink is expected and is not a health outcome.

The fair summary: this is a competent, honestly reported acute study supporting the idea that butterfly-pea flower extract modestly blunts a sugar spike when taken with sugar, most likely by inhibiting carbohydrate digestion at the gut surface. That is a real finding. It is not evidence that blue tea helps anyone manage diabetes, and nobody has tested that.

The High-Fat Meal Trial

Tier: randomized clinical trial — small, acute, mixed result, more relevant population.

The same Bangkok group followed up with a study in overweight and obese men, published in Biology (Basel) in 2021 with Thilavech as first author. Participants ate a high-fat meal on separate occasions — alone and with each of two doses of butterfly-pea flower extract — with blood sampled over six hours.

Reported as positive: at the higher extract dose, the rise in serum triglycerides after the fatty meal was attenuated, free fatty acids later in the curve were lower, and plasma antioxidant status measures and glutathione peroxidase activity were higher.

Reported as negative, and this deserves equal billing: there was no difference in post-meal glucose between conditions in this population, and no change in the inflammatory cytokines interleukin-6, interleukin-1β or TNF-α.

Two observations worth making. First, the study chose a better population than its predecessor — overweight and obese participants are closer to the people who care about post-meal lipids — which is the right direction for a research programme to move. Second, and more instructive, the negative results are informative: the absence of a glucose effect here, in contrast to the earlier trial, is consistent with a mechanism that acts on carbohydrate digestion and therefore needs carbohydrate to act on. A high-fat meal gives it little to do. That coherence is a point in favour of the mechanism being real, even as it further narrows what the mechanism can be claimed to achieve.

Still sixteen or so people, one meal, six hours, no clinical endpoint.

If You Take Diabetes Medication

This is the one place on this page where the practical stakes are real, and it is worth being concrete.

The mechanism above — slowing carbohydrate digestion — is the mechanism of a licensed drug class. If you are already taking a medication that lowers blood glucose, adding something that pushes in the same direction is a genuine interaction consideration rather than a theoretical one.

The reassuring part. Butterfly-pea flower extract did not lower fasting glucose in the human trial, and the mechanism does not create glucose-lowering in the absence of a carbohydrate meal. Culinary use — a cup of flower tea, blue rice — delivers a very small dose. There is no report of hypoglycaemia from drinking butterfly-pea tea.
Where care is warranted. Concentrated extracts and capsules taken regularly with meals, by someone on insulin or a sulfonylurea such as glipizide, gliclazide or glimepiride — the two classes that can drive glucose too low on their own. If you start using concentrated extract while on either, monitor your readings more closely for the first couple of weeks and tell your prescriber.
Overlapping side effects. If you already take acarbose or another α-glucosidase inhibitor, stacking a botanical inhibitor on top may simply add gas and bloating without adding benefit.
Watch out for the reverse error. The bigger practical risk with this plant is not an interaction. It is someone concluding from a headline that blue tea manages their blood sugar and becoming less rigorous about medication, diet or monitoring. Nothing in this literature supports substituting butterfly pea for any part of diabetes care.
Surgery. Stop concentrated botanical extracts around two weeks before planned surgery, following the general principle for any supplement with plausible glucose or platelet effects.

Dose Reality: What Is Actually in a Cup

This section does more to deflate the marketing than any of the mechanistic argument above, and it is simple arithmetic.

What the trials used: one to two grams of concentrated flower extract in 400 mL of liquid, consumed with a meal. An extract represents the soluble material pulled out of a considerably larger mass of dried flower and then concentrated.
What a home-brewed cup contains: eight to ten dried flowers — well under a gram of plant material in total — of which only a fraction dissolves into the water, of which the anthocyanins are only one component. The anthocyanin dose in a cup of butterfly-pea tea is small, and considerably smaller than the trial dose.
Compared with food: a normal serving of blueberries, blackcurrants, blackberries or a glass of dark berry juice supplies substantially more anthocyanin than a cup of flower infusion. Anyone genuinely trying to raise anthocyanin intake should eat berries, which also bring fibre and are cheaper.
Culinary colouring: a small quantity of flower, diluted through an entire dish. Negligible as a dose of anything.

The conclusion this forces is worth stating without hedging. As a beverage, butterfly pea is best understood as a pleasant, caffeine-free, mildly polyphenol-containing drink — not as a dose of an active compound. If you want the amount used in the trials, you need a standardised extract, and at that point you have moved from a traditional food with centuries of use to a concentrated supplement whose long-term human safety has never been studied. That trade is worth noticing before making it.

Evidence Tier Summary

Butterfly-pea flowers contain anthocyanins and other flavonoids. Established analytical chemistry.
Butterfly-pea extract scores well on in-vitro antioxidant assays. Established, and of limited relevance to health.
Butterfly-pea extract inhibits α-amylase and α-glucosidase in vitro. Established in vitro; potency modest; relevance in a real gut unproven.
A concentrated extract taken with sucrose modestly blunted post-meal glucose and insulin in healthy young men over three hours. Randomized clinical trial, one small acute study.
A concentrated extract taken with a high-fat meal attenuated the post-meal triglyceride rise in overweight and obese men over six hours. Randomized clinical trial, one small acute study; no glucose effect and no cytokine effect in that trial.
Butterfly pea raises "plasma antioxidant capacity". Measured, and not a health outcome.
Butterfly pea reduces oxidative damage to human tissues. Not established.
Butterfly pea improves long-term glycaemic control, HbA1c, weight or lipid profile. Not tested. No trial of repeated dosing over weeks or months exists.
Butterfly pea prevents or treats diabetes, heart disease or any other condition. Not established, and not supported by anything above.

