Herbs & Botanicals · Traditional Chinese lung herb

Houttuynia

Q: Does houttuynia help with COVID-19 or colds?

Not shown for houttuynia alone. A meta-analysis of a many-herb formula that includes it found better outcomes when added to standard care, but its authors say the studies were low quality. Lab antiviral effects have not been shown in people.

Q: Is the injection safe?

No. A review of 1,232 reports found that 22.56% of reactions were serious body-wide reactions, mostly anaphylaxis, with deaths. Risk was higher with intravenous use and with macrolide antibiotics.

Q: Was the injection banned in China?

Its hospital use was halted in June 2006 after severe reactions in four children, two fatal. A review says a limited number of manufacturers were later allowed to resume. We did not check its current legal status.

Q: Is tea or food use as risky as the injection?

The injection reactions are not reported for tea or food. We found no case report of anaphylaxis from eating or drinking it, but that is not proof it cannot happen. Reviews say long-term vegetable use has no reported toxicity but ask for better data.

Q: Does it contain aristolochic acid and can it harm the kidneys?

Aristolochic acid I and II were not detected in two chemistry studies. Related aristolactams were found, and they were toxic to kidney cells in a dish. In mice they were much milder than aristolochic acid I. No human kidney harm from houttuynia has been shown, and no human kidney study exists.

Q: Is a tea safer than an extract?

One analysis found no aristolactams in the water extract of any plant part, while the aerial part and flowers held the most. In rats, a strong alcohol extract killed animals at higher doses. Human safety is not tested for either.

Q: Can it harm the liver?

No human case report was found and no LiverTox entry exists. In rats, a strong alcohol extract caused liver and kidney toxicity at higher doses over 28 days.

Q: Is it safe in pregnancy or breastfeeding?

No human data exist, and no LactMed entry was found. Lab embryo studies reported effects on development. Avoid medicinal amounts unless a clinician says otherwise.

Q: Does it interact with medicines?

Evidence is from animals and labs only. An extract raised metformin levels in rats. An isolated compound blocked platelet COX enzymes in mouse blood, so a blood-thinner interaction cannot be ruled out. For the injection, macrolide antibiotics raised reaction risk. Ask a pharmacist.

Q: Can it cause stomach upset?

We found no data on stomach upset from the oral herb. Digestive symptoms were involved in about one-quarter of injection reaction reports.

Houttuynia cordata · also called Fish mint, fish herb, yu xing cao

Houttuynia is a fish-smelling leafy herb eaten in parts of Asia and used in Chinese medicine for colds, chest infections and skin problems. Most of the research is in cells and animals. A few small trials tested multi-herb products, so houttuynia cannot get the credit. A drug-type injection made from it caused serious allergic reactions, including deaths. Talk to a doctor or pharmacist before using strong extracts, especially in pregnancy or with medicines.

General information, not medical advice. Last substantive update October 25, 2026 · Not medically reviewed. A named clinician has not signed off on this page. How this page is made

About Houttuynia

Houttuynia is a fish-smelling herb used in Chinese medicine and as a vegetable. Human evidence is weak and mostly for multi-herb products. The important warnings are about the injection form (fatal allergic reactions), strong extracts in rats, and the aristolactam kidney question, which is unproven in people.

What it is

Houttuynia cordata is a creeping plant in the lizard's-tail family (Saururaceae). Its fishy smell gives it the names fish mint and yu xing cao ('fish-smell herb'). GBIF lists the name as an accepted species, and NCBI Taxonomy holds the same name.

People use the leaves and stems (aerial parts) and the rhizome. Keep four products apart: the fresh herb or tea, strong extracts, sodium houttuyfonate (a drug-type derivative), and the injection. Their benefits and risks are not the same.

^{4 10 16}

What is in it

The oil compound houttuynin (decanoyl acetaldehyde) gives the smell and is reported as antibacterial. It is unstable and turns into 2-undecanone when heated. The plant also holds flavonoids such as quercitrin, hyperoside and rutin, and alkaloids called aristolactams.

Sodium houttuyfonate is made from the oil compound and used as a drug in China. Nobody knows which compound explains any benefit in people.

^{16 8 11}

What the research shows

We found no trial of plain houttuynia for colds, flu or COVID-19. A meta-analysis of seven studies of Lianhua Qingwen (a formula with about a dozen herbs including houttuynia) found better outcomes when added to standard care, but its authors say the studies were low quality and no clear conclusion can be drawn. One small trial of a houttuynia-based formula eased allergic nose symptoms versus regular medicine, with no numbers in the abstract.

For skin, one 60-person trial of a four-extract blend reduced acne lesion counts. A sugar fraction helped eczema-like skin in mice. Evidence for blood sugar, infection and cancer is from cells and animals.

^{7 2 20 5}

At a glance

Traditional uses history, not proof

Colds, fever, cough and pneumonia in Chinese medicine
Lung abscess and pus-forming swellings ('reducing swelling, discharging pus')
Fresh leaves and rhizome eaten as a vegetable in parts of South China and India

Studied for research

COVID-19: only a many-herb formula studied, low-quality trials
Allergic nose symptoms: one small formula trial
Acne: one trial of a four-extract skin blend
Eczema, blood sugar, infection and cancer: cells and animals only
Injection for respiratory infection: harms documented, benefit not shown

Main compounds

Houttuynin (decanoyl acetaldehyde, PubChem CID 122640)
Sodium houttuyfonate (CID 23663544)
Quercitrin (CID 5280459)
Hyperoside (CID 5281643)
Aristolactams, for example aristolactam I (CID 96710)

Common forms

Fresh leaves or tea (food use)
Dried herb (Houttuyniae herba)
Alcohol extract, capsule or essential oil
Injection (a hospital drug, not a supplement)
Skin products, often blends

Research links: Wang 2010: adverse events to Houttuynia injection (PubMed) · Ji 2009: anaphylaxis from nine herbal injections (PubMed) · Xue 2025: aristolactam safety evaluation in mice (PubMed) · Li 2022: Lianhua Qingwen meta-analysis for COVID-19 (PubMed) · Wu 2021: Houttuynia cordata ethnopharmacological review (PubMed) · Research Library · Genes & Biomarkers Library

The plant and its history

Accepted scientific name: Houttuynia cordata Thunb.
Naming authority: Thunb.
Common names: Fish mint (English); Fish herb (English); Yu xing cao (Houttuyniae herba) (Chinese (pinyin)); Dokudami (Japanese)
Botanical family: Saururaceae (lizard's-tail family)
Classification: Plantae > Tracheophyta > Magnoliopsida > Piperales > Saururaceae > Houttuynia > Houttuynia cordata
Native range: Not verified in the sources read. Reviews describe it as eaten as a vegetable in parts of South China and India.
Habitat: Not found in the sources read.
What the plant looks like: Roots: Rhizome (underground stem) is used and eaten; no description found.; Stems: Not described in sources read.; Leaves: Aerial parts (above-ground, mainly leaf and stem) are the usual medicinal part; no description of shape found.; Flowers: Flowers held the most aristolactams in one analysis; no description found.; Fruit: Not described in sources read.
Cultivation: Not found.
Harvest and processing: Not found. Houttuynin is unstable and turns into 2-undecanone when heated, so cooking and drying may change the aroma and compounds.
Look-alikes and adulterants: Name: Houttuynia injection, sodium houttuyfonate and the essential oil; Note: Different products from the same plant. The injection is a drug with documented anaphylaxis. Sodium houttuyfonate is a drug-type derivative. Do not assume tea, extract, oil and injection share the same risks or benefits.; Name: Other plants with aristolactams; Note: Aristolactams occur in several plants. Product labels should name the species and part.
Sustainability and conservation: No conservation assessment was found.

Traditional use and history history, not clinical proof

Tradition or era	Documented use	Source
Traditional Chinese medicine (more than 1,500 years)	Used for 'clearing heat', eliminating 'toxins', reducing swelling and discharging pus; for colds, fever, pneumonia and lung abscess.	p36558101
Food use, South China and India	Leaves and rhizomes eaten as a vegetable; reviews report no long-term toxicity reports from this use and ask for better data.	p34539401
2006, China	Houttuynia (Yuxingcao) injection halted in hospitals after severe reactions in four children (two fatal); a limited number of manufacturers later allowed to resume.	p19559066
2013, China	Listed as a plant that can be used as both food and medicine by the National Health Commission (as reported by a review; the list itself was not read).	p34539401
EMA / HMPC	No herbal monograph found; EMA could not be searched properly. Any Wikipedia-style history claims (dokudami in Japan, diep ca in Vietnam) are NOT verified here.	s-regulators-not-found-houttuynia

Traditional use shows how a plant was used. It does not show that a supplement works or is safe today.

What is inside, and how it may work

What the plant contains and how those compounds act in cells and animals. Laboratory and animal mechanisms are not proven effects in people.

Main compounds

Compound	Family	Where it is concentrated	Activity studied	Level of evidence
Houttuynin (decanoyl acetaldehyde; 3-oxododecanal) PubChem 122640	Volatile oil compound	Aerial parts (essential oil)	Source of the fishy smell and reported as antibacterial. Easily turns into 2-undecanone (CID 8163) at high temperature. Unstable in blood plasma. ^{16 8 11}	Laboratory research
Sodium houttuyfonate PubChem 23663544	Sulfonate derivative of the oil compound	Not a plant compound; made from houttuynin (PubChem structure name: sodium 1-hydroxy-3-oxododecane-1-sulfonate)	Reviews report antibacterial and quorum-sensing effects in the lab. It is a drug-type derivative, not tea or food. ^{16 11}	Laboratory research
Quercitrin PubChem 5280459	Flavonoid glycoside	Aerial parts	Reported in the plant; a computer-docking SARS-CoV-2 report is called idealized by one review. ^{16 11}	Laboratory research
Hyperoside and rutin PubChem 5281643	Flavonoid glycosides	Aerial parts	Reported marker flavonoids; rutin CID 5280805 is not looked up in this file. ^{16 11}	Lab analysis
Aristolactams (AI, AII, BII, FI) PubChem 96710	Phenanthrene lactam alkaloids	Aerial parts and flowers more than underground parts 3.20 to 108.19 ug/g aerial; 0.95 to 11.66 ug/g underground (30 batches)	Relatives of aristolochic acids. Toxic to kidney cells in a dish; much less kidney-toxic than aristolochic acid I in mice. AII CID 148657 (name match). Papers disagree on which are present (AI not detected in 22 samples in one paper). ^{18 19 22 17 11}	Animal research
Aristolochic acid I (comparison, not found in the plant) PubChem 2236	Nitrophenanthrene carboxylic acid	Not detected in 22 samples or in 75 dried herb batches	The kidney toxin used as the positive control in the mouse study. ^{19 22 17 11}	Animal research
Becatamide	Amide	Found in Houttuynia cordata	Blocked COX-1 and COX-2 and lowered platelet activation markers in mouse blood. No CID looked up (name not matched). ¹²	Laboratory research
Polysaccharides	Plant sugars	Not stated	A fraction improved eczema-like skin in mice. ⁵	Animal research

