DHM Research Tracker: Key Studies on Dihydromyricetin and Alcohol (Updated 2026)

Most DHM research is preclinical: a 2026 systematic review of 22 studies on DHM and alcohol-related conditions found consistent liver effects in cells and animals but only two small human trials, both using Hovenia dulcis extract rather than DHM itself. As of October 2026, no randomized trial has tested DHM alone for hangover in people. The only published randomized trials that gave people DHM were in people with fatty liver disease or type 2 diabetes, at 300 to 970 mg a day, and didn't involve alcohol.

Key takeaways

  • A 2026 systematic review included 22 studies (8 in vitro, 17 in animals, 2 clinical) and found no clinical trials of isolated DHM for alcohol-related conditions.1
  • The best-known DHM study, from 2012, was in rats given DHM by injection, and it measured intoxication, not hangover.2
  • On alcohol metabolism the animal data conflict: injected DHM sped clearance in mice, while DHM by mouth didn't change it in rats (4 rats per group).3, 4
  • Three randomized trials that gave DHM (alone, in vine tea or with vitamins) enrolled 195 adults with fatty liver disease or type 2 diabetes, at 300 to 970 mg a day for 1 to 12 months.5, 6, 7
  • At least three registered DHM studies had no published results as of October 2026, including the first human study of how DHM is absorbed.8, 9, 10

The state of DHM research in 2026

DHM research is broad in animals and thin in people. Most studies are in rodents or cells, many used injected DHM, and the human trials either tested a different ingredient (Hovenia extract) or studied people with a disease.

The 2026 systematic review by Skinner and colleagues is the best single summary. DHM "consistently attenuated" alcohol-induced cell damage, oxidative stress, inflammation and fatty liver in animal and cell models, but findings on alcohol metabolism and behavior were "inconsistent," and there were no clinical trials of isolated DHM.1 The review searched PubMed only, through December 2025.

The human evidence on DHM and drinking is a few small Korean trials of Hovenia extracts or drinks and one observational study of a multi-ingredient product. Our own PubMed and ClinicalTrials.gov search on October 9, 2026 found no published randomized trial of DHM alone for hangover. For context, start with what DHM is and how DHM is thought to work.

Timeline of key DHM studies, 2012 to 2026

Each row names the model and route, which decide how much a result tells you about a person swallowing a capsule. "Injected" means into the abdomen.

