evipedia.ai
← Back

Dihydromyricetin for Health & Longevity

Evidence Review created on 11/07/2026 using AI4L / Opus 5.5

Also known as: DHM, DMY, Ampelopsin, Ampeloptin

Motivation

Dihydromyricetin (DHM, also called ampelopsin) is a plant compound related to the colored substances found in tea, grapes and berries. It is most concentrated in vine tea, a climbing plant brewed as a herbal tea in southern China, and also occurs in the Japanese raisin tree. It draws interest because it acts on the brain’s main calming signaling system and on the energy-sensing machinery inside cells, which links it to alcohol’s effects, blood sugar handling and liver fat.

Vine tea has a long history as a folk remedy in China, and the Japanese raisin tree has been used in East Asia against the after-effects of drinking. Interest in the purified compound grew after animal research on alcohol intoxication drew wide attention; it is now sold mainly in after-drinking products and liver and metabolic supplements. A small number of human trials have tested it in people with fatty liver disease and type 2 diabetes.

This review examines what human, animal and laboratory studies show about dihydromyricetin’s effects on metabolic and liver health and on alcohol-related outcomes, together with its safety, interactions, dosing and practical use, for adults who consider it as part of a health and longevity strategy.

Benefits - Risks - Protocol - Conclusion

This section lists expert commentary, high-level overviews and landmark papers that discuss dihydromyricetin directly.

No dihydromyricetin-focused content was found from Peter Attia, Andrew Huberman, Chris Kresser, Life Extension Magazine or Lifespan.io; web and on-site searches returned no matching article or episode.

Grokipedia

  • Ampelopsin

    An encyclopedia-style overview of dihydromyricetin’s chemistry, plant sources, traditional uses, preclinical actions and registered trials, useful as orientation before reading the primary studies.

Examine

Examine has no dedicated dihydromyricetin article; its site search returns no results for the term. Its closest page covers the Japanese raisin tree (Hovenia dulcis), a plant source of the compound:

  • Japanese Raisin Tree

    Summarizes the traditional use of the Japanese raisin tree around drinking and notes a human study in which taking it before alcohol lowered circulating alcohol levels.

ConsumerLab

Systematic Reviews

This section lists the systematic reviews and meta-analyses that address dihydromyricetin directly, all published 2025–2026.

The principal trade-off, safety, is unrepresented: no systematic review or meta-analysis of dihydromyricetin adverse events or interactions exists.

Mechanism of Action

Dihydromyricetin is a flavanonol (a subclass of flavonoids, the plant pigment family) that acts on several targets rather than one receptor:

  • Brain signaling: In rats it binds the benzodiazepine site (the pocket used by sedatives such as diazepam) of GABA-A receptors (the brain’s main calming receptors) and opposes alcohol’s effects there (Shen 2012).
  • Energy pathways: It activates AMPK (the cell’s energy sensor, which shifts metabolism toward fat burning) and lowers liver fat in mice (Silva 2020).
  • Alcohol breakdown (contested): Mice showed more alcohol-processing enzymes (Silva 2020), whereas rats showed unchanged alcohol clearance (Skotnicová 2020).
  • Aging biology: It inhibits DNMT1 (the enzyme copying DNA methylation marks) in skin cells, in work by the skin-care manufacturer Beiersdorf (Falckenhayn 2023), and acts through PRDX2 in aged cells (Xu 2026).

Pharmacology: Selectivity is low. In rats, oral bioavailability is about 4% and oral half-life about 3.7 hours (Liu 2017); human pharmacokinetic and tissue-distribution data are unpublished. A review of animal data describes metabolism by reduction, methylation, glucuronidation and sulfation (attaching sugar-acid or sulfate groups for excretion) (Liu 2019). In human liver preparations it inhibited CYP3A4, CYP2E1 and CYP2D6 (drug-metabolizing liver enzymes) (Liu 2017), but a later study found little CYP3A4 inhibition (Dombi 2024).