Mechanisms, receptors and pathways

Target or pathway	Gene or receptor	What was found	Level
IL-6 release	IL-6	A houttuynia-based formula lowered IL-6 secretion in LPS-stimulated cells. ²	Laboratory research
Mast cell receptor (injection reactions)	Mrgprb2 (mouse), MRGPRX2 (human version)	Mice lacking the gene had less mast cell degranulation, histamine release and paw swelling after the injection. ¹⁴	Animal research
Platelet COX pathway	COX-1, COX-2	Becatamide blocked both enzymes and lowered thromboxane B2 and P-selectin in mouse blood. ¹²	Laboratory research
Drug transporters	OCT2, MATE1 (human transporters in cells); Oct1/Oct2 in rats	The ethanol extract blocked metformin transport in cells and raised metformin exposure in rats. ²¹	Animal research
Drug-metabolizing enzyme	CYP2E1	A review reports lower CYP2E1 activity in mice given a water and ethanol extract at 300 mg/kg/day for 7 days. ¹⁶	Animal research
Embryo development signalling	PI3K-Akt, AKT1	Lower AKT1 and PI3K in embryo cell clusters; computer analysis pointed to oleanolic acid, a lignan and aristolactam AII. ⁹	Laboratory research
Kidney cell stress and fibrosis	TGF-beta1, fibronectin	Three aristolactams raised ROS, TGF-beta1 and fibronectin in human kidney tubule cells. ¹⁸	Laboratory research
Kidney injury in mice	Complement C3a/C3aR, C5a/C5aR (for aristolochic acid I)	Aristolochic acid I raised kidney markers much more than aristolactams, through oxidative stress and complement activation. ¹⁹	Animal research
Skin inflammation	MAPK pathway; IL-1 alpha, IL-8, keratin 16	A polysaccharide fraction lowered inflammatory messengers in mice and skin cells; the blend lowered IL-1 alpha, IL-8 and keratin 16 in human skin biopsies. ^{5 20}	Human (blend) and animal

Absorption, metabolism and elimination

Absorption: No human data. In mice, oral bioavailability was 52.71% for aristolactam I and 39.60% for aristolactam BII, versus 99.83% for aristolochic acid I.
Distribution: Not found.
Metabolism: Aristolactams were broken down into less toxic products in mice; no unchanged aristolactam appeared in urine. Houttuynin is unstable in blood plasma.
Elimination: Not found for people.
Bioavailability: See absorption. For houttuynin, a method to measure it in rat blood was built; values are not in the abstract.

^{19 8 21}

Isolated compounds versus whole extracts

Houttuynia is sold and studied in very different forms, and risks do not carry across. The injection is an extract-based drug with documented anaphylaxis. Sodium houttuyfonate is a derivative of the oil compound. Strong ethanol extracts caused liver and kidney toxicity in rats. Aristolactams were not detected in the water extract of any plant part in one analysis, while the aerial part and flowers held the most in the plant itself. Tea, food and each extract should not be treated as equivalent.

Nutritional composition

Houttuynia is eaten as a leafy and rhizome vegetable in parts of China and India. This file does not report nutrient values.

11benefits and uses looked at

22studies and sources checked

Very low certaintybest evidence rating found

6lab markers tracked in trials

3 very low certainty8 not enough evidence

What is Houttuynia good for?

The studied uses and how strong the human evidence is. For houttuynia it is very weak: the few human results are for multi-herb products, and most findings are from cells and animals.

Benefits, one by one

Each card is a research question, not a promise. Tap a goal to filter. Open a card to see the studies, doses and limits.

COVID-19: formulas that contain houttuynia (not houttuynia alone)

Very low certainty

A meta-analysis of seven studies of Lianhua Qingwen, a formula with about a dozen herbs including houttuynia, found better results when it was added to standard care. The authors say study quality was low. No trial tested houttuynia by itself for COVID-19.

1 studiesFormula capsules or granules combined with standard medicine. No hout…

Studies, doses and limits

What the trials found

One updated meta-analysis (7 studies, 916 people) of the formula plus western medicine versus western medicine alone. The same paper lists Houttuynia cordata aerial part among the formula's ingredients.

Results

Better total efficacy, fewer adverse events, better chest CT findings, fewer people becoming severe, and shorter fever with the formula.

Effect sizes as reported

Total efficacy OR 2.23 (95% CI 1.56 to 3.18); adverse events OR 0.42 (0.18 to 0.97); chest CT OR 1.74 (1.12 to 2.72); conversion to severe OR 0.47 (0.30 to 0.75). Authors: low methodological quality, no clear conclusions.

Why this rating

The effect belongs to a many-herb formula, the studies were rushed and low quality by the authors' own words, and houttuynia cannot be separated out.

Form and dose studied

Formula capsules or granules combined with standard medicine. No houttuynia-only product was tested.

Who was studied

People with COVID-19, mostly mild to moderate.

Limits

A formula result is not a houttuynia result. Pandemic-era trials were often small, open-label or poorly reported.

Why researchers looked

Houttuynia is one ingredient of Lianhua Qingwen, a Chinese formula that was widely used during the pandemic.

Traditional use

Used in Chinese medicine for fever, lung infections and abscesses.

Proposed mechanism Preliminary mechanism

Not shown for this formula in people. In the lab, houttuynia extracts slowed some viruses (see the lab antiviral entry).

A proposed mechanism does not show that the effect happens in people.

No adverse-event rise for the formula was reported in the pooled data; this says nothing about houttuynia alone or the injection.

⁷

Year-round allergic nose symptoms: a houttuynia-based formula

Very low certainty

In a small Taiwan trial, adults with long-standing allergic rhinitis who took a houttuynia-based formula for 4 weeks had better nasal symptoms than those on regular medicine. The abstract gives no numbers.

1 studiesA houttuynia-based formula (other ingredients not listed in the abstr…

Studies, doses and limits

What the trials found

One randomized trial, 4 weeks, formula versus regular western medicine (no placebo).

Results

Improved nasal obstruction, nasal secretion, loss of smell and postnasal drip compared with the control group.

Effect sizes as reported

No numbers reported in the abstract.

Why this rating

One small trial of a multi-herb formula, no numbers, no placebo, blinding not stated; houttuynia cannot be separated out.

Form and dose studied

A houttuynia-based formula (other ingredients not listed in the abstract) for 4 weeks.

Who was studied

Adults 20 to 60 with at least 2 years of moderate to severe year-round allergic rhinitis.

Limits

Not a COVID trial. Single trial, no replication found.

Why researchers looked

Allergic nose symptoms can look like COVID-19 symptoms; the researchers wanted a herbal option to test during the pandemic.

Traditional use

Houttuynia is used traditionally to reduce nasal and chest congestion.

Proposed mechanism Preliminary mechanism

In a cell model the formula reduced the release of IL-6, an inflammation messenger.

A proposed mechanism does not show that the effect happens in people.

The abstract does not report adverse effects.

Houttuynia injection for respiratory infections (benefit not established; serious harm reported)

Not enough evidence

In China a drug-type houttuynia injection was widely given for chest and respiratory infections. The review we read measured its harms, not whether it works, and found serious allergic reactions and deaths. This is not the same as tea or food use.

2 studiesInjection only (hospital drug). Not an oral supplement.

Studies, doses and limits

What the trials found

A systematic review of 1,232 reported reactions across 645 articles. It was not designed to test whether the injection treats infection.

Results

Reaction reports only; no efficacy trial was read.

Effect sizes as reported

Serious systemic reactions were 22.56% of reported reactions; anaphylactic shock made up 58.99% of the most severe grade.

Why this rating

No efficacy trial was read, and the safety record is poor. See the safety tab.

Form and dose studied

Injection only (hospital drug). Not an oral supplement.

Who was studied

People treated with the injection in China.

Limits

Case-report data cannot give how often reactions happen.

Why researchers looked

Respiratory disease was the most common reason the injection had been given in reported reaction cases (52.44% of cases).

Traditional use

Developed from the traditional herb as a modern injectable drug.

Proposed mechanism Animal research

Not a benefit claim. A mouse study links injection reactions to a mast cell receptor (Mrgprb2).

A proposed mechanism does not show that the effect happens in people.

This entry is mainly a safety warning. Reactions were most often with intravenous use and with macrolide antibiotics.

^{15 6}

Colds, cough, pneumonia and lung abscess (traditional use)

Not enough evidence

Chinese medicine has long used houttuynia for fever, colds, lung infections and lung abscess. The reviews we read cite no good human trial of the plain herb for these uses.

2 studiesTea, dried herb or fresh herb in tradition; amounts not studied here.

Studies, doses and limits

What the trials found

Two narrative reviews describe traditional use and lab findings, but no randomized trial of the herb alone for colds or pneumonia was found.

Results

No human result.

Effect sizes as reported

No numbers.

Why this rating

Support is tradition plus lab and animal work only.

Form and dose studied

Tea, dried herb or fresh herb in tradition; amounts not studied here.

Who was studied

Not applicable.

Limits

Tradition is not proof that it works.

Why researchers looked

Traditional first-line use for 'heat' and 'toxin' conditions of the lungs.

Traditional use

Used in Chinese medicine for more than 1,500 years; the fresh herb is also eaten as a vegetable.

Proposed mechanism Preliminary mechanism

Not established in people.

A proposed mechanism does not show that the effect happens in people.

See the safety tab for the injection and for aristolactam questions.

^{16 13}

Antiviral activity (lab and computer work only)

Not enough evidence

Houttuynia extracts and some of its compounds slowed several viruses in cells and in mice. A computer-docking report on SARS-CoV-2 is called 'idealized' by one review. No human trial tests this.

2 studiesExtracts in lab studies; not a product people take.

Studies, doses and limits

What the trials found

Narrative review of cell, mouse and computer studies; no human data.

Results

Lab and computer findings only.

Effect sizes as reported

A mouse coronavirus was slowed in cells (IC50 0.98 mg/ml for an ethyl acetate extract), as reported in the review.

Why this rating

No human trial; computer docking is not evidence of benefit.

Form and dose studied

Extracts in lab studies; not a product people take.

Who was studied

Cells, mice, computer models.

Limits

Lab effects often fail in people.

Why researchers looked

Interest rose during the SARS and COVID-19 outbreaks.

Traditional use

Used for fever and 'toxin' conditions.

Proposed mechanism Laboratory research

Lab studies report slowing of a mouse coronavirus and other viruses; the review says human tolerance has not been properly studied.

A proposed mechanism does not show that the effect happens in people.

None established; see safety tab.

^{16 13}

Acne: a four-extract skin product that includes houttuynia

Very low certainty

In one 60-person trial, a skin product containing four plant extracts, houttuynia among them, reduced acne spots and skin inflammation markers over 8 weeks. It cannot be credited to houttuynia alone.

2 studiesTopical extract blend (mangosteen, Lithospermum officinale, Tribulus …

Studies, doses and limits

What the trials found

One double-blind randomized trial (60 people) of the blend versus a vehicle.

Results

Inflammatory and non-inflammatory lesion counts fell in the extract group (p<.05); skin biopsies showed lower IL-1 alpha, IL-8 and keratin 16 than vehicle; no serious adverse events.

