Houttuynia Cordata Extract
Also known as Yu Xing Cao extract, Fish mint extract, Fish wort extract, Heartleaf extract, Chameleon plant extract, Dokudami extract
“CIR Expert Panel says: not assessed.”
HOUTTUYNIA CORDATA EXTRACT (CAS 164288-50-0; Yu Xing Cao / fish mint / dokudami) is a whole-aerial-parts extract of Houttuynia cordata Thunb. (family Saururaceae), a perennial herb native to East and Southeast Asia and a hallmark K-beauty soothing/anti-inflammatory active. The CIR Expert Panel for Cosmetic Ingredient Safety has NOT yet rendered a Panel-issued safety conclusion: a Scientific Literature Review (SLR) covering 6 Houttuynia cordata-derived ingredients (Extract, Flower/Leaf/Stem Water, Leaf Extract, Leaf Water, Powder, Water) was released for public comment on December 16, 2025 with earliest possible Panel review at the March 12-13, 2026 meeting; review of the published 175th Panel agenda confirms the Houttuynia assessment was NOT taken up at that meeting, so as of authoring (April 2026) only the SLR exists — no Tentative Report and no Final Report. The corpus therefore carries this ingredient as not_assessed in the cir scope and confidence falls accordingly. Per the SLR (RLD 2024 + Council 2025 survey data), Houttuynia Cordata Extract is the most-used of the 6 forms — reported in 1032 cosmetic formulations at maximum concentrations up to 1.5% in leave-on face/neck products. Marketed function is skin-conditioning. Active phytochemistry includes >30 flavonoids (quercitrin, hyperin/hyperoside, isoquercitrin, rutin, kaempferol-3-O-rutinoside), volatile oils dominated by decanoyl acetaldehyde and methyl-n-nonyl ketone (the 'fishy-odor' constituents), polysaccharides, organic acids (p-coumaric acid is the most abundant), and aporphine/aristolactam alkaloids in the aerial parts. Topical-relevant in vitro evidence: PMID 35204104 (Charachit 2022, Antioxidants) shows the ethyl-acetate fraction (HC-EA) at up to 100 ug/mL is non-cytotoxic to HaCaT keratinocytes and attenuates UVB-induced ROS, MMP-1, IL-6, IL-8, COX-2, and iNOS via MAPK/Akt modulation. PMID 34961096 (Mapoung 2021, Plants) reports a hyperoside-enriched fraction is non-cytotoxic to human dermal fibroblasts and exerts photoprotective and collagen-sparing effects against UVB. PMID 38790770 (Foods 2024) reports a fermentation broth (HCT-f) at 1.25-20% v/v is non-cytotoxic, reduces LPS-induced IL-6/IL-8/TNF-alpha, and the authors conclude 'high safety profile' from hemolysis and chick-embryo testing. PMID 31424962 (J Dermatolog Treat 2019/2021) is a 60-patient 8-week double-blinded RCT of a 4-herb topical formulation including Houttuynia cordata in mild-to-moderate acne; significant decrease in inflammatory and non-inflammatory lesion counts and IL-1alpha/IL-8/keratin-16; 'no serious adverse events in both groups' (the 4-herb formulation precludes attributing benefit to Houttuynia alone). Aristolactam-alkaloid safety thread: PMID 32975339 (Chan 2021, J Mass Spectrom) detected NO aristolochic acid I or II in fresh or sun-dried Houttuynia samples (LOD 2 ng/g) — directly addresses the recurring confusion with the nephrotoxic aristolochic acids; PMID 41267271 (Xue 2025, Food Res Int) demonstrates structurally related aristolactams (AL-I, AL-BII) in Houttuynia have substantially lower oral bioavailability (39.6-52.7%) than aristolochic-acid-I (99.83%) and modest renal-marker effects, supporting safe DIETARY consumption (the study did not evaluate topical/cosmetic exposure).