Historical Context & Evolution

Dihydromyricetin’s history runs through two plants. Vine tea (Ampelopsis grossedentata), whose leaves contain up to 30–40% dihydromyricetin, has long been brewed in southern China as a folk remedy (review: He 2024). The Japanese raisin tree (Hovenia dulcis) has been used in East Asia against alcohol intoxication (Kim 2017).

Health-optimization interest began with a 2012 rat study by Jing Liang’s team at the University of California, Los Angeles, which reported that the purified compound counteracted intoxication and withdrawal through GABA-A receptors (Shen 2012). Capsules then entered the United States market as hangover products (review: Liu 2019). A Chinese military-university nutrition group ran the first randomized trials, reporting lower glucose and liver enzymes in fatty liver disease (Chen 2015) and lower glucose in type 2 diabetes (Ran 2019).

Opinion has since moved in both directions. Supporting work added liver protection in alcohol-fed mice (Silva 2020), fruit-fly lifespan extension (Fan 2020) and aged-cell effects in mice (Xu 2026). Contrary findings followed: a Czech rat study found unchanged alcohol clearance (Skotnicová 2020), a 2021 audit found no human data for marketed hangover products (Verster 2021), and a 2026 study showed reactions with worm culture medium that can produce artifacts (Wu 2026). The first independent 12-month trial, published in 2026, met secondary but not primary liver goals (Michailidou 2026). The compound remains well studied in the laboratory, with a thin human record.

Expected Benefits

High 🟩 🟩 🟩

No benefit reaches High: the replicated human findings on glucose and liver enzymes come from randomized trials by a single research group, plus one independent trial that tested a multi-ingredient product.

Medium 🟩 🟩

Lower Fasting Glucose and Insulin Resistance

In a 3-month double-blind randomized controlled trial (RCT) of 60 adults with NAFLD, dihydromyricetin lowered fasting glucose and HOMA-IR (an index of insulin resistance from fasting glucose and insulin) versus placebo (Chen 2015). The same Chongqing group (Mi and colleagues) found lower fasting glucose and glycated albumin (a 2–3-week glucose marker) in 70 completers with type 2 diabetes, using vine tea rather than purified compound, an indirect form (Ran 2019). No trial has enrolled metabolically healthy adults.

Magnitude: Direction only: both Chongqing trials report statistically significant reductions versus their control arms, and their abstracts, the only accessible text, report no outcome figure.

Lower Liver Enzymes in Fatty Liver Disease ⚠️ Conflicted

Chen 2015 found lower ALT, AST and GGT (enzymes released by injured liver cells) versus placebo (Chen 2015). An independent Athens trial gave a commercial product (Hepatrat, Uni-Pharma) combining dihydromyricetin with vitamins C and E and choline for 12 months; the abstract states 300 mg/day, the methods 600 mg/day, and this review follows the methods (Michailidou 2026). Its primary endpoint, ALT improvement, was not met, while combined ALT and GGT normalization was. Net reading: the direction is consistent, but the only long trial missed its primary endpoint.

Magnitude: In the intention-to-treat analysis (all randomized participants analyzed), combined ALT and GGT normalization reached 32.1% versus 3.7% with placebo (P = 0.012, where P below 0.05 conventionally marks statistical significance), while the primary endpoint was 46.4% versus 29.6% (P = 0.315, not significant) (Michailidou 2026).

Sleep Duration and Perceived Stress

A 20-day placebo-controlled trial of 200 mg daily came from Jing Liang’s group, which develops dihydromyricetin; participants chose their own allocation number and self-reported outcomes (Shao 2022). The abstract describes a placebo comparison, but the results report within-group changes, which this review follows. Proposed mechanism: GABA-A support (Shen 2012).

Magnitude: Dihydromyricetin group: sleep rose from 5.45 to 6.52 hours and stress score fell from 2.86 to 1.25; placebo group: no significant change; not compared between groups (Shao 2022).

Low 🟩

Lower LDL Cholesterol ⚠️ Conflicted

Chen 2015 found lower LDL cholesterol (the main artery-clogging cholesterol fraction) and apolipoprotein B versus placebo (Chen 2015), but Ran 2019 found no lipid change after 1 month (Ran 2019); the shorter duration may explain this. Net reading: one positive 3-month trial is balanced by one shorter null trial.