Key DHM and Hovenia studies, 2012 to 2026 (last updated October 9, 2026)
Year Study Model Route Dose Finding Main limitation Link
2012 Shen et al. (UCLA), J Neurosci Rats; rat brain cells Injected (abdomen); mixed into the alcohol solution for the voluntary-drinking test 1 mg/kg Less alcohol intoxication (righting reflex lost about 28 vs. 113 min) and fewer withdrawal signs; blood alcohol not significantly changed. Alcohol and DHM both injected. Measured intoxication, not hangover. PubMed
2014 Liang et al., Neurochem Res Pregnant rats and offspring Not stated in abstract 1 mg/kg, with alcohol Prevented fetal-alcohol deficits in offspring; DHM alone didn't harm litters. One animal study. PubMed
2015 Chen et al., Pharmacol Res 60 adults with fatty liver disease (NAFLD) By mouth 300 or 600 mg/day (wording ambiguous) Lower ALT, AST, GGT, glucose, LDL cholesterol and insulin resistance vs. placebo at 3 months. No alcohol. Its similarity to the same group's resveratrol trial was questioned. PubMed
2017 Liu L et al. (absorption), Pharm Biol Rats By mouth vs. intravenous 20 mg/kg oral; 2 mg/kg IV Oral bioavailability 4.02%; oral half-life 3.7 h. Rats only. PubMed
2017; 2024 Liu L et al., Pharm Biol; Dombi et al., Pharmacol Res Perspect Human liver enzymes Test tube Lab concentrations 2017: inhibited CYP3A4, CYP2E1 and CYP2D6. 2024: no or only minor inhibition of CYP2C9, 2C19 and 3A4. Conflict on CYP3A4. Not tested in people. PubMed; PubMed
2017 Qiu et al., Toxicol Lett Mice on a 6-week alcohol diet By mouth 75 to 150 mg/kg/day Less liver injury, fat buildup and inflammation. Mice. PubMed
2017 Kim H et al., J Ethnopharmacol 26 young Korean men carrying one copy of the ALDH2*2 variant By mouth 2,460 mg Hovenia extract; DHM not reported Blood alcohol and acetaldehyde unchanged; hangover scores fell more than with placebo. Hovenia, not DHM. One genotype, men only. Extract from an ingredient maker. PubMed
2019 Ran et al., Eur J Clin Nutr 80 adults with type 2 diabetes By mouth (vine tea) 970 mg/day Lower fasting glucose and some kidney markers vs. DHM-free vine tea; lipids unchanged. One month. No alcohol. PubMed
2019 Chien et al., J Tradit Complement Med Rats By mouth Herbal mix (36% DHM) up to 1,998 mg/kg/day No signs of toxicity over 28 days. A mixture; only 28 days; company-affiliated authors. PubMed
2020; 2021 Silva et al. (USC), Alcohol Clin Exp Res; Alcohol Mice on chronic alcohol; liver cancer cells Injected 5 to 10 mg/kg Less liver fat; ADH1 and ALDH2 up; blood alcohol and acetaldehyde down; better mitochondrial markers. Injected. Small groups (6 to 10 mice; 3 to 8 per measurement). PubMed; PubMed
2020 Skotnicová et al., Physiol Res Rats; rat liver cells By mouth (stomach tube) 10 mg/kg, with alcohol No change in alcohol clearance or ADH; effect "has not been proven." 2 to 4 rats per group. ALDH not measured. PubMed
2020; 2023 Deng et al., Xenobiotica; Liu H et al., Drug Des Devel Ther Rats Not stated in abstracts 100 mg/kg (and 500 mg/kg for 14 doses) Raised exposure to the drug triptolide by 98% after one dose; didn't change fluconazole levels. Animal drug-interaction studies. PubMed; PubMed
2021 Carry et al., Int J Mol Sci Mice; frog egg cells Injected vs. by mouth 50 mg/kg Injected DHM briefly shortened intoxication; oral exposure 7 to 24 times lower. Mixed receptor findings. PubMed
2021 Ye et al., AAPS PharmSciTech Rats and mice By mouth Not stated in abstract A micelle form roughly doubled absorption vs. plain DHM. Special formulation, not plain DHM. PubMed
2023 Janilkarn-Urena et al., Front Nutr Female mice on an alcohol diet In the liquid diet 6 mg/mL of liquid diet Better liver enzymes, blood fats and inflammatory markers. Mice. Proof of concept. PubMed
2024 Niiya et al., J Ethnopharmacol Rats, mice, cells By mouth (Hovenia) Hovenia extract Extract lowered blood alcohol in rats; in cells, most of its enzyme-inducing activity came from compounds other than DHM. Hovenia. Two authors are company employees. PubMed
2024 Lee KW et al.; Paik et al., Foods 25 and 24 adults By mouth Drinks with 0.475% to 0.731% Hovenia extract, some with kudzu or glutathione-yeast Small blood-alcohol differences. One trial: only the stomach-symptom subscore improved, acetaldehyde higher at 6 h. The other: hangover scores no different from placebo. Same sponsor; in each trial, 6 of 13 authors were employed by the beverage maker. PubMed; PubMed
2026 Michailidou et al., Ann Gastroenterol 55 adults with fatty liver disease (MASLD) By mouth 300 mg/day plus vitamins C and E and choline ALT and GGT both normalized in 35% vs. 5% at 12 months. Multi-ingredient. 9 dropouts (2 vs. 7). PubMed
2026 Park WL et al., Food Funct Rats; 30 adults By mouth Citrus vinegar plus Hovenia extract and glucose Less blood alcohol and acetaldehyde than placebo. Combination product. Small. Erratum published. PubMed
2026 Skinner et al. (systematic review), Nutrients 22 studies Mixed Mixed Consistent preclinical liver effects; inconsistent effects on alcohol metabolism; no clinical trials of isolated DHM. Unregistered. No formal risk-of-bias tool. PubMed
2026 Song et al., Nutrients 90 adults; 2,958 morning surveys By mouth Multi-ingredient DHM and L-cysteine product Small self-rated next-day improvements (effect sizes 0.20 to 0.32). Observational: not randomized, blinded or placebo-controlled. Funded by the maker. PubMed

What have human studies of DHM tested and found?