Effect sizes as reported

No between-group lesion numbers are given in the abstract.

Why this rating

One small trial of a four-extract blend; houttuynia cannot be separated out; funding not stated.

Form and dose studied

Topical extract blend (mangosteen, Lithospermum officinale, Tribulus terrestris, Houttuynia cordata) for 8 weeks.

Who was studied

60 adults with mild to moderate acne in Korea.

Limits

Combination product; no houttuynia-only skin trial found.

Why researchers looked

Houttuynia is called a 'Chinese herbal antibiotic' and is studied for skin inflammation.

Traditional use

Used in Chinese medicine for swellings and pus-forming conditions ('reducing swelling, discharging pus').

Proposed mechanism Laboratory research

Lab work suggests anti-inflammatory and antibacterial effects.

A proposed mechanism does not show that the effect happens in people.

No serious adverse events reported in either group; skin-reaction details are not in the abstract.

^{20 16}

Eczema (atopic dermatitis): mouse study of a houttuynia sugar fraction

Not enough evidence

A fraction of houttuynia polysaccharides (plant sugars) improved eczema-like skin in mice. No human trial was found.

1 studiesPolysaccharide fraction (mice).

Studies, doses and limits

What the trials found

A mouse and cell study only.

Results

Thinner skin, fewer lesions and fewer mast cells in treated mice.

Effect sizes as reported

No numbers in the abstract.

Why this rating

Animal study of an isolated fraction; no human trial.

Form and dose studied

Polysaccharide fraction (mice).

Who was studied

Mice with induced eczema-like skin; mouse and human skin cell lines.

Limits

An isolated sugar fraction is not the same as the herb or a cream.

Why researchers looked

Atopic dermatitis has few targeted treatments.

Traditional use

Not part of the traditional description we read.

Proposed mechanism Animal research

In mice and skin cells the fraction lowered inflammation messengers through the MAPK pathway and restored barrier proteins.

A proposed mechanism does not show that the effect happens in people.

Not tested in people.

⁵

Blood sugar and metabolic effects (animal work)

Not enough evidence

Reviews describe less insulin resistance in diabetic animals given houttuynia extracts. In rats, an extract also raised metformin levels. No human blood-sugar trial was found.

2 studiesEthanol extract with metformin in rats for 28 days.

Studies, doses and limits

What the trials found

Review of animal studies plus one rat and cell study of the metformin combination; no human trial.

Results

Animal improvements only; extract raised metformin exposure in rats.

Effect sizes as reported

No human numbers.

Why this rating

Animal and cell findings only.

Form and dose studied

Ethanol extract with metformin in rats for 28 days.

Who was studied

Rats, mice, cells.

Limits

Animal doses and extracts are not comparable to tea or food.

Why researchers looked

A traditional food-medicine plant now explored for diabetes in animals.

Traditional use

Not a traditional diabetes remedy in the sources read.

Proposed mechanism Animal research

Extracts eased insulin resistance in animals; blocked the metformin transporters OCT2 and MATE1 in cells.

A proposed mechanism does not show that the effect happens in people.

The metformin finding is a reason to ask a pharmacist if you take it (see the safety tab).

^{21 16}

Antibacterial and anti-inflammatory activity (lab and animal only)

Not enough evidence

Cell and animal studies report antibacterial and anti-inflammatory effects of houttuynia extracts and of sodium houttuyfonate, a derivative of its oil compound. No human trial of the herb for infection was found.

3 studiesExtracts, oil, sodium houttuyfonate in lab studies.

Studies, doses and limits

What the trials found

Narrative reviews of lab and animal studies; no human data.

Results

Lab effects only.

Effect sizes as reported

None.

Why this rating

Cell and animal work only.

Form and dose studied

Extracts, oil, sodium houttuyfonate in lab studies.

Who was studied

Cells and animals.

Limits

Sodium houttuyfonate is a drug-type derivative, not the herb.

Why researchers looked

Houttuynin (decanoyl acetaldehyde) is described as an antibacterial compound.

Traditional use

Called a 'Chinese herbal antibiotic'.

Proposed mechanism Laboratory research

Reviews describe effects on bacteria and inflammation pathways in labs. Houttuynin is unstable (it turns into 2-undecanone when heated).

A proposed mechanism does not show that the effect happens in people.

Not a replacement for antibiotics.

^{16 13 8}

Anti-cancer effects in cells (no human evidence)

Not enough evidence

Houttuynia extracts reduced the growth of several cancer cell lines in dishes and some animal models. There is no evidence that it treats cancer in people.

2 studiesExtracts and isolated compounds in dishes.

Studies, doses and limits

What the trials found

Narrative reviews; mostly cell-line work.

Results

Cells and animals only.

Effect sizes as reported

None.

Why this rating

Cell-line work.

Form and dose studied

Extracts and isolated compounds in dishes.

Who was studied

Cell lines.

Limits

Cell results rarely carry over to people.

Why researchers looked

Interest in plant compounds that cause cancer cells to die.

Traditional use

Not a traditional cancer use in the sources read.

Proposed mechanism Laboratory research

Cell studies report cancer-cell death through mitochondria in colon cells and other lines.

A proposed mechanism does not show that the effect happens in people.

No human use should be based on this.

^{16 13}

Eaten as a vegetable and herbal tea (food use, not a treatment)

Not enough evidence

In parts of China and India, the leaves and rhizomes are eaten as a vegetable. China lists the plant as both food and medicine. The reviews found no reports of harm from long-term eating, but they ask for better safety data.

2 studiesFresh leaves or tea in tradition; no amounts studied.

Studies, doses and limits

What the trials found

Narrative review statements; no controlled safety study of eating houttuynia.

Results

No human outcome.

Effect sizes as reported

None.

Why this rating

A history of eating is not a safety trial or proof of benefit.

Form and dose studied

Fresh leaves or tea in tradition; no amounts studied.

Who was studied

General population in food cultures.

Limits

Aristolactam content varies by batch and part (higher in aerial parts and flowers); see the safety tab.

Why researchers looked

Long food history.

Traditional use

Eaten as a fresh herb in parts of South China and India.

Proposed mechanism Pharmacological theory

None needed; this is a food statement.

A proposed mechanism does not show that the effect happens in people.

See the safety tab.

^{16 18}

Connections to the rest of health

How Houttuynia connects to the rest of our health knowledge graph. Every link says what kind of connection it is. A safety link (for example to the liver) is a warning, not a benefit. A study that merely mentions a topic never creates a benefit link.

Explore every connection: goals, body systems, organs, symptoms, genes

Kind of connection Connected to Search

A. Wellness goals

Connected to	Kind of connection	Evidence status	Context	Studies
Immune Support	Potential benefit	Very low certainty	COVID-19: formulas that contain houttuynia (not houttuynia alone): A meta-analysis of seven studies of Lianhua Qingwen, a formula with about a dozen herbs including houttuynia, found better results when it was added to standard care. The authors say study quality was low. No trial tested houttuynia by itself for COVID-19. Year-round allergic nose symptoms: a houttuynia-based formula: In a small Taiwan trial, adults with long-standing allergic rhinitis who took a houttuynia-based formula for 4 weeks had better nasal symptoms than those on regular medicine. The abstract gives no numbers.	^{7 2}
Immune Support	Adverse effect or risk	Not enough evidence	Houttuynia injection for respiratory infections (benefit not established; serious harm reported): In China a drug-type houttuynia injection was widely given for chest and respiratory infections. The review we read measured its harms, not whether it works, and found serious allergic reactions and deaths. This is not the same as tea or food use.	^{15 6}
Immune Support	Traditional use	Not enough evidence	Colds, cough, pneumonia and lung abscess (traditional use): Chinese medicine has long used houttuynia for fever, colds, lung infections and lung abscess. The reviews we read cite no good human trial of the plain herb for these uses.	^{16 13}
Immune Support	Preliminary mechanism	Not enough evidence	Antiviral activity (lab and computer work only): Houttuynia extracts and some of its compounds slowed several viruses in cells and in mice. A computer-docking report on SARS-CoV-2 is called 'idealized' by one review. No human trial tests this. Antibacterial and anti-inflammatory activity (lab and animal only): Cell and animal studies report antibacterial and anti-inflammatory effects of houttuynia extracts and of sodium houttuyfonate, a derivative of its oil compound. No human trial of the herb for infection was found.	^{16 13 8}
Healthy Skin	Potential benefit	Very low certainty	Acne: a four-extract skin product that includes houttuynia: In one 60-person trial, a skin product containing four plant extracts, houttuynia among them, reduced acne spots and skin inflammation markers over 8 weeks. It cannot be credited to houttuynia alone.	^{20 16}
Healthy Skin	Preliminary mechanism	Not enough evidence	Eczema (atopic dermatitis): mouse study of a houttuynia sugar fraction: A fraction of houttuynia polysaccharides (plant sugars) improved eczema-like skin in mice. No human trial was found.	⁵
Blood Sugar Balance	Preliminary mechanism	Not enough evidence	Blood sugar and metabolic effects (animal work): Reviews describe less insulin resistance in diabetic animals given houttuynia extracts. In rats, an extract also raised metformin levels. No human blood-sugar trial was found.	^{21 16}
Metabolic Health	Preliminary mechanism	Not enough evidence	Blood sugar and metabolic effects (animal work): Reviews describe less insulin resistance in diabetic animals given houttuynia extracts. In rats, an extract also raised metformin levels. No human blood-sugar trial was found.	^{21 16}
Inflammation Balance	Preliminary mechanism	Not enough evidence	Antibacterial and anti-inflammatory activity (lab and animal only): Cell and animal studies report antibacterial and anti-inflammatory effects of houttuynia extracts and of sodium houttuyfonate, a derivative of its oil compound. No human trial of the herb for infection was found.	^{16 13 8}
Healthy Aging	Preliminary mechanism	Not enough evidence	Anti-cancer effects in cells (no human evidence): Houttuynia extracts reduced the growth of several cancer cell lines in dishes and some animal models. There is no evidence that it treats cancer in people.	^{16 13}

B. Body systems

Connected to	Kind of connection	Evidence status	Context	Studies
Immune System	Biological association	Very low certainty	Studied for: COVID-19: formulas that contain houttuynia (not houttuynia alone) Studied for: Year-round allergic nose symptoms: a houttuynia-based formula	^{7 2 15 6 16 13 8}
Respiratory System	Biological association	Very low certainty	Studied for: COVID-19: formulas that contain houttuynia (not houttuynia alone) Studied for: Year-round allergic nose symptoms: a houttuynia-based formula	^{7 2 15 6 16 13}
Integumentary System	Biological association	Very low certainty	Studied for: Acne: a four-extract skin product that includes houttuynia Studied for: Eczema (atopic dermatitis): mouse study of a houttuynia sugar fraction	^{20 16 5}
Metabolism Weight	Biological association	Not enough evidence	Studied for: Blood sugar and metabolic effects (animal work)	^{21 16}
Endocrine System	Biological association	Not enough evidence	Studied for: Blood sugar and metabolic effects (animal work)	^{21 16}
Digestive System	Biological association	Not enough evidence	Studied for: Eaten as a vegetable and herbal tea (food use, not a treatment)	^{16 18}