Magnitude: Direction only: a reduction versus placebo in one trial and no change in the other; the accessible abstracts report no outcome figure.

Kidney Function Markers in Type 2 Diabetes

Vine tea supplying 970 mg dihydromyricetin daily lowered cystatin C (a kidney filtration marker) and retinol-binding protein 4 versus dihydromyricetin-depleted vine tea for 1 month (Ran 2019). Indirect: tested as whole-leaf tea, not purified compound.

Magnitude: Direction only: both markers fell significantly versus control; the accessible abstract reports no outcome figure.

Hangover Symptoms

Evidence is indirect: Japanese raisin tree extracts, not isolated dihydromyricetin, were tested. They reduced hangover scores in men carrying an ALDH2 variant (impairing the main acetaldehyde-clearing enzyme) (Kim 2017) and in a trial by the manufacturer Kwangdong Pharmaceutical (Lee 2024).

Magnitude: In an observational cohort sponsored by Cheers Health, testing its multi-ingredient dihydromyricetin and L-cysteine product (indirect), mental clarity was 0.36 points higher on a 1–10 scale (95% confidence interval, the range likely to contain the true effect, 0.21–0.50) than on alcohol-only nights; supplement use was self-selected, not randomized (Song 2026).

Epigenetic Skin Age and Wrinkles

A topical multi-ingredient serum from Beiersdorf, which funded and authored the study, was applied for 8 weeks by 60 adults (Qi 2026). Indirect: skin route and added ingredients; no vehicle control.

Magnitude: Mean epidermal DNA-methylation age fell 2.1 years and wrinkle visibility 13.9% from baseline (no control group) (Qi 2026).

Speculative 🟨

Reduced Senescent-Cell Burden

Dihydromyricetin dampened harmful secretions of senescent cells (aged cells that stop dividing) and cleared aged brain immune cells in mice (Xu 2026). Mechanistic and animal basis only.

Longer Lifespan in Model Organisms ⚠️ Conflicted

It extended fruit-fly (Fan 2020) and worm lifespan (Jo 2026), but depletes amino acids in worm medium (Wu 2026). Animal basis only. Net reading: invertebrate lifespan gains are unconfirmed.

Protection Against Alcohol Liver Injury ⚠️ Conflicted

Alcohol-fed mice had less liver fat and injury (Silva 2020), whereas rat liver cells were only slightly protected (Skotnicová 2020). Net reading: animal results disagree.

Reduced Alcohol Intake and Withdrawal

Rats given dihydromyricetin drank less alcohol and showed fewer withdrawal signs (Shen 2012). No human trial has tested drinking behavior or alcohol use disorder; animal basis only.

Brain Aging and Neurodegeneration

A review of cell and rodent models of Alzheimer’s, Parkinson’s and Huntington’s disease reports antioxidant and nerve-protective effects (Martínez-Coria 2019). No human data exist; preclinical basis only.

Preserved Muscle Mass

Dihydromyricetin limited inflammation-driven muscle wasting in mice (Hou 2021) and in chemically aged rats (Kou 2017). Animal basis only.

Benefit-Modifying Factors

  • Genetic polymorphisms: The Kim 2017 hangover trial enrolled men with one copy of the ALDH2 variant, and CYP2E1 variants (in the enzyme that also oxidizes alcohol) appeared to modify the response (Kim 2017); no variant is known to modify metabolic benefits.
  • Baseline biomarkers: Glucose and liver-enzyme reductions were seen in people starting with raised values (fatty liver with elevated ALT or GGT, type 2 diabetes); effects at normal baseline values are untested.
  • Sex: Women were overrepresented in the Athens dihydromyricetin arm, but sex did not significantly affect outcomes (Michailidou 2026); no sex-specific analyses exist elsewhere.
  • Pre-existing conditions: Human benefit data come from fatty liver disease and type 2 diabetes; people with normal metabolism, the usual profile of the target audience, have not been studied.
  • Age: Trials enrolled adults aged 18–70; the sleep trial found no age dependence between 18 and 60 (Shao 2022), and adults over 70 are unstudied.
  • Formulation and absorption: Low oral absorption in animals means formulation may decide exposure; according to a review, enhanced formulations such as liposomes have not been tested in humans (He 2024).