Human studies fall into three groups, and none is a randomized trial of DHM alone for hangover.

  • Hovenia trials (2017 to 2026). All Korean, small, short and mostly industry-linked, with mixed results.11, 12, 13, 14 None measured DHM in the product or the blood, and Hovenia extract is less than 1% DHM.1
  • DHM trials in people with a disease (2015 to 2026). Three randomized trials in fatty liver disease or type 2 diabetes reported improvements in some lab markers.5, 6, 7 None involved alcohol.
  • An observational study of a DHM product (2026). Small self-rated improvements, with no randomization or placebo. Its authors say a randomized trial is needed "to establish causality."15

Registered studies without published results: a University of Southern California Phase 1 absorption and safety study of purified DHM in 12 healthy volunteers (last listed as not yet recruiting);8 a 288-person trial of DHM for stress-related sleep problems (completed 2022);9 and a randomized, placebo-controlled trial of a DHM-containing powder (completed August 2024).10

What do animal and cell studies suggest?

They suggest DHM is worth studying in people, especially for alcohol-related liver injury, not that it works in people. We walk through the best-known experiments, test by test, in the DHM rat studies.

  • Liver: the most consistent signal. DHM reduced fat buildup, liver enzymes and inflammation in alcohol-fed mice, by mouth, in food or injected.16, 3
  • Intoxication: seen with injected DHM in rats and mice, and short-lived.2, 17
  • Alcohol metabolism: faster clearance with injected DHM in mice; no change with oral DHM in rats.3, 4
  • Absorption: about 4% by mouth in rats.18

FTC guidance says animal and test-tube studies "aren't sufficient to substantiate health-related claims."19 We read them the same way.

How should you read industry-funded and observational studies?

Methods first, funder second. Company money doesn't make a study wrong, but much of the human research near DHM is company-linked.

  • In both 2024 Hovenia trials, 6 of the 13 authors worked for the beverage maker, which sponsored both.12, 13
  • The 2017 Hovenia trial's extract came from the maker of a commercial Hovenia ingredient, according to EU regulators.20
  • The 2026 observational study was funded by the product's maker.15
  • Since January 2025, South Korea has required human data for "hangover relief" claims on foods, which helps explain the wave of small Korean beverage trials.21

Observational designs add problems: people choose when to take the product, know what they took and expect to feel better.

Two more cautions. Some DHM "guide" websites misdescribe studies, for example presenting a 2024 Hovenia drink trial, where only the stomach-symptom subscore improved, as a DHM trial with large drops in hangover severity.12 And we have a conflict of our own: Fuller Health sells a DHM supplement. That's why this page lists the negative and null results alongside the positive ones. For how this research applies to products on shelves, see do hangover pills work? and how to judge hangover supplements honestly.

What we don't know yet: limits of dose, absorption and study quality

We don't know whether DHM does anything for people who drink, at what dose, or whether it's safe at the amounts sold today. Three limits explain why.

  • Route and absorption. Injected DHM skips the gut; in mice, DHM by mouth gave 7 to 24 times lower blood exposure.17 No human absorption data had been published as of October 2026.
  • Dose. Human trials used 300 to 970 mg a day; some products sold today contain far more.
  • Quality. Animal groups as small as 2 to 4, human trials of 24 to 80 people, published questions about one trial, and no searches of Chinese- or Korean-language databases, by us or by the 2026 review.22

Our own product is a case in point. After Party by Fuller Health is a single-ingredient DHM supplement: 650 mg of dihydromyricetin per capsule, 3,250 mg per 5-capsule serving, with nothing else but the cellulose capsule. After Party's 3,250 mg serving is higher than the doses used in published human DHM studies. Those studies gave up to about 970 mg a day, to people with fatty liver disease or type 2 diabetes, not to people who drink.6, 5, 7 Safety at 3,250 mg hasn't been studied. No human data show that a higher dose does more. See DHM dosage and timing and the After Party facts page.

In test-tube studies with human liver enzymes, DHM inhibited CYP3A4, CYP2D6 and CYP2E1, enzymes that help clear many medications.23 That hasn't been tested in people, so ask a pharmacist or doctor first if you take prescription medicines. DHM hasn't been studied in pregnancy or breastfeeding; don't use it then.

What would a good human DHM trial look like?