E. Health conditions

Connected to	Kind of connection	Evidence status	Context	Studies
Anaphylaxis	Adverse effect or risk	Documented risk	Main serious reaction reported to the injection; not shown for tea or food. Most serious reaction reported to the injection; not shown for tea or food. Preparation or compound: Houttuynia injection	^{15 6}
Allergic rhinitis	Potential benefit	Very low certainty	One small trial; no numbers in the abstract. Preparation or compound: Houttuynia-based formula (ZYF)	²

I. Biomarkers

Connected to	Kind of connection	Evidence status	Context	Studies
Nasal symptoms and quality of life (Chinese Rhinosinusitis Outcome Measures)	Biological association	Very low certainty	Measured in trials: lowered. A lab change is not the same as a health benefit.	²
IL-6 release (cell model)	Biological association	Not enough evidence	Measured in trials: lowered. A lab change is not the same as a health benefit.	²
COVID-19 trial outcomes with Lianhua Qingwen (total efficacy, severe conversion, fever duration)	Biological association	Very low certainty	Measured in trials: mixed. A lab change is not the same as a health benefit.	⁷
Acne lesion counts (inflammatory and non-inflammatory)	Biological association	Very low certainty	Measured in trials: lowered. A lab change is not the same as a health benefit.	²⁰
Skin IL-1 alpha, IL-8 and keratin 16 (skin biopsy)	Biological association	Very low certainty	Measured in trials: lowered. A lab change is not the same as a health benefit.	²⁰
Blood and liver metformin levels (rats)	Biological association	Not enough evidence	Measured in trials: raised. A lab change is not the same as a health benefit.	²¹

J. Genes, receptors and enzymes

Connected to	Kind of connection	Evidence status	Context	Studies
Mrgprb2 / MRGPRX2	Preliminary mechanism	Animal	Mast cell receptor linked to injection reactions in mice.	¹⁴
COX-1 and COX-2	Preliminary mechanism	In vitro	Blocked by becatamide in mouse blood.	¹²
CYP2E1	Preliminary mechanism	Animal	Lower activity in mice.	¹⁶
IL-6	Preliminary mechanism	In vitro	Release lowered in a cell model by the formula.	²

K. Biological pathways

Connected to	Kind of connection	Evidence status	Context	Studies
OCT2 and MATE1	Preliminary mechanism	Animal	Blocked by the ethanol extract in cells; raised metformin in rats.	²¹
PI3K-Akt	Preliminary mechanism	In vitro	Lowered in embryo cell clusters.	⁹

L. Active compounds

Connected to	Kind of connection	Evidence status	Context	Studies
Houttuynin (decanoyl acetaldehyde; 3-oxododecanal)	Biological association	In vitro	Source of the fishy smell and reported as antibacterial. Easily turns into 2-undecanone (CID 8163) at high temperature. Unstable in blood plasma. Preparation or compound: Aerial parts (essential oil)	^{16 8 11}
Sodium houttuyfonate	Biological association	In vitro	Reviews report antibacterial and quorum-sensing effects in the lab. It is a drug-type derivative, not tea or food. Preparation or compound: Not a plant compound; made from houttuynin (PubChem structure name: sodium 1-hydroxy-3-oxododecane-1-sulfonate)	^{16 11}
Quercitrin	Biological association	In vitro	Reported in the plant; a computer-docking SARS-CoV-2 report is called idealized by one review. Preparation or compound: Aerial parts	^{16 11}
Hyperoside and rutin	Biological association	Lab analysis	Reported marker flavonoids; rutin CID 5280805 is not looked up in this file. Preparation or compound: Aerial parts	^{16 11}
Aristolactams (AI, AII, BII, FI)	Biological association	Animal	Relatives of aristolochic acids. Toxic to kidney cells in a dish; much less kidney-toxic than aristolochic acid I in mice. AII CID 148657 (name match). Papers disagree on which are present (AI not detected in 22 samples in one paper). Preparation or compound: Aerial parts and flowers more than underground parts	^{18 19 22 17 11}
Aristolochic acid I (comparison, not found in the plant)	Biological association	Animal	The kidney toxin used as the positive control in the mouse study. Preparation or compound: Not detected in 22 samples or in 75 dried herb batches	^{19 22 17 11}
Becatamide	Biological association	In vitro	Blocked COX-1 and COX-2 and lowered platelet activation markers in mouse blood. No CID looked up (name not matched). Preparation or compound: Found in Houttuynia cordata	¹²
Polysaccharides	Biological association	Animal	A fraction improved eczema-like skin in mice. Preparation or compound: Not stated	⁵

Each connection is stored with the ingredient, the connected record, the kind of connection, the evidence status, the supporting studies, the preparation or compound involved, and a review date (last reviewed October 25, 2026).

Evidence at a glance: what changed in 6 lab markers

Tap a number to show only those markers. Counts are markers, not people.

Inflammatory1 Other311

The evidence

The research centre for Houttuynia. Every record here was checked against PubMed or the publisher's page; we never list a study we have not verified.

Evidence by outcome

Bars show how many studies we found for each outcome, not how well it works. People are added up across trials only (reviews are not added, to avoid double counting).

How we rate evidence

High certainty

Several well-run trials or reviews agree; further research is unlikely to change the picture much.

Moderate certainty

Good trials point the same way, but some doubt remains.

Low certainty

Small, short or inconsistent trials. The finding could change with better research.

Very low certainty

Very little or very weak human data. Treat as a possibility only.

Mixed findings

Trials or reviews disagree with each other.

Not enough evidence

Not enough human research to say either way.

Evidence against

Good research found no benefit, or found harm.

We take the reviewers' own risk-of-bias and GRADE statements where they exist, and otherwise judge from study size, design, consistency, directness and whether a product matches what is sold. Many publications do not make evidence high. Method note: We started from our existing houttuynia summary, read the abstract of every PubMed study cited, read key review sections in full text via Europe PMC, checked the plant name in GBIF and NCBI Taxonomy and compounds in PubChem. NCCIH, Memorial Sloan Kettering, MedlinePlus, LiverTox and LactMed were checked and nothing was found; EMA could not be checked properly.