Potential Risks & Side Effects

High 🟥 🟥 🟥

No risk reaches High: no randomized trial has documented an adverse event attributable to dihydromyricetin.

Medium 🟥 🟥

No risk reaches Medium: the only controlled safety data, a 55-person, 12-month randomized trial, recorded no drug-related adverse events in either arm (Michailidou 2026).

Low 🟥

Additive Glucose Lowering with Diabetes Medication

Dihydromyricetin lowered glucose in type 2 diabetes (Ran 2019), so low glucose is plausible with insulin or sulfonylureas (diabetes drugs that make the pancreas release insulin). Indirect: trials measured glucose reduction, not low-glucose events.

Magnitude: Not quantified in available studies. No trial has measured low-glucose episodes in people combining dihydromyricetin with glucose-lowering drugs.

Heavier Drinking Enabled by Hangover Relief ⚠️ Conflicted

Hangover products may lower a barrier to heavy drinking (Verster 2021). Contrary: users did not increase intake (Song 2026) and rats drank less (Shen 2012). Net reading: no human evidence of increased drinking.

Magnitude: Direction: stable AUDIT-C scores (a three-question alcohol-intake screen) over 6 weeks in a sponsor-funded observational cohort; the accessible report gives no outcome figure (Song 2026).

Speculative 🟨

Altered Levels of Co-Administered Drugs ⚠️ Conflicted

It inhibited CYP3A4, CYP2E1 and CYP2D6 in vitro (Liu 2017), but another study found minor inhibition (Dombi 2024). Net reading: unresolved; no human data.

Altered Response to Sedatives

Binding at the benzodiazepine site of GABA-A receptors (Shen 2012) could change the effect of sedatives in either direction. Mechanistic basis only; no human data.

Reduced Iron Absorption

A review notes that its pyrogallol structure binds iron and other metals (Liu 2019), which may reduce absorption of dietary iron. Mechanistic basis only; never measured in humans.

Unknown Long-Term Safety

Human data stop at 12 months; a European Food Safety Authority (EFSA) panel could not establish the safety of a dihydromyricetin-containing raisin tree extract (Turck 2020). Basis: missing data, not reported harm.

Risk-Modifying Factors

  • Genetic polymorphisms: People with reduced CYP2D6 activity (genetic “poor metabolizers”) rely more on other clearance routes, so any CYP3A4 inhibition may matter more; this is inferred from enzyme data, with no human study (Liu 2017).
  • Baseline biomarkers: People with normal or low-normal fasting glucose or a history of low-glucose episodes have the least margin for additional glucose lowering; baseline ALT allows later liver changes to be interpreted.
  • Sex: No sex differences in risk are reported; pregnancy and breastfeeding are unstudied (Turck 2020).
  • Pre-existing conditions: Diabetes treated with insulin or sulfonylureas, advanced liver disease, alcohol use disorder and transplant or seizure medications that depend on CYP3A4 raise the stakes of any interaction or glucose effect.
  • Age: Older adults more often take several medications, raising interaction exposure; no trial enrolled people over 70.

Key Interactions & Contraindications

Documented interactions:

  • Alcohol (tested with Japanese raisin tree extract): Unchanged blood alcohol and acetaldehyde in one trial (Kim 2017); lower blood alcohol at 0.5 and 6 hours in a Kwangdong Pharmaceutical trial (Paik 2024). Consequence: impairment not reliably reduced. Mitigation: driving fitness judged by intake alone.