Randomized, double-blind and placebo-controlled, testing DHM alone in healthy adults who drink, with outcomes registered in advance. In our view, it should also:

  • verify the DHM content of the product by independent analysis;
  • give a controlled amount of alcohol and measure blood alcohol, acetaldehyde and blood DHM;
  • use a validated hangover scale as the main outcome, with next-day safety outcomes such as reaction time;
  • compare more than one dose, including the higher amounts sold today, such as 3,250 mg, to test both dose-response and safety;
  • record adverse events, disclose funding and publish whatever the result.

The FTC calls randomized controlled human trials "the most reliable form of evidence."19

How we keep this page updated (changelog)

Last updated October 9, 2026. We plan to update this page as significant new studies appear.

  • October 9, 2026: page published, covering PubMed and ClinicalTrials.gov through that date, including the four 2026 papers in the table.

Method: we search PubMed and ClinicalTrials.gov for dihydromyricetin, ampelopsin, vine tea and Hovenia with alcohol, hangover and clinical-trial terms, and read each abstract, plus the full text where available. Know of a study we missed? Tell us.

Frequently asked questions

Is there human research on DHM?

Yes, but not much, and not on hangovers. Three small randomized trials gave DHM to people with fatty liver disease or type 2 diabetes at 300 to 970 mg a day, and some lab markers improved; none involved alcohol.5, 6, 7

Has DHM been tested on hangovers in people?

Not on its own. As of October 2026, no randomized trial of DHM alone for hangover had been published in a PubMed-indexed journal; the human hangover trials tested Hovenia extracts or multi-ingredient products.1, 15

Who funds DHM research?

A mix. Much of the animal work came from university labs, while several human Hovenia trials and the 2026 observational study were run or funded by companies selling the products.12, 15 Funding doesn't make a study wrong, but it's a reason to read the methods. Fuller Health sells DHM too.

What did the 2012 rat study find?

In a 2012 UCLA study, DHM injected into the abdomen of rats at 1 mg/kg reduced signs of alcohol intoxication and withdrawal without significantly changing blood alcohol.2 The researchers linked the effect to GABA-A receptors. It didn't study hangover or people.

Is DHM FDA approved?

No. NIH's LiverTox notes that DHM isn't approved for any use or disease in humans in the United States.28 FDA says it doesn't have the authority to approve dietary supplements before they're sold.29 And a supplement claiming to cure or prevent hangovers is making a drug claim; FDA warned seven companies about that in July 2020.30