Study library: every study we checked 22

Design Research type Plant part Year from Search

Study	Design	People	Product, dose, length	Result
AlternativeMed editorial check 2026 Regulator and reference-site check for Houttuynia (NCCIH, MSK, MedlinePlus, EMA, LiverTox, LactMed) Web and NCBI Bookshelf check, 2026-10-25 · Source	Database record not applicable		Houttuynia cordata not applicable Not applicable Not applicable · vs Not applicable	NCCIH, Memorial Sloan Kettering and MedlinePlus pages for houttuynia returned HTTP 404. The EMA site search returned HTTP 401 (automated access refused), so EMA/HMPC could not be checked. A search of NCBI Bookshelf for houttuynia in LiverTox and LactMed returned 0 records. FDA, WHO and Commission E were not checked. Methods, limits and funding Limitations: Not found is not proof that no document exists, especially for EMA. Read: web-check
Chang 2022 A Houttuynia cordata-based Chinese herbal formula improved symptoms of allergic rhinitis during the COVID-19 pandemic J Chin Med Assoc · PMID 35421875 · DOI	Controlled trial	Adults aged 20 to 60 with at least a 2-year history of moderate to severe year-round allergic rhinitis (hay-fever type nose symptoms), Taiwan	Zheng-Yi-Fang (ZYF), a Houttuynia cordata-based Chinese herbal formula (other ingredients not listed in the abstract) unspecified Not stated in abstract 4 weeks · vs Regular western medicine	People taking ZYF improved nasal obstruction, nasal secretion, loss of smell and postnasal drip compared with the control group. ZYF also reduced IL-6 release in a lab model of inflammation (cells exposed to LPS). Better nasal symptom scores than the comparison treatment (no numbers in the abstract) Methods, limits and funding Outcomes: Nasal symptoms and quality of life (Chinese Rhinosinusitis Outcome Measures); IL-6 in a cell model Limitations: The abstract gives no numbers, no group sizes and does not say whether participants or assessors were blinded. It tested a multi-herb formula in allergic rhinitis, not COVID-19 infection and not Houttuynia alone. The comparison was ordinary medicine, not a placebo. Authors declare no conflicts of interest. Funding and conflicts: Authors declare no conflicts of interest. Read: abstract
Chen 2021 Acute and subacute toxicity evaluation of Houttuynia cordata ethanol extract and plasma metabolic profiling analysis in both male and female rats J Appl Toxicol · PMID 34057207 · DOI	Animal research animal	Male and female Sprague-Dawley rats	95% ethanol extract of Houttuynia cordata (HCE) whole plant (parts not stated in abstract) Single dose 2000 mg/kg; subacute 250, 500 and 1000 mg/kg/day by mouth Single dose, or 28 days · vs Control rats	The single 2000 mg/kg dose was not toxic. In the 28-day study, no significant adverse effect was seen at 250 mg/kg/day. Five rats died in the 500 and 1000 mg/kg/day groups, which showed liver and kidney toxicity. Male rats were more sensitive. Plasma tests showed disturbances in energy, amino acid and lipid metabolism. The authors call 250 mg/kg/day the no-observed-adverse-effect level in rats over 28 days and say long-term use of large doses showed some liver and kidney toxicity. Liver and kidney toxicity and deaths in rats at higher doses of a strong alcohol extract Methods, limits and funding Outcomes: Blood counts, chemistry, tissue under the microscope, plasma metabolic profile Limitations: Rats, strong alcohol extract, 28 days. Rat doses cannot be converted into human amounts. Not tested on tea or food forms. Read: abstract
GBIF Secretariat 2026 GBIF Backbone Taxonomy: Houttuynia cordata Thunb. (key 5384931) GBIF API (api.gbif.org), read 2026-10-25 · Source	Database record not applicable		Houttuynia cordata not applicable Not applicable Not applicable · vs Not applicable	GBIF returns Houttuynia cordata Thunb. as an ACCEPTED species (exact match, confidence 99), family Saururaceae, order Piperales, class Magnoliopsida. Methods, limits and funding Limitations: Taxonomy databases can change. Read: api
Huang 2025 Houttuynia cordata polysaccharides ameliorate atopic dermatitis in mice through modulation of skin immune barrier and lipid metabolism Int J Biol Macromol · PMID 40393586 · DOI	Animal research animal	Mice with MC903-induced atopic dermatitis; mouse macrophage and human skin cell lines	Houttuynia cordata polysaccharides (HCP) unspecified Not stated in abstract Not stated in abstract · vs Untreated mice	HCP reduced skin thickness, skin lesions, mast cell infiltration and inflammatory messengers, and restored barrier proteins in mice. In cells, it reduced inflammatory messengers through the MAPK pathway. The authors call it a promising topical candidate. Eczema-like skin changes improved in mice Methods, limits and funding Outcomes: Skin thickness, lesions, mast cells, inflammatory messengers, skin barrier proteins, lipid metabolism Limitations: Mouse and cell study of an isolated sugar fraction, not the whole herb. No human data found. Read: abstract
Ji 2009 Comments on serious anaphylaxis caused by nine Chinese herbal injections used to treat common colds and upper respiratory tract infections Regul Toxicol Pharmacol · PMID 19559066 · DOI	Systematic review	150 Case reports of severe allergic shock or death after a Chinese herbal injection given for a common cold or upper respiratory infection (150 cases, 9 injection types)	Chinese herbal injections, including Yuxingcao (Houttuynia cordata) injection unspecified Not stated Not applicable (case reports) · vs None	Full text read. The authors report that Yuxingcao herbal injection, made from an extract of Houttuynia cordata, caused severe reactions in four children, two of them fatal, in the first half of 2006. In response the Chinese drug regulator (SFDA) halted its use in hospitals in June 2006. After reviewing manufacturing and quality control it allowed a limited number of manufacturers to resume production. The authors conclude that the risk of serious or lethal anaphylaxis should rule out herbal injections for common colds and upper respiratory infections, and suggest using them only for severe disease. Authors advise against herbal injections for colds and mild infections Methods, limits and funding Outcomes: Anaphylaxis risk and regulatory response Limitations: A commentary and review of published case reports, not a measure of how often reactions occur. Covers nine injections, not Houttuynia alone. The current legal status of the Houttuynia injection was not checked. Read: full-text
Li 2022 Efficacy and Safety of Chinese Medicine Lianhua Qingwen for Treating COVID-19: An Updated meta-Analysis Front Pharmacol · PMID 35721166 · DOI	Meta-analysis	916 People with COVID-19 in 7 studies (916 participants) (randomized trials or case-control studies)	Lianhua Qingwen (LQ), a multi-herb Chinese formula that includes Houttuynia cordata aerial part among about a dozen ingredients, given together with western medicine aerial part (as one ingredient) Not stated in abstract Not stated in abstract · vs Western medicine alone	Compared with the control group, the LQ group had better total efficacy (OR 2.23), fewer adverse events (OR 0.42), better chest CT findings (OR 1.74) and a lower rate of conversion to severe cases (OR 0.47), plus faster symptom improvement and shorter fever. The authors say the studies had low methodological quality and rushed designs, so the meta-analysis cannot draw clear conclusions. Possible benefit of the whole formula added to standard care; certainty low because of study quality (Total efficacy OR 2.23 (95% CI 1.56 to 3.18); adverse events OR 0.42 (0.18 to 0.97); chest CT OR 1.74 (1.12 to 2.72); conversion to severe OR 0.47 (0.30 to 0.75)) Methods, limits and funding Outcomes: Total efficacy, adverse events, chest CT, conversion to severe disease, symptoms, fever duration Limitations: This is a formula of many herbs, so nothing can be credited to Houttuynia. Low-quality, rushed studies according to the authors. Formula composition (including Houttuynia) comes from the full text of the same paper. Registration: PROSPERO CRD42020190757 Read: abstract
Liu 2021 Development and Verification of a Precolumn Derivatization LC-MS/MS Method for the Pharmacokinetic Study of Houttuynine of Houttuynia Essential Oil Molecules · PMID 33923761 · DOI	Animal research animal	Rats given Houttuynia essential oil by mouth and by injection	Houttuynia essential oil (houttuynine, the main active and quality marker) unspecified Not stated in abstract Not stated in abstract · vs Not applicable	Houttuynine (houttuynin) is active but not stable in plasma, so blood levels were hard to measure. The authors built a method that locks it into a stable compound, validated it, and applied it to rats given the oil by gavage and by injection. Houttuynin is unstable in blood; no human data Methods, limits and funding Outcomes: Method to measure houttuynine in rat blood Limitations: A lab-method paper. The abstract gives no half-life or concentration values. Read: abstract
Liu 2023 The Mechanism of Houttuynia cordata Embryotoxicity Was Explored in Combination with an Experimental Model and Network Pharmacology Toxins (Basel) · PMID 36668893 · DOI	Laboratory research in-vitro	Zebrafish embryos and embryoid bodies (stem-cell cell clusters); computer network analysis	Houttuynia cordata (preparation not stated in abstract) unspecified Not stated in abstract Not stated in abstract · vs Untreated embryos	The authors report that Houttuynia cordata affected embryonic development in these lab models and lowered AKT1 and PI3K in the embryoid body. Computer analysis pointed to oleanolic acid, a lignan and aristolactam AII as the main compounds linked to the toxicity. They suggest it may affect embryonic development through the PI3K-Akt pathway. Lab warning signal for development of embryos; not shown in pregnancy Methods, limits and funding Outcomes: Embryonic development; gene activity in the PI3K-Akt pathway Limitations: Zebrafish, cell clusters and computer modelling only. No dose, no pregnancy data. A reason for caution, not proof of harm to a human baby. Read: abstract
NCBI 2026 NCBI Taxonomy: Houttuynia cordata (taxid 16752) NCBI Taxonomy (E-utilities), read 2026-10-25 · Source	Database record not applicable		Houttuynia cordata not applicable Not applicable Not applicable · vs Not applicable	NCBI Taxonomy returns taxid 16752 as the species Houttuynia cordata. Methods, limits and funding Limitations: Taxonomy databases can change. Read: api
NCBI PubChem 2026 PubChem compound records for Houttuynia constituents PubChem PUG REST, read 2026-10-25 · Source	Database record not applicable		Houttuynia cordata not applicable Not applicable Not applicable · vs Not applicable	Name look-ups returned: houttuynin (3-oxododecanal) CID 122640, C12H22O2; sodium houttuyfonate CID 23663544, C12H23NaO5S (sodium 1-hydroxy-3-oxododecane-1-sulfonate); 2-undecanone CID 8163; quercitrin CID 5280459; hyperoside CID 5281643; aristolactam I CID 96710; aristolactam AII CID 148657 (name match); aristolochic acid I CID 2236 (comparison compound). Methods, limits and funding Limitations: A name match is not proof that a paper meant the same structure. Several aristolactam names (BII, FI) matched other trivial names, so no CID is given for them. Read: api
Park 2015 Becatamide Found in Houttuynia cordata Suppresses P-selectin Expression Via Inhibiting COX Enzyme, Not Increasing cAMP in Platelets Phytother Res · PMID 26112157 · DOI	Laboratory research in-vitro	Mouse blood and platelet lab tests	Becatamide, a compound found in Houttuynia cordata, and related compounds unspecified Becatamide 0.25 uM in mouse blood Lab test · vs Related compounds; untreated blood	Becatamide blocked COX-1 (IC50 0.27 uM) and COX-2 (IC50 0.78 uM). In mouse blood treated with becatamide, thromboxane B2 and P-selectin expression fell by 35% and 28%. The author concludes it may inhibit platelet activation by blocking COX. An isolated compound reduced platelet activation markers in mouse blood in the lab Methods, limits and funding Outcomes: COX-1 and COX-2 inhibition, thromboxane B2, P-selectin Limitations: One isolated compound in the lab, not the whole herb. No human or bleeding data. Used here only to explain why a theoretical interaction with blood thinners cannot be ruled out. Read: abstract
Rafiq 2022 Pharmacological Effects of Houttuynia cordata Thunb (H. cordata): A Comprehensive Review Pharmaceuticals (Basel) · PMID 36145299 · DOI	Review review	Narrative review of Houttuynia cordata pharmacology	Houttuynia cordata (many preparations) unspecified Varies Varies · vs Not applicable	Not re-opened for this file. Our earlier reading notes say: lab and animal anti-inflammatory, antiviral, antibacterial and anti-tumour effects; traditional food and medicine use. Overview source for lab findings Methods, limits and funding Outcomes: Pharmacological effects Limitations: Not re-opened for this file; used only to support statements already in our summary. Read: summary-only
Sun 2022 Mrgprb2 gene plays a role in the anaphylactoid reactions induced by Houttuynia cordata injection J Ethnopharmacol · PMID 35104575 · DOI	Animal research animal	Mice, including mice lacking the gene Mrgprb2 (the mouse version of the human receptor MRGPRX2)	Houttuynia cordata injection unspecified Not stated in abstract Not stated in abstract · vs Normal mice versus Mrgprb2 knockout mice	Mice without Mrgprb2 had less mast cell degranulation, less histamine release and less paw swelling after the injection. The authors conclude this gene plays a role in the pseudo-allergic (anaphylactoid) reactions the injection causes. Suggests one non-classic allergy route for injection reactions in mice Methods, limits and funding Outcomes: Mast cell degranulation, histamine release, paw swelling, gene activity Limitations: Mouse study of the injection; not shown in people. It explains a possible mechanism, not who will react. Read: abstract
Wang 2010 Adverse events to Houttuynia injection: A systematic review J Evid Based Med · PMID 21349062 · DOI	Systematic review	1232 Reported cases of adverse reactions to Houttuynia injection in China (645 articles, 1,232 cases, searched to June 2010)	Houttuynia injection (a drug-type product made from the herb; not tea or food) unspecified Not stated in abstract Not applicable (case reports) · vs Injection used alone versus with other drugs; intravenous versus intramuscular	Respiratory diseases accounted for 52.44% of cases treated. Reactions involved the respiratory system (37.42%), skin (34.66%), digestive system (25.49%), circulatory system (25.41%) and nervous system (23.96%). Serious systemic reactions were 22.56% of all reactions. Using the WHO grading, 22.56%, 36.28%, 16.48% and 24.68% of cases were grades I, II, III and IV. Anaphylactic shock made up 58.99% of the most serious (grade I) reactions. All deaths were from allergic shock except one, from multiple organ failure after anaphylactic purpura. The fastest three reactions began within one minute of injection. Of 286 cases with a described allergy history, 15.03% were allergic to penicillin. Combined use with macrolide antibiotics carried a higher reaction risk than the injection alone (RR 8.80). Intravenous use carried a higher risk than intramuscular use (OR 6.86). Serious, sometimes fatal allergic reactions; risk higher with intravenous use and with macrolide antibiotics (Macrolide combination RR 8.80 (95% CI 6.12 to 12.65); intravenous vs intramuscular OR 6.86 (95% CI 1.88 to 56.95)) Methods, limits and funding Outcomes: Types, severity and risk factors of adverse reactions Limitations: Built from published case reports, so it cannot give how often reactions happen in everyone treated (no count of all people treated). Reports are voluntary and can be incomplete. Applies to the injection only. Read: abstract
Wu 2021 Houttuynia cordata Thunb: An Ethnopharmacological Review Front Pharmacol · PMID 34539401 · DOI	Review review	Narrative review of laboratory, animal and a few human reports on Houttuynia cordata	Houttuynia cordata (many preparations) aerial parts and rhizome Varies Varies · vs Not applicable	Reads in full text via Europe PMC. Describes decanoyl acetaldehyde (houttuynin) as the source of the fishy smell and an antibacterial compound that easily turns into 2-undecanone when heated. Lists flavonoids (rutin, hyperoside, quercetin, quercitrin) and alkaloids including aristololactam. Reports lab and animal anti-inflammatory, antiviral, antibacterial and anti-cancer findings. Mentions a four-extract acne trial and an eye-drop combination, with no numbers. Says the SARS-CoV-2 computer-docking report seems idealized. On toxicity: no reports of toxicity from long-term eating as a vegetable in some areas of China and India, and the National Health Commission of China listed it as both food and medicine in 2013, but the authors call for better safety data and note possible liver and kidney toxicity. Overview source; mostly lab and animal findings Methods, limits and funding Outcomes: Chemistry, pharmacology, toxicity, clinical mentions Limitations: A narrative review, not a systematic review. Some citation and typing errors are noted in it. Only the relevant sections were searched, not every line. Read: full-text
Wu 2022 Rapid Analysis of Aristolochic Acids and Aristolactams in Houttuyniae Herba by LC-MS/MS Molecules · PMID 36558101 · DOI	Laboratory research in-vitro	75 75 batches of Houttuyniae Herba (dried herb) from different origins	Houttuyniae Herba (dried Houttuynia cordata) dried herb Not applicable Not applicable · vs None	Six alkaloids (including aristololactam AII, FI and BII) were found and measured in 75 batches. No aristolochic acids were found at a detection limit of 4 ng/mL or lower. The abstract also says no aristololactams were found, which conflicts with the aristololactams it lists; the body text was used. No aristolochic acids detected in 75 batches; several aristolactams detected Methods, limits and funding Outcomes: Content of 7 alkaloids and 5 aristolochic acids by LC-MS/MS Limitations: Chemistry only; no toxicity tested. The abstract and the body text disagree about aristolactams; the body text was used. Read: full-text
Xu 2023 In vitro nephrotoxicity and quantitative UPLC-MS analysis of three aristololactams in Houttuynia cordata J Pharm Biomed Anal · PMID 36812795 · DOI	Laboratory research in-vitro	30 Human kidney tubule cells (HK-2) and 30 batches of Houttuynia cordata from different regions and plant parts	Three aristolactams isolated from Houttuynia cordata (AL AII, AL FI, AL BII) aerial and underground parts, flowers Cells: IC50 3.88 to 20.63 uM Cell tests · vs Aristolactam I	All three aristolactams were about as toxic to the cells as aristolactam I (IC50 3.88 to 20.63 uM), raised reactive oxygen species and the fibrosis markers TGF-beta1 and fibronectin, and changed cell shape toward fibrosis. Content varied a lot between 30 batches. The aerial part held much more (3.20 to 108.19 ug/g) than the underground part (0.95 to 11.66 ug/g), flowers the most. No aristolactams were detected in the water extract of any part. Aristolactams in the plant are toxic to kidney cells in a dish; content is higher in aerial parts; water extracts had none detected Methods, limits and funding Outcomes: Cell toxicity, reactive oxygen species, fibrosis markers, aristolactam content by batch and part Limitations: Cell and chemistry study only. The real exposure from tea, food or supplements was not measured. Read: abstract
Xue 2025 Safety evaluation of aristolactams from the edible herb Houttuynia cordata: implications for dietary exposure and food safety Food Res Int · PMID 41267271 · DOI	Animal research animal	BALB/c mice and human kidney tubule cells (HK-2) in a lab; subacute toxicity, metabolism and bioavailability	Aristolactam I (AL I), aristolactam BII (AL BII) and an aristolactam fraction from Houttuynia cordata (HCA), with aristolochic acid I (AA I) as the harmful comparison whole plant fraction (parts not stated in abstract) Not stated in abstract Subacute (length not stated in abstract) · vs Aristolochic acid I as positive control	AL I, AL BII and the HCA fraction caused only modest rises in body weight, blood kidney markers and urine injury markers. AA I caused significantly larger increases, through oxidative stress and complement activation. In cells, AL I was more toxic to HK-2 cells (IC50 2.76 uM) than AA I (79.85 uM), but its toxicity in mice was lower. Oral bioavailability was 52.71% for AL I and 39.60% for AL BII versus 99.83% for AA I, the aristolactams were broken down into less toxic products, and no unchanged aristolactam appeared in urine. The authors say the findings support safe consumption at typical dietary levels, and stress monitoring aristolactam content. In mice, Houttuynia aristolactams were much less kidney-toxic than aristolochic acid I despite higher toxicity in cell tests Methods, limits and funding Outcomes: Body weight, blood kidney markers, urine injury markers, cell toxicity, oral bioavailability Limitations: Mice, not people. The abstract gives no dose, duration or group sizes. The authors conclude "typical dietary levels" are safe; that is their reading of mouse data, not a human safety study. No competing interests declared. Funding and conflicts: Authors declare no known competing interests. Read: abstract
Yang 2021 Clinical efficacy of herbal extracts in treatment of mild to moderate acne vulgaris: an 8-week, double-blinded, randomized, controlled trial J Dermatolog Treat · PMID 31424962 · DOI	Controlled trial	60 60 people with mild to moderate acne vulgaris, Korea	Topical herbal extract blend of mangosteen, Lithospermum officinale, Tribulus terrestris and Houttuynia cordata (parts and amounts not stated in abstract) unspecified Not stated in abstract 8 weeks · vs Vehicle (same base without the extracts), double-blind	Both inflammatory and non-inflammatory acne lesion counts decreased in the herbal extract group (p<.05). In skin biopsies, IL-1 alpha, IL-8 and keratin 16 fell significantly in the extract group compared with the vehicle group. No serious adverse events in either group. Fewer acne lesions and lower skin inflammation markers with a four-extract blend; cannot be credited to Houttuynia alone Methods, limits and funding Outcomes: Inflammatory and non-inflammatory lesion counts, Investigator Global Assessment, satisfaction, safety, skin biopsy markers Limitations: Four plant extracts combined, so the effect cannot be assigned to Houttuynia. The abstract gives no between-group lesion numbers or funding statement. Small trial (60 people) over 8 weeks. Read: abstract
You 2018 Houttuynia cordata extract increased systemic exposure and liver concentrations of metformin through OCTs and MATEs in rats Phytother Res · PMID 29480578 · DOI	Animal research animal	HEK-293 cells expressing human transporters, and diabetic and normal rats	Houttuynia cordata ethanol extract (HCT) given with metformin unspecified Not stated in abstract 28 days in rats · vs Metformin alone	HCT blocked metformin transport by the human transporters OCT2 and MATE1 in cells. In rats treated for 28 days, HCT reduced the kidney excretion of metformin, so more metformin stayed in the blood, and more went into the liver, with better glucose lowering and no hypoglycaemia or kidney function impairment. The authors suggest the combination is applicable for efficacy and safety. In rats, the extract raised metformin levels Methods, limits and funding Outcomes: Metformin transport, blood and liver levels, glucose lowering, kidney function Limitations: Cells and rats only. Whether this happens in people is unknown. Higher metformin levels could matter for someone who already takes it. Read: abstract
Yu 2022 Study of the Contents of Analogues of Aristolochic Acid in Houttuynia cordata by Ultra-High Performance Liquid Chromatography Tandem Mass Spectrometry Foods · PMID 35159454 · DOI	Laboratory research in-vitro	22 22 Houttuynia cordata samples from 22 regions of China	Houttuynia cordata unspecified Not applicable Not applicable · vs None	None of the three carcinogenic compounds (aristolochic acid I, aristolochic acid II and aristolactam I) was detected in any of the 22 samples. Aristolactam AII, which the authors say has not been reported to have genotoxicity, was found in all samples. No aristolochic acid I or II or aristolactam I detected in 22 samples Methods, limits and funding Outcomes: Content of aristolochic acids I and II, aristolactam I and aristolactam AII by UHPLC-MS/MS Limitations: Chemistry only, 22 samples. Another paper in this file (Xue 2025) treats aristolactam I as a main aristolactam, so papers do not fully agree on which aristolactams are present. Read: abstract