Theoretical interactions (inferred from mechanism, not studied in humans):

  • Narrow-margin CYP3A4 drugs (tacrolimus, cyclosporine, simvastatin): Inferred from in vitro CYP3A4 inhibition (Liu 2017); raised drug levels could cause toxicity such as kidney injury or muscle damage. Mitigation: drug-level or symptom monitoring with the prescriber.
  • CYP2D6 drugs (metoprolol, codeine, tamoxifen, dextromethorphan): Inferred from in vitro CYP2D6 inhibition (Liu 2017); slowed heart rate with metoprolol, dizziness with dextromethorphan, or weaker pain relief from codeine and reduced tamoxifen effect. Mitigation: pharmacist review.
  • Acetaminophen (over-the-counter): Inferred from in vitro CYP2E1 inhibition (Liu 2017), the enzyme forming acetaminophen’s liver-toxic by-product; the direction of any effect on liver risk is unknown. Mitigation: acetaminophen kept within labeled doses.
  • Glucose-lowering drugs (insulin, glipizide, metformin): Additive glucose lowering inferred from trial glucose reductions (Chen 2015); consequence: low blood glucose. Mitigation: home glucose checks during the first weeks.
  • Benzodiazepines and Z-drugs (diazepam, alprazolam, zolpidem; Z-drugs are non-benzodiazepine sleep drugs acting at the same site): Inferred from shared binding site (Shen 2012); altered sedation. Mitigation: no simultaneous start of both before driving.
  • P-glycoprotein substrates (digoxin, dabigatran): P-glycoprotein is a drug-export pump; inferred from rat data in which dihydromyricetin raised triptolide exposure (Deng 2020); consequence: digoxin toxicity or bleeding with dabigatran. Mitigation: level or clotting monitoring.
  • Glucose-lowering supplements (berberine, cinnamon extract, bitter melon): Additive glucose lowering inferred from shared effect; consequence: low blood glucose. Mitigation: introduction of one at a time.
  • Sedating supplements (kava, valerian): Additive or altered calming effect inferred from GABA-A action; consequence: drowsiness. Mitigation: no combination before driving.
  • Iron supplements and iron-rich meals: Metal binding inferred from chemistry described in a review (Liu 2019); consequence: lower iron absorption. Mitigation: separation by at least 2 hours.

Populations who should avoid Dihydromyricetin:

  • Pregnancy and breastfeeding: EFSA’s assessment of a Japanese raisin tree extract excluded pregnant and lactating women from the target population (Turck 2020); no human data.
  • Advanced liver disease (Child-Pugh Class B or C, a scoring of liver-failure severity): EFSA’s assessment excluded people with liver malfunction (Turck 2020); inferred from untested clearance in a failing liver, no human data.
  • Children and adolescents under 18: Inferred from its in vitro inhibition of drug-metabolizing liver enzymes, whose activity is still maturing in children, and untested clearance; no human data.
  • Alcohol use disorder: Inferred from the possibility that hangover relief sustains heavy drinking (Verster 2021); no human data in this group.
  • Users of narrow-margin CYP3A4 drugs (tacrolimus, cyclosporine): Inferred from in vitro CYP3A4 inhibition (Liu 2017); no human data.

Risk Mitigation Strategies

Doses and timings below follow common practice unless cited.

  • Glucose checks on diabetes medication: Home checks in weeks 1–4 catch low blood glucose; the alert level is below 70 mg/dL (American Diabetes Association and European Association for the Study of Diabetes statement), whose members gain no revenue from it.
  • Liver enzyme check: ALT at baseline and 3 months detects liver injury; values above the American College of Gastroenterology healthy limit, 33 U/L (men) or 25 U/L (women), prompt medical review (Kwo 2017); its members gain no revenue from it.
  • Medication review: A pharmacist check for CYP3A4, CYP2D6 and P-glycoprotein drugs before starting limits the risk of altered drug levels; drug-level monitoring applies to tacrolimus or cyclosporine.
  • No reliance for alcohol impairment: Treating blood alcohol as unchanged prevents impaired driving, because human data on alcohol clearance are inconsistent (Kim 2017; Paik 2024).
  • Lowest studied dose first: Starting at 200 mg/day, the lowest studied oral dose (Shao 2022), limits exposure while long-term safety remains unknown.
  • Iron spacing: Taking dihydromyricetin at least 2 hours apart from iron supplements limits reduced iron absorption.
  • Dose-disclosed products: Choosing products that state the dihydromyricetin amount avoids unknown exposure; 59% of dihydromyricetin-containing hangover products did not report it (Verster 2021).
  • Sedative caution: Avoiding first-time combination with benzodiazepines, Z-drugs, kava or valerian before driving prevents unexpected drowsiness.