Key Research Papers

Cited as PubMed topic searches rather than numeric identifiers. Where exact metadata or a numeric result is not certain, the finding is described and the search provided.

Chusak, Thilavech, Henry and Adisakwattana, BMC Complementary and Alternative Medicine, 2018 — the acute randomized crossover trial of a Clitoria ternatea flower beverage on glycaemic response and antioxidant capacity in healthy subjects. The central human study. Search PubMed.
Thilavech, Adisakwattana and colleagues, Biology (Basel), 2021 — butterfly-pea flower extract, postprandial lipaemia and plasma antioxidant status after a high-fat meal in overweight and obese participants. Search PubMed.
Adisakwattana and colleagues, BMC Complementary and Alternative Medicine, 2012 — in-vitro inhibition of intestinal α-glucosidase and pancreatic α-amylase by plant-based foods and their combinations. The mechanistic groundwork. Search PubMed.
Manach, Williamson, Morand, Scalbert and Rémésy, American Journal of Clinical Nutrition, 2005 — the review of 97 human polyphenol bioavailability studies that placed anthocyanins at the low end of absorbed classes. Search PubMed.
Czank, Cassidy and colleagues, American Journal of Clinical Nutrition, 2013 — the carbon-13 tracer study of human metabolism and elimination of cyanidin-3-glucoside, which reframed anthocyanin bioavailability around metabolites. Search PubMed.
de Ferrars and colleagues, British Journal of Pharmacology, 2014 — the pharmacokinetics of anthocyanins and their metabolites in humans. Search PubMed.
Williamson and Clifford, Biochemical Pharmacology, 2017 — the roles of small intestine, colon and microbiota in determining the metabolic fate of polyphenols. Search PubMed.
Bjelakovic and colleagues, JAMA, 2007 — mortality in randomized trials of antioxidant supplements for primary and secondary prevention: the meta-analysis that closed the simple antioxidant hypothesis. Search PubMed.
Klein and colleagues, JAMA, 2011 — the SELECT trial of vitamin E and selenium and prostate cancer risk. Search PubMed.
The Alpha-Tocopherol Beta-Carotene Cancer Prevention Study Group, New England Journal of Medicine, 1994, and Omenn and colleagues (CARET), New England Journal of Medicine, 1996 — the beta-carotene trials in high-risk populations. Search PubMed.
Nair and colleagues, Journal of Agricultural and Food Chemistry, 2015 — ternatin anthocyanins and quercetin glycosides from butterfly-pea petals against lipopolysaccharide-induced inflammation in macrophages. Cell culture. Search PubMed.
Escher and colleagues, Food Research International — the broad in-vitro panel on butterfly-pea petal bioactives, including enzyme inhibition and LDL oxidation. Search PubMed.
On the limitations of ORAC and related capacity assays as predictors of biological effect — the methodological literature behind the USDA withdrawal. Search PubMed.
On Nrf2 activation by dietary polyphenols as an alternative to direct radical scavenging. Search PubMed.

Connections

All Herbs
Butterfly Pea (Clitoria ternatea) — the main topic page, with the full trial descriptions, forms and dosage.
Anthocyanins and Why It Changes Colour — the pigment chemistry that the antioxidant claim is built on top of.
Antioxidants — the category overview, including how the antioxidant hypothesis was tested and what replaced it.
Anthocyanins — sources, absorption and the wider human evidence for the pigment class.
Anthocyanins — Benefits — a fuller treatment of bioavailability and metabolite biology.
Quercetin — Benefits — a flavonol with the same bioavailability problem and a similar claims gap.
EGCG — the green-tea catechin, another polyphenol whose in-vitro power far exceeds its in-vivo reach.
Insulin Resistance — what post-meal glucose findings are usually stretched to cover.
Prediabetes — the population butterfly pea is marketed to and has never been tested in.
Diabetes — the condition, and why an acute postprandial study is not evidence about it.
Hemoglobin A1C — the outcome that would need to move before a glycaemic claim was meaningful.
Lipid Panel — where a fasting triglyceride effect would have to show up, as opposed to a six-hour post-meal curve.
Hibiscus — Benefits — a flower tea whose human trial base is genuinely larger, for calibration.
Blueberries — a much larger anthocyanin dose per serving, with fibre attached.

Safety and disclaimer. This page is educational and is not medical advice, and nothing on it should be used to change how you manage a diagnosed condition. Butterfly-pea flower as a tea or food colouring has a long culinary record and no established toxicity signal at those amounts; concentrated extracts and capsules have not been tested for long-term human use. If you take insulin, a sulfonylurea or any other glucose-lowering medication, speak to your prescriber before starting concentrated butterfly-pea extract and monitor your readings. Avoid concentrated extracts in pregnancy and breastfeeding, do not give supplements to children, and do not use any herbal product as a substitute for prescribed treatment for diabetes or cardiovascular disease.

Back to Table of Contents