What we do not know yet

No trial of plain houttuynia (tea, capsule or dried herb) for any infection.
COVID-19 evidence is for a many-herb formula from low-quality studies.
No human kidney or liver safety data; the aristolactam question is not settled.
No human interaction studies with blood thinners, metformin or CYP-metabolised drugs.
The injection's current status was not checked; EMA, FDA and WHO could not be properly checked.
No pregnancy or breastfeeding data in people.

How new research is added

New studies are found by a saved PubMed search and placed in a review queue. A person reads each one before it changes any rating or safety wording.

Lab markers

What changed in blood or saliva tests during trials. A changed number is not the same as feeling or being healthier, and we give no "normal ranges".

Inflammatory 1

▼

IL-6 release (cell model)

Lowered in trials

An inflammation messenger that rises in allergic nose disease and in COVID-19.

Not enough evidence

Results and what it means

Why it was measured

To explain how the formula might work.

Statistical vs. clinical meaning

A change in a cell model says nothing about benefit in people.

Safety implication

None.

Study	Result	People	Product
²	The formula inhibited IL-6 secretion in cells stimulated with LPS (lab model, not measured in the trial participants).	Cell model	ZYF formula

Other 5

▼

Nasal symptoms and quality of life (Chinese Rhinosinusitis Outcome Measures)

Lowered in trials

A questionnaire score of nasal blockage, discharge, smell and postnasal drip.

Very low certainty

Results and what it means

Why it was measured

To see if the formula eases allergic nose symptoms.

Statistical vs. clinical meaning

A symptom score is what patients feel, but there are no numbers or blinding details in the abstract.

Safety implication

None reported in the abstract.

Study	Result	People	Product
²	Improved nasal obstruction, nasal secretion, hyposmia and postnasal drip compared with the control group (no numbers in the abstract).	Adults with moderate to severe year-round allergic rhinitis	ZYF houttuynia-based formula, 4 weeks, versus regular western medicine

↕

COVID-19 trial outcomes with Lianhua Qingwen (total efficacy, severe conversion, fever duration)

Mixed results in trials

Combined clinical outcomes reported by the meta-analysis.

Very low certainty

Results and what it means

Why it was measured

To judge whether the formula added to standard care helps.

Statistical vs. clinical meaning

Pooled results look favourable but come from low-quality studies of a many-herb formula.

Safety implication

Fewer adverse events were reported with the formula in the pooled data.

Study	Result	People	Product
⁷	Total efficacy OR 2.23; conversion to severe OR 0.47; adverse events OR 0.42; fever duration shorter (p<0.05). Authors say no clear conclusion because of low study quality.	7 studies, 916 people with COVID-19	Lianhua Qingwen plus western medicine versus western medicine alone

⁷

▼

Acne lesion counts (inflammatory and non-inflammatory)

Lowered in trials

Number of acne spots counted by the investigator.

Very low certainty

Results and what it means

Why it was measured

To judge acne severity.

Statistical vs. clinical meaning

A within-group fall; a blend; small trial.

Safety implication

No serious adverse events.

Study	Result	People	Product
²⁰	Both inflammatory and non-inflammatory lesion counts decreased in the extract group (p<.05); no between-group numbers in the abstract.	60 adults with mild to moderate acne	Four-extract topical blend including houttuynia, 8 weeks, versus vehicle

²⁰

▼

Skin IL-1 alpha, IL-8 and keratin 16 (skin biopsy)

Lowered in trials

Inflammation and skin-turnover markers measured by staining skin samples.