Sources

  1. Skinner SG, et al. Therapeutic Effects of Dihydromyricetin on Wholly Alcohol-Attributed Conditions: A Systematic Review. Nutrients. 2026;18(14):2221. PubMed
  2. Shen Y, et al. Dihydromyricetin as a novel anti-alcohol intoxication medication. J Neurosci. 2012;32(1):390-401. PubMed
  3. Silva J, et al. Dihydromyricetin Protects the Liver via Changes in Lipid Metabolism and Enhanced Ethanol Metabolism. Alcohol Clin Exp Res. 2020;44(5):1046-1060. PubMed
  4. Skotnicová A, et al. Does dihydromyricetin impact on alcohol metabolism. Physiol Res. 2020;69(Suppl 4):S573-S581. PubMed
  5. Chen S, et al. Dihydromyricetin improves glucose and lipid metabolism and exerts anti-inflammatory effects in nonalcoholic fatty liver disease: A randomized controlled trial. Pharmacol Res. 2015;99:74-81. PubMed
  6. Ran L, et al. Ampelopsis grossedentata supplementation effectively ameliorates the glycemic control in patients with type 2 diabetes mellitus. Eur J Clin Nutr. 2019;73(5):776-782. PubMed
  7. Michailidou E, et al. Dietary supplement based on dihydromyricetin in metabolic dysfunction-associated steatotic liver disease: a double-blind, placebo-controlled, randomized clinical trial. Ann Gastroenterol. 2026;39(1):54-62. PubMed
  8. University of Southern California. Phase I, Dose-Escalation Study of Dihydromyricetin (DHM) to Treat Alcohol-Associated Liver Disease (NCT05623501). ClinicalTrials.gov. Status unknown; no results posted as of October 9, 2026. ClinicalTrials.gov
  9. University of Southern California. Stress-, Anxiety-, and Cellphone Use-induced Sleep Deficits and a Potential Complementary Therapy (NCT05280561). ClinicalTrials.gov. Completed in 2022; no publication found as of October 9, 2026. ClinicalTrials.gov
  10. ClinicalTrials.gov record NCT06575972: randomized, placebo-controlled trial of a DHM-containing recovery powder on alcohol aftereffects. Completed August 2024; no results posted as of October 9, 2026. ClinicalTrials.gov
  11. Kim H, et al. A standardized extract of the fruit of Hovenia dulcis alleviated alcohol-induced hangover in healthy subjects with heterozygous ALDH2: A randomized, controlled, crossover trial. J Ethnopharmacol. 2017;209:167-174. PubMed
  12. Lee KW, et al. Clinical Evaluation of Hovenia dulcis Extract Combinations for Effective Hangover Relief in Humans. Foods. 2024;13(24):4021. PubMed
  13. Paik DH, et al. Efficacy of Hovenia dulcis Fruit Extract in Hangover Mitigation: Double-Blind Randomized Clinical Evaluation. Foods. 2024;13(24):4084. PubMed
  14. Park WL, et al. Citrus unshiu fermented vinegar and Hovenia dulcis fruit extract formulation enhances alcohol metabolism and attenuates alcohol-induced hepatic injury. Food Funct. 2026;17(9):4033-4050. PubMed
  15. Song S, et al. Decentralized Assessment of a Dihydromyricetin- and L-Cysteine-Based Multi-Ingredient Dietary Supplement on Post-Alcohol Recovery: An Observational Cohort Study. Nutrients. 2026;18(15):2424. PubMed
  16. Qiu P, et al. Dihydromyricetin modulates p62 and autophagy crosstalk with the Keap-1/Nrf2 pathway to alleviate ethanol-induced hepatic injury. Toxicol Lett. 2017;274:31-41. PubMed
  17. Carry E, et al. Identification of Dihydromyricetin and Metabolites in Serum and Brain Associated with Acute Anti-Ethanol Intoxicating Effects in Mice. Int J Mol Sci. 2021;22(14):7460. PubMed
  18. Liu L, et al. Determination of dihydromyricetin in rat plasma by LC-MS/MS and its application to a pharmacokinetic study. Pharm Biol. 2017;55(1):657-662. PubMed
  19. Federal Trade Commission. Health Products Compliance Guidance. December 2022. FTC
  20. Turck D, et al. Safety of hot water extract of fruits and peduncles of Hovenia dulcis as a novel food pursuant to Regulation 1(EU) 2015/2283. EFSA J. 2020;18(8):e06196. PubMed
  21. NutraIngredients. Hangover relief products must be backed by human tests: South Korea's MFDS. January 21, 2025. NutraIngredients
  22. Rotman Y. Similarity between studies of dihydromyricetin and reservatrol [sic] for NAFLD. Pharmacol Res. 2015;100:335. PubMed
  23. Liu L, et al. In vitro inhibitory effects of dihydromyricetin on human liver cytochrome P450 enzymes. Pharm Biol. 2017;55(1):1868-1874. PubMed
  24. U.S. Department of Agriculture; U.S. Department of Health and Human Services. Dietary Guidelines for Americans, 2025-2030. Released January 7, 2026. PDF
  25. National Institute on Alcohol Abuse and Alcoholism. Hangovers. Updated December 2025. NIAAA
  26. Substance Abuse and Mental Health Services Administration. SAMHSA's National Helpline. SAMHSA
  27. MedlinePlus, U.S. National Library of Medicine. Alcohol withdrawal. Reviewed January 1, 2025. MedlinePlus
  28. National Institute of Diabetes and Digestive and Kidney Diseases. LiverTox: Dihydromyricetin. NCBI Bookshelf. Last updated August 16, 2023. NCBI Bookshelf
  29. U.S. Food and Drug Administration. Dietary Supplements: Report Adverse Events to FDA (consumer fact sheet). May 2022. FDA (PDF)
  30. U.S. Food and Drug Administration. FDA Sends Warning Letters to Seven Companies Illegally Selling Hangover Products. Constituent update, July 29, 2020. FDA

Published by Fuller Health, which makes After Party. This article is general information, not medical advice: talk to a doctor or pharmacist about your own health, medicines and drinking. Last reviewed October 9, 2026. How we research and write.

Back to blog