Therapeutic Protocol

Timing, cycling and titration parameters without a citation reflect common practice.

  • Metabolic and liver protocol: Two 150 mg capsules twice daily for 3 months, per the abstract (Chen 2015); 600 mg/day in a multi-ingredient tablet for 12 months (Michailidou 2026); 300 mg twice daily in a registered diabetes trial (NCT03606694).
  • Whole vine tea alternative: 10 g of vine tea daily supplying about 970 mg dihydromyricetin for 1 month (Ran 2019); popularized by the Chongqing nutrition group of Mi Mantian.
  • Sleep and stress protocol: 200 mg once daily for 20 days, taken at the same time each day (Shao 2022); popularized by Jing Liang’s laboratory, now at the University of Southern California.
  • Alcohol-related use: No trial of isolated dihydromyricetin sets a dose; 2,460 mg of Japanese raisin tree extract was taken with alcohol (Kim 2017). Commercial products are often taken before or during drinking.
  • Topical approach: A dihydromyricetin serum applied twice daily for 8 weeks (Qi 2026) is a separate approach, popularized by Beiersdorf, with no overlap with oral protocols.
  • Time of day: Trials used morning and evening dosing with meals or a fixed daily time; no study compared times of day.
  • Half-life: Oral half-life in rats is about 3.7 hours (Liu 2017); no human half-life is published.
  • Single or split dosing: The short animal half-life explains why metabolic trials split the daily amount into two doses; once-daily dosing was used only for sleep.
  • Genetic polymorphisms: No variant is known to change dose; ALDH2 and CYP2E1 variants may modify hangover response (Kim 2017).
  • Sex differences: No sex-specific dosing or efficacy data exist; trials used the same dose for men and women.
  • Age: Doses were tested in adults 18–70; adults over 70 have no dosing data, so the lowest studied dose is the usual starting point.
  • Baseline biomarkers: Responses were documented in people with raised glucose or liver enzymes; at normal values, a measurable change is less likely.
  • Pre-existing conditions: Fatty liver disease and type 2 diabetes are the only studied conditions; other conditions have no protocol data.

Discontinuation & Cycling

  • Duration: No lifelong use has been studied; trials ran from 20 days to 12 months (Michailidou 2026), so use is typically framed as a time-limited trial reassessed with laboratory tests.
  • Withdrawal effects: None reported in human trials; Liang’s group reports that dihydromyricetin did not produce tolerance in rodents, unlike benzodiazepines (Shao 2022).
  • Tapering: Not required; no rebound or withdrawal syndrome is documented, so stopping at once is the usual practice.
  • Cycling: No evidence shows that cycling preserves effect; benefits on glucose and liver enzymes were seen with continuous daily dosing, and a recheck of laboratory values after 3 months guides continuation.

Sourcing and Quality

  • Source: Most dihydromyricetin is extracted from vine tea leaves, which contain up to 30–40% of the compound according to a review (He 2024); Japanese raisin tree extracts contain only small amounts (Turck 2020).
  • Purity: Trial-grade material was at least 98% pure by high-performance liquid chromatography (a laboratory purity test) (Shao 2022); labels or certificates of analysis stating purity indicate comparable quality.
  • Stated dose: Many hangover blends omit the dihydromyricetin amount (Verster 2021); single-ingredient products with a stated dose allow trial doses to be matched.
  • Stability: A review reports that dihydromyricetin degrades with heat, alkaline conditions and metal ions (Liu 2019); cool, dry storage and avoidance of mixing into alkaline drinks preserve potency.
  • Third-party testing: Seals from independent testers such as USP or NSF (nonprofit standards organizations) verify identity and contaminants; ConsumerLab’s dihydromyricetin article (ConsumerLab DHM answer) is member-only.
  • Named suppliers: Master Herbs, Inc. supplied purified material for University of Southern California studies (NCT05623501); Hepatrat (Uni-Pharma) was the Athens trial product (Michailidou 2026). Naming them reflects trial use, not independent quality testing.