Very low certainty

Results and what it means

Why it was measured

To confirm immune changes in the skin.

Statistical vs. clinical meaning

A marker change, not a clinical outcome; blend.

Safety implication

None.

Study	Result	People	Product
²⁰	IL-1 alpha, IL-8 and keratin 16 fell significantly in the extract group compared with the vehicle group from baseline to week 8.	60 adults with acne	Four-extract blend including houttuynia, 8 weeks

²⁰

▲

Blood and liver metformin levels (rats)

Raised in trials

How much metformin stays in the body.

Not enough evidence

Results and what it means

Why it was measured

To explain why houttuynia plus metformin lowered glucose more in rats.

Statistical vs. clinical meaning

Animal finding; shows an interaction risk in theory rather than a benefit.

Safety implication

Higher drug exposure could matter for people who take metformin.

Study	Result	People	Product
²¹	The extract reduced kidney excretion of metformin and raised its blood exposure and liver uptake in rats after 28 days.	Rats	Houttuynia ethanol extract plus metformin

²¹

Safety

What studies and official sources say about side effects, medicines and who should check first. Each point shows what kind of evidence it comes from.

Main points to know 8

The injection can cause serious allergic reactions, including death. A systematic review of 1,232 reported reactions to Houttuynia injection found that 22.56% were serious body-wide reactions. Anaphylactic shock made up 58.99% of the most severe grade, and reactions began within one minute in the fastest cases. All deaths were from allergic shock except one (multiple organ failure after anaphylactic purpura). Intravenous use and use with macrolide antibiotics raised the risk. This is a hospital drug product, not tea or food. The reports have no count of everyone treated, so the true rate cannot be worked out. Systematic review ¹⁵
Chinese regulators halted hospital use of the injection in 2006, then allowed limited production. A review of herbal injections reports that Yuxingcao (Houttuynia) injection caused severe reactions in four children, two fatal, in the first half of 2006. The Chinese drug regulator (SFDA) halted its use in hospitals in June 2006. After reviewing manufacturing and quality control, it let a limited number of manufacturers resume production. The same authors argue that herbal injections should not be used for common colds. We did not check the injection's current legal status. Authoritative guidance ⁶
Herb, injection and food are not the same thing. The reaction data come from the injection. We found no published case report of anaphylaxis or organ injury from eating the fresh herb or drinking the tea in our PubMed searches. That is not proof it never happens. Reviews say no long-term toxicity has been reported from eating it as a vegetable in some areas of China and India, and they ask for better data. Narrative review ^{16 15}
Aristolactams: a kidney question, not a proven human harm. Houttuynia contains aristolactams, relatives of the kidney-damaging aristolochic acids. Two chemistry papers did not detect aristolochic acid I or II (75 batches; 22 samples) but did find aristolactams. In kidney cells, three of them were about as toxic as aristolactam I. In mice, aristolactam I, aristolactam BII and an aristolactam fraction from the plant raised kidney markers only modestly, while aristolochic acid I raised them much more. No human kidney outcome has been studied, and the papers do not agree on which aristolactams are present. Content is higher in aerial parts and flowers, and none was detected in water extracts. Animal research ^{19 18 17 22}
Strong extracts caused liver and kidney damage and deaths in rats. Rats given a 95% ethanol extract for 28 days had no significant adverse effect at 250 mg/kg/day, but five rats died at 500 and 1000 mg/kg/day, with liver and kidney toxicity. Males were more sensitive. This is an animal result for a concentrated extract, not for tea or food. Animal research ³
Pregnancy and breastfeeding: no human safety data. A lab study in zebrafish embryos and embryo cell clusters reported effects on embryonic development. This is a warning signal, not proof of harm in pregnancy. No LactMed entry was found. Avoid medicinal amounts and concentrated extracts unless a clinician says otherwise. Laboratory research ^{9 1}
May raise levels of some medicines (theory and animal data only). In rats and cell tests, an extract raised metformin levels. An isolated compound blocked COX enzymes and platelet markers in mouse blood. No human study of houttuynia with blood thinners, metformin or other medicines was found. Ask a pharmacist before combining. Animal research ^{21 12}
No oral safety trial in people. The only human trials we found were short (4 and 8 weeks) trials of a multi-herb formula and a four-extract skin blend. They report few safety details in their abstracts. 'Natural' or 'eaten as a vegetable' is not a safety test. Authoritative guidance ^{2 20}

Side effects 4

Effect	How often	Severity and recovery	Evidence
Anaphylactic shock and other serious body-wide reactions (injection) Injection only. Fastest onset within one minute.	Serious systemic reactions were 22.56% of 1,232 reported reactions (no count of all people treated)	Life-threatening; deaths reported Not applicable (emergency)	Systematic review ¹⁵
Breathing, skin, digestive, circulatory and nervous system symptoms (injection) Digestive-system involvement was about one-quarter of injection reaction reports. We found no oral-herb stomach upset data.	Among reported reactions: respiratory 37.42%, skin 34.66%, digestive 25.49%, circulatory 25.41%, nervous 23.96%	Mixed, grades I to IV Not stated in abstract	Systematic review ¹⁵
Death and organ toxicity in rats at high extract doses 28-day study of a 95% ethanol extract; no significant effect at 250 mg/kg/day.	Not established	Five rats died at 500 and 1000 mg/kg/day; liver and kidney toxicity Not tested	Animal research ³
Skin product with houttuynia in a blend Four-extract blend, 60 people, 8 weeks; mild skin reactions are not described in the abstract.	Not established	No serious adverse events in either group Not applicable	Controlled trial ²⁰

Liver

No LiverTox entry was found (a search of NCBI Bookshelf returned no record), and no human liver injury case report from houttuynia was found in our searches. In rats, 28 days of a strong ethanol extract at 500 and 1000 mg/kg/day caused liver and kidney toxicity and deaths, with no significant effect at 250 mg/kg/day. Some reviews describe liver-protecting effects in animals, but a review also says liver and kidney toxicity cannot be ignored. ^{3 16 1}

How often. No frequency is available for people.

Monitoring. No monitoring advice was found. Stop and get medical help for yellow skin or eyes, dark urine, pale stools or severe tiredness.

This is a safety connection. It is not evidence that Houttuynia supports liver health.

Medicines and supplements 5

With	Why it matters	Importance	Evidence	What to do
Macrolide antibiotics (for example azithromycin), and penicillin or cephalosporin antibiotics, with the INJECTION prescription	The injection was often given with these antibiotics. Macrolide co-use carried a higher reaction risk than the injection alone.	Higher risk of serious reaction; RR 8.80 (95% CI 6.12 to 12.65) for macrolides	Systematic review Low certainty ¹⁵	Injection only. It is a hospital product; do not self-inject. Tell clinicians about all medicines.
Metformin prescription	In cells and rats an ethanol extract blocked the transporters OCT2 and MATE1, which clear metformin through the kidney, raising metformin levels.	Unknown in people; higher drug exposure is possible	Animal research Not enough evidence ²¹	Ask a pharmacist before combining, especially with kidney problems.
Blood thinners (warfarin, apixaban), aspirin, clopidogrel and NSAIDs prescription	An isolated compound, becatamide, blocked COX-1 and COX-2 and lowered platelet activation markers in mouse blood. Whether this applies to the whole herb or to people is unknown.	Unknown; a bleeding interaction cannot be ruled out and has not been tested	Laboratory research Not enough evidence ¹²	No study of houttuynia with warfarin or any anticoagulant was found. Ask a pharmacist if you take these.
Drugs broken down by liver (CYP) enzymes prescription	A review reports that a houttuynia extract lowered CYP2E1 activity in mice at 300 mg/kg/day for 7 days. We found no human study of CYP3A4, CYP2C9 or other enzymes.	Unknown	Animal research Not enough evidence ¹⁶	Check with a pharmacist, especially for narrow-margin medicines.
Other diabetes medicines prescription	Animal studies report blood-sugar lowering; additive effects are possible in theory.	Unknown	Pharmacological theory Not enough evidence ^{21 16}	Monitor blood sugar and tell your prescriber.

Who should check first 6

Situation	Advice	Evidence
Houttuynia injection in people with a history of drug allergy, and use for colds or mild infections	Reports describe many reactors with earlier allergy (for example penicillin allergy in 15.03% of 286 cases with a described history). Authors advise against herbal injections for colds. Use only under hospital care.	Systematic review ^{15 6}
Children (injection)	Severe reactions in four children in 2006, two fatal, led to the hospital halt. No oral safety data for children.	Authoritative guidance ⁶
Pregnancy and breastfeeding	Avoid medicinal amounts and concentrated extracts. No human data; lab embryo studies are a warning. No LactMed entry found.	Laboratory research ^{9 1}
Kidney disease, or taking medicines that stress the kidneys	Be cautious, ask a doctor. This is a precaution from aristolactam and rat data; no human kidney harm from houttuynia has been shown.	Animal research ^{19 3}
Liver disease	Be cautious with concentrated extracts. This is a precaution from rat data.	Animal research ³
Taking blood thinners or metformin	Ask a pharmacist first; see interactions.	Animal research ^{21 12}

Overdose, product quality and regulators

Summary. There is no human toxicity study of the oral herb. The warnings are: serious allergic reactions to the injection (human reports), liver and kidney toxicity with strong extracts in rats, and aristolactams that are toxic to kidney cells in a dish but much less toxic than aristolochic acid I in mice.

Animal. Rats given a 95% ethanol extract for 28 days: no significant adverse effect at 250 mg/kg/day; five deaths and liver and kidney toxicity at 500 and 1000 mg/kg/day; males more sensitive. Mice given aristolactam I, aristolactam BII or a plant aristolactam fraction: modest rises in kidney markers; aristolochic acid I raised them much more.

Human doses studied. No oral safety trial. Short trials of a multi-herb formula (4 weeks) and a topical blend (8 weeks) reported no serious adverse events in their abstracts.

High dose or long term signals. Only 28-day rat and subacute mouse studies exist. Long-term and human kidney effects are unknown.

Overdose response. Not found.

Genotoxicity. Aristolochic acid I and II and aristolactam I were not detected in 22 samples, and one paper says aristolactam AII has not been reported to have genotoxicity. Another paper found aristolactams in the plant and treated aristolactam I as a main one, so this is not fully settled. Embryo toxicity was suggested in lab models.

Toxic plant parts. In 30 batches, aerial parts held much more aristolactam (3.20 to 108.19 ug/g) than underground parts (0.95 to 11.66 ug/g), flowers the most. No aristolactams were detected in the water extract of any part. Alcohol extracts were not reported.

Supplements are not checked before sale in the US. Houttuynia content of aristolactams varies a lot by batch, region and part. We found no adulteration or heavy metal data for houttuynia products.