Skin-conditioning function per CIR Dictionary; widely used in K-beauty soothing/calming positioning. RLD 2024 data report 1032 formulations at up to 1.5% in leave-on face and neck products — by far the most-used of the 6 Houttuynia INCI forms
UVB-photoprotective and antioxidant activity demonstrated in HaCaT keratinocytes — quercitrin and hyperoside attenuate UVB-induced inflammatory mediators (IL-6, IL-8, COX-2, iNOS, MMP-1) via MAPK/Akt modulation, non-cytotoxic up to 100 ug/mL
Anti-skin-aging activity in human dermal fibroblasts via the MAPK pathway — hyperoside-enriched fraction reduces ROS and inflammatory cytokines, increases collagen synthesis, downregulates MMP-1 with no observed cytotoxicity
Anti-inflammatory and barrier-restoring activity in LPS-stimulated keratinocytes via MAPK/NF-kB suppression — fermentation broth at 1.25-20% v/v non-cytotoxic; authors conclude 'high safety profile' from hemolysis and chick-embryo (HET-CAM) testing
Clinical signal in mild-to-moderate acne in an 8-week RCT (PMID 31424962, n=60 randomized, double-blind, vehicle-controlled): a 4-herb topical formulation containing Houttuynia cordata reduced inflammatory and non-inflammatory lesion counts and decreased IL-1alpha, IL-8, and keratin 16 expression; no serious adverse events. Caveat: 4-herb formulation precludes single-ingredient attribution
Reassuring aristolochic-acid screening: the most directly nephrotoxic compounds in the aristolochic-acid family (AA-I and AA-II) were below detection in fresh and sun-dried Houttuynia samples across multiple Chinese sourcing regions, separating Houttuynia from the documented Aristolochia-genus nephropathy outbreaks
Plant: Houttuynia cordata Thunb. (Saururaceae); aerial parts (stems, leaves, flowers) are the source for cosmetic extracts. Native to China, Japan, Korea, and Southeast Asia. Long traditional Chinese / Korean / Japanese culinary and medicinal history; called dokudami in Japan, eo seong cho in Korea, Yu Xing Cao in Mandarin
CIR safety assessment is INCOMPLETE: only an SLR (December 16, 2025) exists. The Panel had not taken up the Houttuynia assessment at the 175th meeting (March 12-13, 2026), so no Panel-issued Tentative or Final conclusion is yet on record. Information explicitly sought by CIR per the SLR includes chemical properties, manufacturing data, dermal toxicity data, reproductive/developmental data, and DERMAL IRRITATION AND SENSITIZATION DATA AT MAXIMUM CONCENTRATION OF USE — these endpoints are MISSING from the cosmetic-ingredient literature
Aristolactam alkaloids (aristolactam B-II, aristolactam A) are present in the aerial parts — these are structural analogues of the IARC Group 1 carcinogen and renal toxin aristolochic acid I. Two recent papers triangulate the issue but do not resolve it for cosmetic use: PMID 32975339 (2021) found NO aristolochic acid I/II at LOD 2 ng/g in fresh/sun-dried plant material; PMID 41267271 (2025) shows aristolactams have lower oral bioavailability and 'favorable safety profile' for DIETARY exposure, but neither paper evaluates dermal absorption or chronic topical exposure. Topical-route risk has not been formally characterized
Heavy metal accumulation: the SLR explicitly notes Houttuynia cordata is 'highly prone to accumulating heavy metals such as lead and cadmium, based on the soil and environment in which it is grown' (no quantitative cosmetic-grade limit set). Sourcing from contaminated soils is a quality-assurance concern that finished-product testing should cover but ingredient-grade specifications do not always require
Embryotoxicity signals in early-development models: hiPSC-derived embryoid bodies were reduced in diameter at all tested aqueous-extract concentrations (150-350 ug/mL) and embryoid body formation was inhibited at 350 ug/mL (per SLR); zebrafish embryo LC50 = 2052 ug/mL with cardiovascular system affected most strongly. These are screening-level developmental signals, not human pregnancy-exposure data, but they are why CIR specifically requested reproductive/developmental data
Distinct from the Houttuynia cordata INJECTABLE: SFDA has suspended several injectable Houttuynia formulations in China after adverse events including respiratory symptoms, rash, anaphylactic shock, and death (per SLR). These events are route-of-administration-specific (parenteral) and do not transfer to topical cosmetic use, but the trade-name confusion in marketing literature is a frequent source of inflated safety claims in either direction
Strong fishy odor from decanoyl acetaldehyde and methyl-n-nonyl ketone is a formulation/aesthetic challenge but not a safety concern; processed extracts (deodorized, fermented) may have markedly different volatile-oil and flavonoid profiles than crude extracts, with implications for both efficacy and the topical-safety inferences drawn from any one published study