Practical Considerations

  • Time to effect: Glucose changes appeared within 1 month (Ran 2019); liver-enzyme changes appeared at 3 months (Chen 2015) and became clearer between 6 and 12 months (Michailidou 2026).
  • Common pitfalls: Assuming lower blood alcohol, buying blends with undisclosed doses, and treating fruit-fly, worm or mouse lifespan findings as human longevity evidence are the most common mistakes.
  • Regulatory status (United States): Dihydromyricetin is sold as a dietary supplement; the Food and Drug Administration (FDA) does not approve dietary supplements before they are marketed (FDA dietary supplement Q&A).
  • Regulatory status (European Union): An EFSA panel concluded that the safety of a Japanese raisin tree fruit extract proposed as a novel food had not been established (Turck 2020).
  • Cost and access: Dihydromyricetin is inexpensive and widely available without prescription; cost is not a barrier.

Interaction with Foundational Habits

  • Sleep: Potentiating, low certainty: a 20-day trial reported about one hour more self-reported sleep within the dihydromyricetin group (Shao 2022), proposed through GABA-A receptor support. Evening dosing matches that trial’s rationale; alcohol’s own sleep disruption is not offset by any shown effect.
  • Nutrition: Indirect, potentiating: glucose and lipid effects were seen in people with fatty liver or diabetes, so a lower-sugar diet addresses the same targets. Taking it with meals, as in twice-daily trials, and away from iron-rich meals limits iron binding.
  • Exercise: None documented in humans; mouse data suggest preserved muscle through AMPK (Hou 2021). Whether antioxidant activity blunts training adaptations, as reported for some antioxidants, has not been studied.
  • Stress management: Potentiating, low certainty: self-reported stress scores fell in the dihydromyricetin group of one trial (Shao 2022), proposed through calming GABA-A signaling; it complements but has not been compared with behavioral stress techniques.

Monitoring Protocol & Defining Success

Baseline testing before starting covers fasting glucose, HbA1c (glycated hemoglobin, a 3-month average of blood glucose), fasting insulin, ALT, GGT and LDL cholesterol. These record the starting point against which any change is judged and identify people already near low-glucose or abnormal liver ranges. Medication lists are reviewed at the same time for CYP3A4, CYP2D6 and glucose-lowering drugs.

Ongoing monitoring repeats the same panel at 3 months, then every 6–12 months while use continues. People on glucose-lowering drugs add home glucose checks during the first 2–4 weeks. Success is defined as a measurable improvement from the individual’s own baseline in glucose, insulin resistance, liver enzymes or LDL cholesterol without a rise in ALT, together with stable qualitative markers. No change after 3 months suggests limited personal response.

Biomarker Optimal Functional Range Why Measure It? Context/Notes
Fasting glucose 70–99 mg/dL (standard reference range) Expected to change: trials report reductions 8–12 hour fast, morning draw; pair with fasting insulin for HOMA-IR
HbA1c 4.0–5.6% (standard reference range) Expected to change: tracks 3-month glucose No fasting needed; anemia and recent blood loss distort results
HOMA-IR No established target; track change from own baseline Expected to change: fell in Chen 2015 Calculated from fasting glucose and insulin drawn together in the morning
ALT 29–33 U/L or less (men), 19–25 U/L or less (women), American College of Gastroenterology guideline Safety check and expected to change Source: Kwo 2017; members gain no direct revenue from these values; conventional laboratory upper limits are often about 40–55 U/L; avoid heavy exercise and alcohol 24–48 hours before
GGT Below the laboratory upper limit, often about 40–60 U/L (standard reference range) Expected to change: fell in fatty liver trials Raised by alcohol within days; abstain 48 hours before testing; pair with ALT
LDL cholesterol Below 100 mg/dL, National Cholesterol Education Program “optimal” category Expected to change: fell in one trial Source: National Cholesterol Education Program Adult Treatment Panel III report; conventional laboratory reference limits are often below 130 mg/dL; fasting or non-fasting panel