Aristolactam content varies. Among 30 batches, the aerial part held 3.20 to 108.19 ug/g and the underground part 0.95 to 11.66 ug/g. Six alkaloids were measured in 75 batches of the dried herb. ^{18 17}

Aristolochic acids not detected. Neither aristolochic acid I nor II was found in 22 samples, nor any aristolochic acid in 75 dried herb batches at a detection limit of 4 ng/mL or lower. ^{22 17}

Body	Action	Date
China drug regulator (SFDA, now NMPA) (China)	Halted hospital use of Yuxingcao (Houttuynia) injection in June 2006 after severe reactions in four children (two fatal); later let a limited number of manufacturers resume after a quality review. Current status not checked. source	2006-06
NCCIH (United States)	No houttuynia page found (the expected address returned 404). source	2026-10-25
Memorial Sloan Kettering and MedlinePlus (United States)	No houttuynia page found (expected addresses returned 404). source	2026-10-25
LiverTox and LactMed (NCBI Bookshelf) (United States)	No houttuynia entry found in a Bookshelf search. source	2026-10-25

When to get medical help

After an injection or a new product: hives, swelling of the lips, face or throat, wheezing, trouble breathing, a fast heartbeat or feeling faint. This can be anaphylaxis, so call emergency services.
Yellow skin or eyes, dark urine, pale stools, or severe tiredness or belly pain: stop and get medical help.
Passing much less urine, new swelling of the legs or face, or back pain with nausea: get medical help.

Stop the supplement and contact a clinician or emergency service. In the US, Poison Control is 1-800-222-1222.

General information only, not medical advice. Ask a clinician or pharmacist before use, especially with medicines, in pregnancy or with a health condition. See the Natural Remedy Safety guide.

Frequently asked questions

Does houttuynia help with COVID-19 or colds?

Not shown for houttuynia alone. A meta-analysis of a many-herb formula that includes it found better outcomes when added to standard care, but its authors say the studies were low quality. Lab antiviral effects have not been shown in people.

^{7 16}

Is the injection safe?

No. A review of 1,232 reports found that 22.56% of reactions were serious body-wide reactions, mostly anaphylaxis, with deaths. Risk was higher with intravenous use and with macrolide antibiotics.

¹⁵

Was the injection banned in China?

Its hospital use was halted in June 2006 after severe reactions in four children, two fatal. A review says a limited number of manufacturers were later allowed to resume. We did not check its current legal status.

⁶

Is tea or food use as risky as the injection?

The injection reactions are not reported for tea or food. We found no case report of anaphylaxis from eating or drinking it, but that is not proof it cannot happen. Reviews say long-term vegetable use has no reported toxicity but ask for better data.

^{15 16}

Does it contain aristolochic acid and can it harm the kidneys?

Aristolochic acid I and II were not detected in two chemistry studies. Related aristolactams were found, and they were toxic to kidney cells in a dish. In mice they were much milder than aristolochic acid I. No human kidney harm from houttuynia has been shown, and no human kidney study exists.

^{22 17 18 19}

Is a tea safer than an extract?

One analysis found no aristolactams in the water extract of any plant part, while the aerial part and flowers held the most. In rats, a strong alcohol extract killed animals at higher doses. Human safety is not tested for either.

^{18 3}

Can it harm the liver?

No human case report was found and no LiverTox entry exists. In rats, a strong alcohol extract caused liver and kidney toxicity at higher doses over 28 days.

^{3 1}

Is it safe in pregnancy or breastfeeding?

No human data exist, and no LactMed entry was found. Lab embryo studies reported effects on development. Avoid medicinal amounts unless a clinician says otherwise.

^{9 1}

Does it interact with medicines?

Evidence is from animals and labs only. An extract raised metformin levels in rats. An isolated compound blocked platelet COX enzymes in mouse blood, so a blood-thinner interaction cannot be ruled out. For the injection, macrolide antibiotics raised reaction risk. Ask a pharmacist.

^{21 12 15}

Can it cause stomach upset?

We found no data on stomach upset from the oral herb. Digestive symptoms were involved in about one-quarter of injection reaction reports.

¹⁵

Does it work for acne or eczema?

One trial of a four-extract skin blend that included houttuynia reduced acne lesions, but houttuynia cannot be credited. For eczema there is only a mouse study.

^{20 5}

Has any health agency reviewed it?

China lists it as food and medicine, as a review reports. No NCCIH, MedlinePlus, Memorial Sloan Kettering, LiverTox or LactMed page was found. EMA could not be checked properly, so that is not proof none exists. Ask a doctor or pharmacist about amounts.

^{16 1}

Explore by health goal

Immune Support Healthy Skin Blood Sugar Balance Metabolic Health Inflammation Balance Healthy Aging

Editorial transparency

Written by: AlternativeMed editorial team. Research compiled with AI assistance, then checked against PubMed records.
Medical review: None. This page has not been reviewed by a physician or pharmacist.
First published: 2026-10-25
Last substantive update: 2026-10-25
How we research: We started from our existing houttuynia summary, read the abstract of every PubMed study cited, read key review sections in full text via Europe PMC, checked the plant name in GBIF and NCBI Taxonomy and compounds in PubChem. NCCIH, Memorial Sloan Kettering, MedlinePlus, LiverTox and LactMed were checked and nothing was found; EMA could not be checked properly.
How we rate evidence: See "How we rate evidence" in the Evidence tab.
Independence: No product is promoted on this page.

Sources 22

Each source was opened and checked. Where we read only the abstract, it says so. The small numbers in the text point here.

AlternativeMed editorial check. Regulator and reference-site check for Houttuynia (NCCIH, MSK, MedlinePlus, EMA, LiverTox, LactMed). Web and NCBI Bookshelf check, 2026-10-25. 2026. (web-check)
Chang KW, Lin TY, Fu SL et al.. A Houttuynia cordata-based Chinese herbal formula improved symptoms of allergic rhinitis during the COVID-19 pandemic. J Chin Med Assoc. 2022. PMID 35421875. DOI 10.1097/JCMA.0000000000000732. (abstract)
Chen H, Sha X, Luo Y et al.. Acute and subacute toxicity evaluation of Houttuynia cordata ethanol extract and plasma metabolic profiling analysis in both male and female rats. J Appl Toxicol. 2021. PMID 34057207. DOI 10.1002/jat.4198. (abstract)
GBIF Secretariat. GBIF Backbone Taxonomy: Houttuynia cordata Thunb. (key 5384931). GBIF API (api.gbif.org), read 2026-10-25. 2026. (api)
Huang R, Zhang W, Hu Y et al.. Houttuynia cordata polysaccharides ameliorate atopic dermatitis in mice through modulation of skin immune barrier and lipid metabolism. Int J Biol Macromol. 2025. PMID 40393586. DOI 10.1016/j.ijbiomac.2025.144264. (abstract)
Ji K, Chen J, Li M et al.. Comments on serious anaphylaxis caused by nine Chinese herbal injections used to treat common colds and upper respiratory tract infections. Regul Toxicol Pharmacol. 2009. PMID 19559066. DOI 10.1016/j.yrtph.2009.06.008. (full-text)
Li Y, Xiao P, Liu N et al.. Efficacy and Safety of Chinese Medicine Lianhua Qingwen for Treating COVID-19: An Updated meta-Analysis. Front Pharmacol. 2022. PMID 35721166. DOI 10.3389/fphar.2022.888820. (abstract)
Liu Y, Yang Y, Wang B et al.. Development and Verification of a Precolumn Derivatization LC-MS/MS Method for the Pharmacokinetic Study of Houttuynine of Houttuynia Essential Oil. Molecules. 2021. PMID 33923761. DOI 10.3390/molecules26082327. (abstract)
Liu Y, Yang G, Yang C et al.. The Mechanism of Houttuynia cordata Embryotoxicity Was Explored in Combination with an Experimental Model and Network Pharmacology. Toxins (Basel). 2023. PMID 36668893. DOI 10.3390/toxins15010073. (abstract)
NCBI. NCBI Taxonomy: Houttuynia cordata (taxid 16752). NCBI Taxonomy (E-utilities), read 2026-10-25. 2026. (api)
NCBI PubChem. PubChem compound records for Houttuynia constituents. PubChem PUG REST, read 2026-10-25. 2026. (api)
Park JB. Becatamide Found in Houttuynia cordata Suppresses P-selectin Expression Via Inhibiting COX Enzyme, Not Increasing cAMP in Platelets. Phytother Res. 2015. PMID 26112157. DOI 10.1002/ptr.5391. (abstract)
Rafiq S, Hao H, Ijaz M et al.. Pharmacological Effects of Houttuynia cordata Thunb (H. cordata): A Comprehensive Review. Pharmaceuticals (Basel). 2022. PMID 36145299. DOI 10.3390/ph15091079. (summary-only)
Sun X, Lv W, Wang Y et al.. Mrgprb2 gene plays a role in the anaphylactoid reactions induced by Houttuynia cordata injection. J Ethnopharmacol. 2022. PMID 35104575. DOI 10.1016/j.jep.2022.115053. (abstract)
Wang L, Cui X, Cheng L et al.. Adverse events to Houttuynia injection: A systematic review. J Evid Based Med. 2010. PMID 21349062. DOI 10.1111/j.1756-5391.2010.01091.x. (abstract)
Wu Z, Deng X, Hu Q et al.. Houttuynia cordata Thunb: An Ethnopharmacological Review. Front Pharmacol. 2021. PMID 34539401. DOI 10.3389/fphar.2021.714694. (full-text)
Wu Y, Liu J, Kang S et al.. Rapid Analysis of Aristolochic Acids and Aristolactams in Houttuyniae Herba by LC-MS/MS. Molecules. 2022. PMID 36558101. DOI 10.3390/molecules27248969. (full-text)
Xu Z, Wang C, Bao W et al.. In vitro nephrotoxicity and quantitative UPLC-MS analysis of three aristololactams in Houttuynia cordata. J Pharm Biomed Anal. 2023. PMID 36812795. DOI 10.1016/j.jpba.2023.115289. (abstract)
Xue S, Wang T, Zhou J et al.. Safety evaluation of aristolactams from the edible herb Houttuynia cordata: implications for dietary exposure and food safety. Food Res Int. 2025. PMID 41267271. DOI 10.1016/j.foodres.2025.117663. (abstract)
Yang JH, Hwang EJ, Moon J et al.. Clinical efficacy of herbal extracts in treatment of mild to moderate acne vulgaris: an 8-week, double-blinded, randomized, controlled trial. J Dermatolog Treat. 2021. PMID 31424962. DOI 10.1080/09546634.2019.1657792. (abstract)
You BH, Chin YW, Kim H et al.. Houttuynia cordata extract increased systemic exposure and liver concentrations of metformin through OCTs and MATEs in rats. Phytother Res. 2018. PMID 29480578. DOI 10.1002/ptr.6036. (abstract)
Yu X, Gao Y, Xu Y et al.. Study of the Contents of Analogues of Aristolochic Acid in Houttuynia cordata by Ultra-High Performance Liquid Chromatography Tandem Mass Spectrometry. Foods. 2022. PMID 35159454. DOI 10.3390/foods11030302. (abstract)

Education, not medical advice. This page is for general information. It does not diagnose or treat any condition and is not a substitute for care from a qualified health professional. Natural does not automatically mean safe. Ask a clinician or pharmacist before changing your care, especially if you take medication, are pregnant or breastfeeding, or have a health condition.