Qualitative markers:

  • Self-rated sleep duration and morning mood
  • Perceived stress level
  • Next-day well-being after drinking, where used around alcohol
  • Digestive comfort and any new drowsiness when combined with other medications

Emerging Research

No recruiting or active dihydromyricetin trial is registered on ClinicalTrials.gov as of 11/07/2026; the registered trials are described with their status.

  • First-in-human dose escalation: NCT05623501, University of Southern California, phase 1, 12 healthy volunteers, purified dihydromyricetin; primary endpoints pharmacokinetics and adverse events. Registered completion 2025 passed; status unknown, no results posted. Good absorption and tolerability would supply the first human dosing basis; poor exposure would deepen absorption concerns.
  • Dihydromyricetin versus metformin in type 2 diabetes: NCT03606694, University of Guadalajara, phase 2, 24 participants, 300 mg twice daily for 12 weeks versus metformin; completion 2023 passed, status unknown, no results posted. A positive result would add independent glucose evidence; a null result would weaken it.
  • Commercial hangover powder: NCT06575972, sponsored by JoyRise, 60 participants, placebo-controlled; completed 2024, no results posted. Positive results would add direct product evidence; null results would weaken hangover claims.
  • Completed and published trials: NCT05052515 (Athens fatty liver trial) and NCT05280561 (sleep and stress) are published; their results appear under Expected Benefits (Michailidou 2026; Shao 2022).
  • Aged-cell targeting: Mouse work showing reduced aged-cell signaling and removal of aged brain immune cells (Xu 2026) points to human studies of aging markers that could strengthen the longevity case.
  • Experimental artifacts: Dihydromyricetin reacts with amino acids in worm culture medium, delaying growth (Wu 2026), which could weaken invertebrate lifespan claims if those findings prove medium-dependent.
  • Alcohol metabolism dispute: Mouse data showing faster alcohol clearance (Silva 2020) conflict with rat data showing none (Skotnicová 2020); the 2026 systematic review rates the evidence inconsistent (Skinner 2026).
  • Topical epigenetic effects: Beiersdorf’s uncontrolled serum study (Qi 2026) leaves open whether a vehicle-controlled trial would confirm skin DNA-methylation age reduction or attribute it to other ingredients.

Conclusion

Dihydromyricetin is a plant compound, most abundant in vine tea, that acts on the brain’s calming signals and on how cells sense their energy supply. For health-focused adults it is mainly encountered in after-drinking products and as a liver and blood-sugar supplement.

The strongest human evidence is modest. Small trials against placebo in people with fatty liver disease or type 2 diabetes point to lower blood sugar and better liver blood tests, but most come from one research group, and the only year-long independent trial tested a combination product and missed its main goal. Effects on cholesterol are inconsistent. Hangover relief rests on extracts of another plant that contains the compound, not on the purified compound, and the sleep, stress and skin-aging findings come from weakly controlled or uncontrolled studies. The case for longer life, including effects on aged cells and lifespan, exists only in cells, worms, flies and rodents, and some worm results may reflect quirks of the laboratory setup.

No side effects attributable to it have been reported in trials, but safety data stop at one year, and laboratory work suggests it may change how the liver handles some medicines and may add to the blood-sugar lowering of diabetes treatment. The evidence base is small and partly shaped by commercial interests: a skin-care manufacturer produced the skin research, supplement makers sponsored hangover and after-drinking studies, and the year-long liver trial tested a branded product. For people with healthy metabolism, whether dihydromyricetin changes long-term health remains unknown.

Top - Benefits - Risks - Protocol